sign in sign up
<<
Home , David Rivett, Eric Burhop, Ernest Titterton, Harrie Massey, Harry Boot, John Madsen (physicist)

UNIVERSITY OF NEW SOUTH WALES SCHOOL OF HISTORY AND PHILOSOPHY

MARK OLIPHANT FRS AND THE

A thesis submitted for the award of the degree of

By David Ellyard B.Sc (Hons), Dip.Ed, M.Ed.

December 2011

ORIGINALITY STATEMENT

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.

20 December 2011

2

COPYRIGHT STATEMENT

I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only).

I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

26 November 2012

AUTHENTICITY STATEMENT

I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.

26 November 2012

3

Acknowledgements

In submitting this thesis, I wish to acknowledge the powerful ongoing support I have received from my supervisor Professor David Miller. We have had many discussions over the last four years from which I have derived great benefit and he has reviewed the many drafts of this work with insight, patience and a keen eye. My interactions with him have been a source of pleasurable intellectual stimulation. I have also received support from my co-supervisor, Dr Nick Rasmussen.

In undertaking this work, I am of course conscious of the influence of Professor , whose career in experimental nuclear first engaged my attention when I co-wrote his biography in 1981. I was also honoured to enjoy his friendship. Since that time, it has been my intention to examine more rigorously his contribution in that field, and in particular his role in the development of the proton synchrotron, the technology which has dominated research into the structure of matter to the present day. This thesis is the outcome of that intention.

I wish also to thank the many people who have helped me access the documentary materials on which this thesis has drawn so heavily, including the staff at the various archives I have used.

My friend Ann Turner has ruthlessly proof-read this thesis. It is a much better piece of work for her participation.

Finally I thank my family, especially my wife, for their support and forbearance. Four years is a long time, and they have been with me all the way.

4

Abstract

The years immediately after World War II saw the development of a new generation of particle accelerators known as “proton ”. These provided beams of particles carrying energy an order of magnitude greater than previously available, permitting study of phenomena not previously accessible for examination. The first such machine to be proposed, funded, designed and commenced was initiated at the by Australian-born Mark Oliphant FRS. Nearly concurrently, two similar machines were commenced in the United States, the Cosmotron at Brookhaven and the Bevatron at Berkeley.

While it is generally acknowledged that Oliphant was one of three researchers (the others being the American McMillan and the Russian Veksler) to have independently come upon the operating principles of the synchrotron, this thesis demonstrates that he was more than simply “first among equals” in this field. Developed by examination of primary sources not previously systematically studied, a chronology of Oliphant's activities in this field clearly shows that he was well in advance of others in proposing the use of such a machine to accelerate . Furthermore, his ideas had significant influence on the teams building the American machines.

We also demonstrate that the Birmingham accelerator was in large measure an embodiment of Oliphant's own personality and style, for better and for worse. Without his initiative and influence, and the utilization of the considerable “capital” accumulated through his career, the machine would not even have been commenced in economically- stressed immediate post-war Britain. The enterprise reflected his preferred way of working; a minimum of detailed, reliance on innovation to solve problems as they arose and inadequate use of engineering

5

expertise. For these and other reasons, his accelerator was not the first to generate a beam, despite its lead time.

The thesis sets this pioneering endeavour against a number of backgrounds: Oliphant's long involvement with accelerator building; the growth of the technology of experimental through previous decades and the growth of the phenomenon of Big Science. It also recounts in detail the conception, funding, design, construction and impact of the machine up until its shutdown in 1967.

TABLE OF CONTENTS

1. Introduction: Nuclear physics in transition 9

2. The Proton Synchrotron: Who Did What When? 21

3. The quest for higher energies: Experimental nuclear 39 physics to 1932

4. Building “capital”: Oliphant in the 1930s 73 6

5. Oliphant at War: 1939 to 1945 160

6. The Birmingham Proton Synchrotron 221 1944 to 1946: Conception and Funding

7. The Birmingham Proton Synchrotron 314 1946 to 1953: Design and Construction

8. The Birmingham Proton Synchrotron 379 1953 to 1967: Operation and Impact

9. Was the Birmingham Enterprise Big Science? 44768

10. Conclusions 483

APPENDIX of synchrotron-related PhDs 505

Sources 507

7

Mark Oliphant FRS at of 38 just prior to World War II

This image gives some sense of scale of the Birmingham proton synchrotron, by comparing the human figures against the magnet in the background.

8

CHAPTER ONE Introduction Nuclear physics in transition

In 1946, a team of and technicians at the University of Birmingham, led by -born Mark Oliphant, began to construct a large of radically new design, intended to generate beams of protons of unprecedented energy for research in nuclear physics. Across the Atlantic, similar developments were underway, though some distance behind those in Birmingham. Collectively, these initiatives would lead before the mid 1950s to the inauguration of three first-generation machines of this new type, known as “proton synchrotrons”.

In the history of experimental nuclear physics (also becoming known at this time as “high-energy physics” and later as “”), the years immediately following World War II marked the start of a major transition. Over the previous decade and a half advances in nuclear physics had depended increasingly on bombarding targets with beams of high-energy particles, such as , protons or deuterons, artificially-accelerated by equipment of growing size, complexity and cost. Experimenters using such machines were seeking to initiate some form of that would throw light on the way atomic nuclei, and the particles that comprise them, interacted and on the forces that controlled those interactions.

Machines for accelerating the bombarding particles had been of two basic designs; “linear accelerators” which imparted a steady and continuous acceleration using high electric voltages (hundreds of thousands or even millions of volts), and “”, which imparted a large number of discrete accelerations to particles as they spiralled

9

around a magnetic field. A third method, the “betatron”, was available for accelerating electrons. As new generations of equipment were created, some balance was sought between two demands; the need for a substantial number of bombarding particles, since that would make interactions more likely, and the quest to give those particles higher energies, since that was expected to reveal new sorts of interactions. The first required a high beam current, the second a high beam energy.

As nuclear physicists began to return to their laboratories after war-time enterprises such as the development of and the building of the first atom bombs, the limitations of the existing methods to accelerate particles, particularly protons, were beginning to show. Impediments to generating ever higher energies were appearing, some natural (relativistic effects), some economic (the growing cost of the equipment). A new approach was needed.

The proton synchrotron was one such innovation. One of the motivations for the push to higher energies was a desire to recreate under controlled conditions in the laboratory phenomena which had previously been observed only in cosmic radiation. At those energies, new phenomena had already been seen, including the production of previously unknown particles such as and mesons. As we shall see, Oliphant used the desire to produce “artificial cosmic rays” to justify the substantial investment needed for his proton synchrotron. Another of the first generation machines was labeled the Cosmotron for the same reason.

The transition in the field went beyond the growing size and capacity of the equipment. In the prewar years, the construction of research equipment for nuclear physics lay within the capacity of the small recurrent grants available for universities, increasingly supplemented by

10

support from philanthropists. In this regard, and in the British context, we will be referring to the contribution of public-spirited industrialists such as Ludwig Mond, Lord Austin and Lord Nuffield. In the United States similar support could be secured from industrial firms and from philanthropic bodies such as the Rockefeller Foundation. However by 1946, the increasing scale of the research apparatus was such that only the public purse could provide the necessary funds, and they therefore needed to be linked to broader public objectives beyond simply the acquisition of knowledge.

For the first time separate laboratories were established for this research, standing outside the university system though closely linked to it. These laboratories grew to substantial size, as did the university- based teams working in the field. This was another element of the transition. Where previously a nuclear physics experiment might be undertaken by a lone researcher or by two or three working together, now teams numbered in the dozens, or even hundreds, would be required to design, build and operate the apparatus. These considerations are dimensions of what was later thought of as a new phenomenon, dubbed Big Science. Experimental nuclear (or particle) physics was often cited as an archetype of this new way of working.

On this transition, Galison wrote

In particle physics, by the late 1950s, virtually all experiments were attached to accelerators, and the move from stations to accelerator laboratories marks one of the most significant long-term trends in the discipline. Studies of individual accelerator laboratories therefore have a crucial role to play. They can depict the history of high energy physics within a

11

framework that complements case studies organized around resolution of specific questions1.

Galison’s comment can be taken to support the present enterprise. Not only is it a study of an “individual accelerator laboratory”, since we propose to examine and document the life of the Birmingham proton synchrotron over a period of more than two decades, but it takes as a central theme the achievement of an engineering goal (energies in the cosmic ray realm), and a resulting capacity for advances in broad areas of nuclear physics, rather than a push to solve any particular scientific question.

Setting the contexts

The construction by Mark Oliphant and his team at the University of Birmingham of a proton synchrotron, which is the focus of this thesis, can be set in a number of contexts, all of which had an influence on both the process and the outcomes of the endeavour.

These contexts include:

o The growth of experimental nuclear physics in the several decades before the Birmingham enterprise, including the motivation to seek to secure ever-higher energies, and Oliphant’s own involvement in several generations of particle accelerators2. o Oliphant’s career at the and the University of Birmingham and his involvement in the ,

1 Galison (1997), p. 316. 2 For some general perspectives on the growth of experimental nuclear physics and in particular on the development of particle accelerators, see Livingstone (1952), Livingstone and Blewett (1962), Ratner (1963), Kollath (1967), Livingstone (1969) , Courant (1994), Dahl (2002), Sessler and Wilson (2007), Halpern (2010). 12

especially the growth of his network of international contacts, of his own reputation and authority and of other forms of “capital”.3 o The impact of wartime activities in nuclear energy on post-war developments in experimental nuclear physics, in particular on the genesis of Oliphant’s own insights into ways of reaching very high energies. o The construction, in much the same time frame as the Birmingham endeavour, of two similar machines in the United States (the Cosmotron at the Brookhaven National Laboratory on Long Island, NY, and the Bevatron at the University of California, Berkeley), and issues of comparison, contrast and interaction. o The development of what has become generally known as Big Science4, of which the postwar construction of increasingly large particle accelerators is widely taken as a prime example. . By undertaking such a “contexting” exercise, we will be able to assess the and significance of Oliphant's role in the development of the proton synchrotron, and indeed what particular influence Oliphant

3 In the context of this thesis, the term “capital”, when placed within quotation marks, has a specific, though broad, meaning. It refers to the totality of the personal factors that Oliphant had gathered through his career and was able to bring to bear on the Birmingham enterprise. It included his own innate technical skills, experience gathered through designing, building and managing large-scale facilities, a capacity to secure funding, high-level communication skills which enabled him to secure support from bureaucrats and loyalty among colleagues, his personal reputation, particularly enhanced through his leading role in the British contribution to the Manhattan Project, and a network of influential contacts, many of whom were close friends. This concept of “capital” as used here did not originate with me. It was suggested by my supervisor, Professor David Miller, and derives, primarily, from the writings of Pierre Bourdieu. In the Science of Science and Reflexivity Bourdieu uses ‘scientific capital’ as an amalgam of economic, symbolic, social and cultural capitals. This concept, which plays a vital role in power relationships in science, incorporates factors such as visibility and peer recognition. See Bourdieu (2004), pp 55-56. “Capital”, as used here, is broader than Bourdieu’s concept, as it includes native and experience-derived factors which may not necessarily be externally recognised but can significantly affect the capacity to successfully act in particular circumstances, such as Oliphant was able to do in initiating the Birmingham proton synchrotron. 4 This will hereafter be the way in which this term is written, though contributors to the literature have expressed it in other ways, such as without the capitals and enclosed in quotation marks. As we shall see in Chapter Nine, the term was first used in this context by Weinberg, and this was his preferred usage. For the sake of consistency, it will be used here. 13

exerted. At a higher level we will at the same time be illuminating a case study of the factors influencing the growth and diffusion of scientific and technical knowledge which may be of wider interest. This must raise the question of the meaning of "discovery" or "invention". To what extent can Oliphant, or indeed anyone, be credited with the discovery or invention of the proton synchrotron? The answer to this question must take account of the increasingly team-based nature, from the postwar years onwards, of research into nuclear physics.

With regard to the growth of Oliphant’s own ideas about the proton synchrotron this thesis is able to make a significant contribution to knowledge. Through examination of a number of primary sources, most or all of which appear to have been little studied, we are able to develop a chronology of that growth from late 1944 (the date we propose for the “Oliphant Memorandum” on accelerating particles to high energies, long accepted as being written in 1943), through to the formal proposals for funding for the machine made in mid 1945, and their subsequent amendment. This will show that Oliphant, in proposing to apply the “synchrotron principle” to the acceleration of protons, and in addressing many of the technical issues involved, was up to two years in advance of any similar proposal from the United States.

Though it does not form part of the present study, we should not lose sight of one larger context for the work of Oliphant and others in building the first proton synchrotrons. Essentially all other large particle accelerators, at least those for the acceleration of heavy , built in the last 60 years have been based on the "synchrotron principle", including the much-publicised Large Hadron Collider5 at the European research

5 For background to this facility, see CERN Communication Group (2008), p. 44. See also Halpern (2010). 14

centre at CERN6, currently the largest such machine in the world. Through the application of new technologies, such as "strong focusing", superconducting magnets and colliding beams, these machines are able to generate effective energies of bombardment many orders of magnitude greater than those possible in the immediate post-war years. Yet their operating principles, in which the particles being accelerated are confined to a narrow annulus by simultaneous variation of the constraining magnetic field and the accelerating radio frequency, are those which Oliphant and others devised. Their work marked the beginnings of a technological revolution in experimental particle physics, which has vastly enriched our understanding of the fundamental structure of matter.

Oliphant’s background7

Given the title of this thesis and how large Oliphant looms in it, we should give at the start some of Oliphant’s background and history. He was born in , in 1901. His initial academic career at the had not been outstanding but his performance

6 CERN is the multinational European Centre for Nuclear Research, located outside Geneva and founded in the early 1950s. It currently hosts the world’s largest proton synchrotron, the Large Hadron Collider (LHC). For general information, see http://public.web.cern.ch/public/. 7 For general information on Oliphant’s background and career, see Cockburn and Ellyard (1981), Carver et al (2004), Bleaney (2001). Cockburn and Ellyard, a biography published to mark Oliphant’s 80th birthday in 1981, was written for a general audience and covered the whole of Oliphant’s career. This determined the “popular” manner in which the building of the Birmingham proton synchrotron, the subject of this thesis, was presented. In addition, many of the sources which have informed the current exercise had not been studied. In writing Cockburn and Ellyard, the latter prepared the chapters dealing essentially with his Oliphant’s scientific activities, while the former covered his private and public life. It was the writing of that book that I assembled the cache of documents, drawn largely from Oliphant’s own files, which now forms the Ellyard Biographical Archive (EBA). It had always been my intention to return to the subject when I could and write a more substantial treatment covering in detail some aspect of his scientific career. For a variety of reasons that task was successively delayed, ultimately by several decades, and has only recently become feasible. This helps explain why the documents in EBA have remained in in my possession and have not been place in a public archive. Plans are now in place to have these documents transferred to the Barr Smith Library at the University of Adelaide which holds Oliphant’s other papers. . 15

improved in the latter years of his course. What had been evident from a young age was his outstandingly high level of technical skill in the manufacture and management of scientific apparatus8. This was to remain a characteristic of his work throughout his career. He was able to pay his way through his undergraduate studies by working as a laboratory assistant to Professor ; the major stimulus to his scientific growth came from the influence of the lecturer Dr Roy Burdon.

He had been strongly influenced by the visit made to Adelaide in 1925 by New Zealand-born Nobel Laureate , then the director of the Cavendish Laboratory at the . Oliphant learned of the exciting progress being made at the Cavendish in the relatively new field of nuclear physics. Once he had decided that he wished to work under Rutherford at the Cavendish, his goal was made possible by his securing an “1851” scholarship 9 which made his goal possible, as they had likewise advanced the careers of other young antipodean researchers. He arrived at the Cavendish in the autumn of

8 Cockburn and Ellyard reproduced information from an application made by Oliphant for a post at the Commonwealth Institute of Science and Industry, containing details of scientific apparatus he claimed to have made. More than 20 pieces of chemical and physical apparatus are listed. The authors commented “It is a formidable list of apparatus for a 17 year old boy to have put together, some of it sufficiently useful to have been put to work into schools, which were doubtless chronically short of adequate equipment for their science laboratories” . See Cockburn and Ellyard (1981), p. 20-21. 9 The correct title of the award Oliphant secured was an “1851 Exhibition Overseas Studentship”. These highly influential scholarships had been made possible by the proceeds of the Great Exhibition held in London in 1851, hence their title. Though not their only purpose, the scholarships enabled students from the “dominions” to travel to Britain to undertake postgraduate research leading to the PhD degree which was commonly not able to be undertaken in the country of origin. Rutherford himself had been the beneficiary of one of these scholarships. For an account of the influence of these scholarships on the building of young careers, see Dean (2003). That specific issue can form one of several starting points for the consideration of “imperial relations” in science, namely the interaction between scientific cultures, activities and institutions in Britain on the one hand and the “colonies” on the other. As Dean has pointed out, in her PhD thesis (see Dean (2005)) and elsewhere, these relationships changed over time, but they continued to have an impact in “post-colonial” times. One relevant example discussed here (see p 339/40) is the manner in which Australia sought to have Oliphant released from his UK obligations after WW2 so that he might take up a role at the fledgling Australian National University. In general however, this theme has not been pursued at length here.

16

1927, with his new wife Rosa, where Rutherford accepted him as a research student. He was admitted initially to College. He completed his PhD in 192910, and continued to work at the Cavendish under a 1851 Senior Studentship and from 1932 under a series of Messell Research Fellowships provided by the Royal Society. 11

As we shall see in Chapter Four, his career advanced rapidly from the early to mid 1930s; in quick succession he became a Fellow of St John's College, Deputy Director of Research at the Cavendish, a Fellow of the Royal Society, and, in his first move to professorial rank, the Poynting Professor of Physics at the University of Birmingham. It is fair to say that these achievements, all secured while Oliphant was not yet 40, reflect the influence of his increasingly close association with Rutherford. Oliphant would have admitted as much himself. It was also in the 1930s that he established his scientific reputation, with a relatively small number of influential scientific papers dealing principally with the reactions between ions of the newly discovered of known as and with the masses of the light elements. Though the early research was undertaken in association with Rutherford, it is clear that Oliphant took the leading role in the construction of the apparatus and the gathering of data. From 1932 he became involved in the design, construction and operation of what would prove to be a succession of particle accelerators, again initially through the influence of Rutherford, though he soon made his mark independently. That involvement would characterise his career from that time onwards, and generated the major enterprise which is the subject of this thesis.

10 Oliphant's PhD thesis was entitled The Neutralisation of Positive Ions at Metal Surfaces and the Emission of Secondary Electrons. According to the abstract, “a method is described for producing a beam of gaseous ions of high intensity and very uniform velocity, by attracting ions from the region of intense ionisation surrounding a low voltage arc”. This experience in the generation of beams of positive ions would prove of great value in Oliphant's later work with accelerated beams of such ions. See Oliphant (1931), p. 120. 11 Oliphant’s first paper was published in 1928, communicated by Rutherford. See Oliphant (1928). 17

While his later career is not of significance for our current purposes, we can briefly recount it for the sake of completeness. In the middle of 1950, after nearly 15 years in Birmingham, Oliphant returned to his homeland to take up a post at the newly-established Australian National University in . There he headed the Research School of Physical Sciences, and began to build a new style of particle accelerator which he planned would exceed the capability of the machine in Birmingham, still incomplete at the time of his departure. The Canberra venture did not succeed, with only its power source, a homo-polar generator, being completed and used for other experimental work. Oliphant resigned his post at the University in 1963/4, though he maintained some research interests and enjoyed a significant public profile. He served as Governor of from 1971 to 1976 and died in 2000, a few months short of his 99th birthday.

What is our hypothesis?

What then are the assertions to be tested against the evidence in the course of this thesis? We here advance both a principal and a secondary hypothesis. In the first place we will examine the contention that, among a handful of influential early players in the development of the proton synchrotron, Oliphant is at least entitled to be considered the “first among equals”, though the evidence presented here goes considerably beyond that. It will be argued that on almost every consideration Oliphant was in advance of the other players, with a lead time of up to two years, and that his influence extended beyond his own machine.

The secondary hypothesis deals with the machine itself, rather than with its genesis. We will argue that the Birmingham proton synchrotron can

18

be seen in large measure as embodying Oliphant's own style and personality, to both its benefit and its detriment. The project was made possible through the accumulated “capital” (in the sense we have already defined) that Oliphant was able to bring to bear upon it, and was initially sustained by his influence and inspiration. The withdrawal of that direct involvement from 1950 onwards was one of a number of factors which contributed to the failure of the machine to be the first to "produce a beam”.

The thesis which follows is divided into two parts. The first five chapters, including the present one, set the context and necessary background. Chapter Two contains an attributional survey, which examines various published accounts of the development of the proton synchrotron in order to discern how the roles of the leading contributors, including Oliphant, have been described. In Chapter Three we will examine the early development of experimental nuclear physics, particularly at the Cavendish, with emphasis on the perceived need to impart as much energy as possible to the particles being used in the bombardment experiments which were the key process for advancing knowledge in this field. This work established the platform from which Oliphant launched his own career as a builder of particle accelerators, culminating in the proton synchrotron.

Chapters Four and Five focus on Oliphant's own activities, through the 1930s in Cambridge and Birmingham, and through the war years in Birmingham and the United States. In these chapters we will be particularly concerned to examine Oliphant's accumulation of “capital” and the development of his habits of working which would influence the post-war project.

19

From there onwards, our focus is on the Birmingham project itself, examining the conception of the machine (Chapter Six), its funding and construction (Chapter Seven) and its utilisation and impact (Chapter Eight). In Chapter Nine we will examine the project in the light of the debate over Big Science. Chapter Ten contains a summary of our findings, in the light of our hypotheses.

20

CHAPTER TWO The Proton Synchrotron: Who Did What When?

As a starting point in this investigation, it is instructive to read the various reviews of progress in this field made over the more than 60 years since the building of the first proton synchrotrons. In this way we can see how priority is assigned among the five key players (Veksler, McMillan, Brobeck, Livingstone, and Oliphant) by various observers and commentators, and how those attributions were shaped by time and circumstance. In the present context, a particular interest lies in the manner in which Oliphant’s contributions were assessed. As we shall see in more detail later, those contributions were of two kinds; the discovery or invention of the operational basis of these machines (the “synchrotron principle”), and the embodiment of that principle in an accelerator for protons. Of the five names listed above, only Oliphant made both types of contribution. McMillan and Veksler spelt out the principle but did not built proton synchrotrons (McMillan never, Veksler only much later). Brobeck and Livingstone built machines, but on principles discovered by others.

The references cited below come from two types of sources. Some are documents which explicitly addressed the historical context; others are technical overviews from which some historical information may be gleaned. In the summation that follows, those overviews generated nearest to the events in question, usually by participants in the “action”, are presented first. These two considerations, proximity in time and degree of personal engagement, raise the issue of whether such views should or should not be given additional weight. It can be argued that such accounts are not necessarily objective, because of personal

21

involvement with one or other of the pioneering projects. Another limitation of the early accounts is that their authors may not have had access to documentation and other sources which later became available for study. This appears to be the case in early assessments of Oliphant’s contribution.

Lawrence (1946)12. A document prepared by Lawrence in August 1946 as text for a lecture to be given at Princeton in September is currently the earliest survey of experimental methods in nuclear physics as they existed, or were emerging, in the immediate post-war years. Lawrence noted that many laboratories were at work on new schemes for accelerating electrons and ions to great energy, and that only multiple acceleration methods of the kind he had pioneered were up to the task. He referred to the betatron (suitable only for electrons) and then to the synchrotron, though again only in the context of acceleration. He noted that a number of such machines were being built at a variety of locations in the United States, including at his own laboratory in Berkeley. He made no explicit mention of a "proton synchrotron”, perhaps implying that no such machine was yet contemplated13. He did however note that the synchrotron had been independently conceived in the United States, Russia and , though he did not mention Oliphant by name.

Lofgren (1950)14. At the time of writing of this overview, Lofgren was working at Berkeley where the Bevatron was under construction. After discussing the limitations placed on the maximum energy obtainable from a by relativistic mass increase, he went on, “The theory of

12 E.O. Lawrence Experimental Methods in Nuclear Physics, 29 August 1946. EOL 72/117/40/30. EO Lawrence Papers, Bancroft Library, University of California at Berkeley [hereafter designated as “LPBL”] 13 As we shall see, no firm evidence of a proposal to build such a machine in Berkeley is available from before about November 1946, some months after Lawrence prepared this overview, though there is no doubt the idea was being discussed there earlier. 14 Lofgren (1950). 22

the synchrotron - electron or proton - and the synchrocyclotron which circumvented this difficulty was worked out independently by V Veksler, EM McMillan and ML Oliphant and depends on the phase stability of an in a magnetic field under certain conditions”.

The reference cited to document Oliphant’s claim was from early 194715. Lofgren went on to discuss how the principle of “phase stability”, newly discovered, could be used to accelerate heavy ions, as in the synchrocyclotron and the synchrotron. He notes that the energy limit of the former, with its fixed magnetic field and changing accelerating frequency, is set at 500 or 1000 MeV by economic considerations, chiefly because of the very large magnet structure required.

He continued

The remaining possibility is that the magnetic field as well as the frequency should vary. The orbital radius can then be held constant, requiring only an annular magnet, which is very much cheaper than the corresponding cyclotron type of magnet. Three of these proton synchrotrons are being built at present. One, at the University of Birmingham, England, is designed for 1.3 BeV16. Another under construction at the Brookhaven National Laboratory will accelerate protons to about 3 BeV and is called a cosmotron. A third is building at Berkeley and will initially operate at 3.5 BeV but can be modified to produce protons of about 6 BeV. This machine has been named a bevatron. The only physical limit to the energy of this class of accelerator is that

15 Oliphant, Gooden, and Hide (1947). 16 There has been some difference of usage in terminology for high energies. Originally the term BeV (for billion electron volts) was used an alternative to 1000 million electron volts. This gave rise to the name Bevatron given to the first proton synchrotron in Berkeley. Due to the different meaning assigned to the word “billion’ in different countries, the modern preference is for the term 1 GeV (giga electron volts) to describe 1000 MeV (1000 mega electron volts). 23

imposed by the loss of energy due to the electromagnetic radiation from a charged particle moving in a circular path. For protons, this limit is so high that the cost, which is one to two million dollars per BeV, is the governing limitation.17

Livingstone (195218 and 195419). American physicist and accelerator pioneer Stanley Livingstone’s review of developments in the field was published in 1952 in the Annual Reviews of Nuclear Science. This review dealt only with proton synchrotrons, but the material was expanded two years later to form part of a book on the broad range of high-energy accelerators. Livingstone’s views are worth giving additional weight, because of his deep involvement in the events he describes, especially at Brookhaven. His account was also written relatively close to those events.

In his 1952 review, Livingstone wrote

The synchrotron principle of magnetic particle acceleration arose from three independent sources at about the same time, another example of parallelism in scientific development throughout the world. The first proposal for a proton accelerator was made by Oliphant at the University of Birmingham to the British Directorate of Atomic Energy in 1943; the details were not published until 1947. Meantime McMillan at the University of California and Veksler of the USSR published papers in 1945 describing the principle.20

In 1954, the expanded text read

17 Lofgren (1950), p. 297. 18 Livingstone (1952). 19 Livingstone (1954). 20 Livingstone (1952), p. 169. 24

The first proposal for a proton accelerator using a ring magnet, in which both the magnetic field and the electric field are varied, was made by Professor ML Oliphant, of the University of Birmingham, England, to the British Directorate of Atomic Energy in 1943. Because of wartime security restrictions, it was not published at the time. It is reported in a detailed design study by Oliphant, Gooden and Hide in 1947, accompanied by a theoretical analysis of orbit stability by Gooden, Jensen and Symonds. An accelerator for protons at 1.0 GeV following these designs has been under construction for some years.21

Livingstone went on to comment that the 1.3 BeV Birmingham machine was the first proton synchrotron to be started. He also noted that design studies for the other two first-generation machines, the Cosmotron at the Brookhaven National Laboratory (BNL) on Long Island, New York and the Bevatron at the Lawrence Radiation Laboratory, University of California at Berkeley did not start until early 1947, around the time the Birmingham design was published. A preliminary design study for the Bevatron, planned initially to run at 10 GeV, was published in 1948, and one for the 3 GeV Cosmotron in 1950.

Blewett (1956)22. This review, coming at the end of the first decade of proton synchrotron development, contained a comprehensive comparison of the three machines then in operation (Birmingham, Berkeley, and Brookhaven). In briefly reviewing the history of the proton synchrotron, Blewett cited two dates: 1943, in which a suggestion had been made by Oliphant that charged particles could be accelerated in a

21 Livingstone (1954), p. 99.

22 Blewett (1956), p. 39-42. Blewett was a colleague of Livingstone in the building of the Cosmotron. 25

magnetic accelerator in which, in contrast to the cyclotron, the radius of the particle orbit could be kept constant by variation of the magnetic field, and 1947, which he said saw the beginning of the published history of the technology with the appearance of a paper by Oliphant, Gooden and Hide23 "describing the proposed design and beginning of construction of the accelerator at the University of Birmingham".

According to Blewett, there was, "at this time"24, no certainty that the particles would remain in phase with the accelerating field, but he ascribed the solution to that issue as coming from the discovery of the principle of "phase stability" by McMillan and Veksler, published in 1945. By 1947, according to Blewett’s account, only Oliphant had proposed to use the technology to accelerate protons, since that application "presented a number of technical difficulties", solutions to which were included in the 1947 paper. Design studies for the two accelerators planned for the USA were published later, and construction of these machines was authorised by the United States Atomic Energy Commission early in 1948.

Ratner (1964)25. In contrast to sources noted above, which were essentially written from an American point of view, this review gave a Soviet perspective on the origin of the new technology, with a particular emphasis on the contribution of Veksler. The latter was credited with discovering "auto-phasing" (his preferred terminology for "phase stability"), though there was a brief mention of McMillan in this context. Descriptions of applications of the principle were restricted almost entirely to Soviet machines, notably a 680 MeV "phasatron" (synchro- cyclotron) and a 10 BeV "synchro-phasatron" (synchrotron), completed in 1957. Both the Cosmotron and Bevatron received brief mentions, but

23 Oliphant, Gooden and Hide (1947). 24 Meaning at the time of Oliphant’s purported 1943 proposal. 25 Ratner (1967), chapter 12. 26

the Birmingham accelerator did not, though there was a reference to the machine Oliphant was at that time planning to build in Canberra (and which indeed had been all-but abandoned by that time). Oliphant himself was not mentioned, and Livingstone, the builder of the Cosmotron was referred to only in connection with the technology of "strong focusing".

Kollath (1967)26. This overview was light on historical detail, though it contained extensive descriptions of the seven proton synchrotrons then in use or under construction. A footnote commented that the "synchrotron principle" (that is, confining the protons being accelerated to a circular orbit by simultaneously varying the magnetic field strength and the frequency of the accelerating voltage) was first publicly stated by Oliphant and his colleagues in a paper in 1947.

Glasson (1967)27. In chapter 9 of his Source Book on Atomic Energy, Glasson briefly reviewed the history of technology for the acceleration of charged particles. His treatment of the history of the proton synchrotron began

In the late 1940s, when the most energetic particles had energies of 350 to 400 MeV, attention had turned to the possibility of attaining energies of thousands of MeV, i.e. in the billion (109 electron volt or GeV) range. Because electrons of high energy revolving in an orbit lose energy by radiation, attempts to produce those particles in the energy range of interest did not appear to offer much promise at the time. With protons, however, the situation is different. The rate of energy loss of the particle … is inversely proportional to the fourth power of the rest mass of the particle. Since the rest mass of the proton is nearly two thousand

26 Kollath (1967), p. 152. A first edition of this overview had been published in 1954 but is not available currently for study. 27 Glasson (1967), Chapter Nine. 27

times that of the electron, the energy of a proton can be raised to extremely high values without any detectable loss by radiation.

After reviewing the very high cost of increasing the size of cyclotrons or synchrocyclotrons in search of higher energies (noting, for example, that the magnet of the 184 inch cyclotron at Berkeley already weighed 4000 tonnes), and the obvious cost advantages of an accelerator that confined the particles to an orbit of constant radius, Glasson continued

The idea of using a ring-shaped magnet for a proton accelerator, with a varying magnetic field and frequency modulation, was proposed by MLE Oliphant in Great Britain in 1943, although details were not available until 1947. The same possibility was studied independently by WM Brobeck in Berkeley in 1946. From these considerations came the development of the proton synchrotron, capable of attaining energies of several billion electron volts without the use of unreasonably heavy magnets.

Glasson then briefly described the building of the Cosmotron at Brookhaven and the Bevatron in Berkeley, and their first operation in 1952 and 1954 respectively, but said nothing of the firing up of the Birmingham accelerator in 1953, though he did list England among the countries in which first-generation (or “weak focusing”) machines had been constructed. This would not have referred exclusively to the Birmingham machine (which was indeed at that time about to be shut down), since the 7 GeV Nimrod accelerator at the Rutherford-Appleton Laboratory was in operation by 1967.

Livingstone (1969)28. While drawing on his earlier commentary, Livingstone added some significant comment in this treatment.

28 Livingstone (1969), p. 50. 28

Accepting the widely-quoted 1943 date for Oliphant's earliest proposals for such a machine, he remarked "this anteceded the discovery of phase stability by two years. But at that time it was no theoretical assurance that the proposal was sound or technically practical”. He noted that an accelerator based on these ideas had been built and operating in Birmingham at 1 GeV by 1952, but the Birmingham machine lacked "some of the advantageous features developed at Brookhaven and Berkeley and operated at a rather low beam intensity”.

Crowley-Milling (1983)29. In his 1983 summation of progress in high- energy particle accelerators, Crowley Milling wrote

With the cost of a cyclotron-type magnet going up as about the cube of the pole diameter, another idea was needed before higher energies could be achieved economically. Oliphant, Veksler and McMillan independently proposed that if the magnetic field was low at the time of injection of the particles and was increased appropriately as the particles gained energy, then only an annulus of the magnet pole would be required, thus reducing considerably the volume of magnetic field that needed to be provided, and also allowed the replacement of the huge dees30 by simpler RF cavities. The success of such an approach depended on two things: phase stability and focusing.

He went on to the review the insights of McMillan and Veksler on the key issue of phase stability, and the development of “strong focusing” to replace the “weak focusing” employed in the first generation of machines. In terms of machines actually built, he referred only to the Brookhaven Cosmotron (operational in 1952 at 3 GeV) and the Dubna

29 Crowley-Milling (1983), p. 58. 30 “dees” were an essential element of the cyclotron, the technology mostly displaced by the synchrotron. Their purpose will be explained later. 29

syncrophasatron in the which generated 10 GeV in 1957. Neither the Birmingham nor Berkeley accelerators were mentioned.

Courant (1995)31. Courant’s personal recollections of the development of proton synchrotrons dated from mid 1947, at which time he joined the project at Brookhaven under Livingstone that would give rise to the Cosmotron. Of the early developments he wrote

Not long after the first synchrocyclotrons and electron synchrotrons were built, it was realised that there seemed to be nothing standing IN the way of much higher energies for proton synchrotrons. Several proposals for proton synchrotrons appeared, notably one for a 1 BeV machine at Birmingham, England (Oliphant, Gooden and Hide) and one for a 10 BeV machine by Brobeck at Berkeley. Discussion between Leland Haworth, Director of Brookhaven, at Berkeley and the AEC32 authorities led to the decision that both Brookhaven and Berkeley, instead of competing for the 10 BeV prize, would each build a smaller proton synchrotron, one about 3 BeV and one at 6 [Bev]. Haworth chose the smaller size in the hope of getting it finished faster; in later years he often said this was one of the best decisions he ever made.

That is his only reference to the Birmingham project, though it must be said that he passes very quickly over the early years, moving on to developments like “strong focusing”33.

31 Courant (1995). A briefer presentation of this narrative (including the quotation given) is found in Courant (1985). 32 (US) Atomic Energy Commission. 33 A essential element of the “second generation” of proton synchrotrons. 30

Seidel (1983)34. More detail on the timing of key decisions on the early accelerators was provided by Seidel in a study of the evolution of the Lawrence Radiation Laboratory in the postwar years. He noted that a proposal for what became the Bevatron had been made by Berkeley physicist William Brobeck.

Brobeck had conceived the machine after participating in a course for accelerator building for engineers in the summer of 1946. By November he had plans ready. II Rabi, then visiting the laboratory, borrowed copies to take back to Brookhaven, where he urged a similar project on the accelerator committee there. In Birmingham, England, Mark Oliphant has already embarked on a project inspired by his wartime stint at the Radiation Laboratory and McMillan’s principle of phase stability35.

The timing here set out by Seidel has been confirmed by McMillan in correspondence with the author36. This stated, in part

The possibility of using the principle of phase stability as a basis for designing both electron and particle accelerators is inherent in the theory, which depends on the type of particle only through the rest mass which occurs in the equations. This was recognised in my first memo on the subject, written at Los Alamos and dated July 4, 1945, not long after I had the idea.

The first draft of my letter to the editor of the was also written at Los Alamos and was dated August 26, 1945... In

34 Seidel (1983), p. 292. 35 This statement is not correct. As we shall discuss later, Oliphant was not aware of McMillan’s (and Veksler’s) work at the time he proposed the building of the synchrotron in Birmingham. 36 McMillan to Ellyard, 5 November 1980. Ellyard Biographical Archive [hereafter cited as “EBA”]. For more information, see Sources. 31

this letter, the application to both electrons and heavy particles was discussed, and my specific proposal for an electron accelerator was given. Such a proposal, leading eventually to the completion of the 184 inch as a synchrocyclotron, was originated at Berkeley shortly after my return, I think by JR Richardson.

These proposals were for machines in which the magnetic field or the frequency was varied. A more complicated design in which both of these parameters are varied together in such a way as to keep the orbit radius constant (now called the “proton synchrotron”) was proposed at Berkeley by William M. Brobeck sometime in 1946. I recall many discussions with him on this topic, but have no notes or documentary evidence of dates, except for a drawing dated November 12, 1946, which is a conceptual design of the Bevatron in considerable detail.

Shortly after this drawing was made, II Rabi of Columbia University visited Berkeley and took a copy back with him. That is how the proton synchrotron got to Brookhaven.

Sessler and Wilson (2007)37. Sessler and Wilson’s Engines of Discovery: A Century of Particle Accelerators is the most recent survey of the history of the field. The section on synchrotrons was mostly a reworking and expansion of material published by Wilson in the 1990s under the heading Fifty Years of Synchrotrons38. In the abstract of that paper, Wilson wrote “The idea of a pulsed magnet ring, fundamental to the synchrotron, appeared in a proposal by Oliphant in 1943 and was followed by the independent discovery of phase stability by Veksler in 1944 and McMillan in 1945”.

37 Sessler and Wilson (2007). 38 Wilson (1997). 32

In the chapter on synchrotrons in the 2007 book, Oliphant received the first mention, with the citing of the 1943 date, though both McMillan and Veksler had previously been discussed in terms of the phase stability issue in cyclotrons. Oliphant was clearly credited as being the first to start construction of a proton synchrotron.

Being more than fifty years distant from the events they describe, Sessler and Wilson clearly drew most of their information on Oliphant’s contribution from Oliphant’s own 1967 account39 and the 1947 paper40, though they do not cite the former. This is plain from the use of terms like “owl watch” and from references to letters to Sir Wallace Akers and to the McMillan and Veksler papers as “comprehensive and beautiful”, none of which are found in other accounts41. It follows that this account is only as reliable as Oliphant’s recollections allow.

Typical of the editorialising in the account is the following42: “[In his paper] McMillan described Oliphant’s pulsed ring-magnet idea and announced his own plan to build such a machine – without a single reference to Oliphant”. In his earlier summary, Wilson wrote “Oliphant recalls that, after his return to Birmingham, he was “shattered by the publication of the comprehensive and beautiful papers by McMillan and Veksler”, which, he admits, did prove powerful support for his proposal to the DSIR for the construction of the 1000 GeV (sic) synchrotron. One presumes that he was disappointed to find no reference to his pulsed ring magnet idea”. The implication appears to be that McMillan was

39 Oliphant (1967), Part 2. 40 Oliphant, Gooden and Hide (1947). 41 The significance of these terms and expressions., which are found in Oliphant’s own accounts, will become clearer as we proceed, for example in Chapter 6. 42 Wilson and Sessler (2007), p. 56. 33

aware of Oliphant’s plans and essentially stole the idea. As we shall see, these assertions are not supported by the evidence43.

Who did what when?

What can be deduced by collating these various accounts of accelerator history, allowing for the fact that many of them drew on the same meagre documentary resources? On the basis of the evidence presented, it seems reasonable to say the following statements are supported, though not all overviews addressed all these issues. o The “synchrotron principle” (confining the accelerated particles to an orbit of mostly constant radius) was independently discovered (or developed) by Oliphant, McMillan and Veksler in 1945 or earlier. o Oliphant was first to give serious consideration to the possibility of using the synchrotron principle to accelerate protons, rather than to accelerate electrons which was the concern of Veksler and McMillan. o Oliphant’s first proposal for what was later called a proton synchrotron was contained in a memorandum written to British authorities in 1943, although details are not available until 1947, due to a variety of factors including wartime security. o Doubt existed as to whether the Birmingham proposal was technically feasible, in other words, whether the orbits of the particles would be stable), and this issue was not satisfactorily settled until the

43 It should be noted that Sessler and Wilson are not reliable on matters of significant detail. They assert, for example, that Oliphant was “charged with overseeing Britain’s coastal defences” and that he first took the magnetron to the USA in 1941. Neither of these statements is correct. 34

publication of the concept of phase stability/auto-phasing by McMillan and Veksler. o A proposal to build a proton synchrotron was independently developed by Brobeck at Berkeley in 1946, leading to the construction of the Bevatron. o Through the agency of I.I. Rabi, the concept of the proton synchrotron was transmitted to Brookhaven where it became the stimulus for the building of the Cosmotron. o The first published design study for a proton synchrotron was for the University of Birmingham machine, and was contained in a paper published in early 1947, by which time the machine was already under construction. Design studies for the American machines were published a year or more later. o The Birmingham synchrotron was the first to be commenced, up to 2 years in advance of the commencement of construction of the American machines.

These statements, summarising the outcomes of the attributional survey are generally correct, according to research undertaken in connection with this thesis, though a 1943 date for Oliphant’s first proposal for such a machine is not supported by the evidence. However, they give only an outline of the early development of this important new technology, and they underplay the role taken by Oliphant and the Birmingham team. We are now in a position to indicate to what extent Oliphant was in advance of other workers in this field, and in what manner the work at Birmingham influenced that taking place in the United States, particularly at Berkeley and to a lesser extent at Brookhaven. These

35

considerations will be shown to support the hypothesis that Oliphant was considerably more than the "first among equals”, among the pioneers of the proton synchrotron.

If Oliphant is in fact entitled to greater credit in these developments, as we shall endeavour to show, why is such credit so sparingly given, especially in more general histories of physics44 in the 20th century? If his name occurs, it gains only the briefest of mentions. This may be thought inevitable given the very large amount of ground such overviews have to cover, but it is compounded by a tendency to emphasise theoretical developments, and their supporting experimental observations, rather than experimental techniques. This results in the brief accounts given of accelerator development taking on a formulaic character, with only one or two leading figures named45.

A further influence arises from what Hughes has called, in a broader context, “bomb historiography"46. The term refers to an essentially Whiggish treatment of the development of nuclear physics, for military purposes at least, which looks backwards from the Manhattan Project and the Big Science exploits to the 1950s and tells the story of the rise in nuclear physics in a selective fashion. The story that results fails to capture the true character of the physics of the interwar and immediate post-World War II periods. As the above quotations show, the history of the proton synchrotron as traditionally told exhibits these features,

44 For example, Brown, Pais and Pippard (1995), Kragh ((1997), Mary Jo Nye Cambridge History of the Physical Sciences/ 45 For example, Brown, Pais and Pippard devoted only 25 of more than 2000 pages to explicit discussion of particle accelerators. In so brief a summary, it is not surprising that only McMillan and Veksler are mentioned in connection with the development of the “synchrotron principle”. Kragh covered the early history of the proton synchrotron in less than a single paragraph, with again only McMillan and Veksler being mentioned by name. See Kragh, p. 302. Brobeck, who designed and built the Bevatron, was not mentioned; Livingstone, who led the construction of the Cosmotron was mentioned only in connection with his involvement with Lawrence in the early development of the cyclotron. 46 See Hughes (2004). 36

notably in projecting backwards from the dominant machines of this type, the Bevatron and the Cosmotron, in accounting for its origins. The common attribution of the discovery of the synchrotron principle to McMillan and Veksler alone as a result of this historiography will be shown here to be problematic. Restoring Oliphant fully to the picture can correct the currently US-dominated linear historiography.

It is not possible to escape the linear historiography altogether in recounting the broader context in which Oliphant operated but, by using the notion of the accumulation of “capital” to account for Oliphant's arrival at the point where the Birmingham machine became possible, many of the problems of the linear historiography have been avoided. Constructing the story on the basis of primary documentation, as we shall be doing, also enables an approach to give due attention to the immediate material, economic and political circumstances of Oliphant's work.

We have seen in the materials presented above many references to some form of a document or memorandum outlining a proposal for a proton synchrotron, reputedly written by Oliphant in 1943. Oliphant himself was first to make public reference to this document in the March 1947 paper, and other authors have quoted this date, though, as we shall see, some reference to a document of this kind was made at least some months earlier.

The nature and content of such a document have been a matter of some controversy. If it existed, and could be proven reliable, it would predate the proposals contained in the published papers by McMillan and Veksler and, by a larger time span, the proposals to build the accelerators at Berkeley and Brookhaven. As we shall see, some have identified it with the so-called “Oliphant Memorandum", a copy of which

37

exists today. That document appears to have gone missing for up to 20 years after it was reputedly created. This led some observers, notably McMillan, to doubt its veracity, so implying that Oliphant was not entitled to the precedence that such a document would suggest. However, in the work that follows, we are able to document the steps by which the Oliphant Memorandum was created, and the role of the concepts it contained played in initiating the construction of the Birmingham machine, though we will also propose a different chronology for its creation. This will demonstrate beyond any reasonable doubt the precedence which Oliphant is entitled to enjoy in the genesis of the proton synchrotron.

38

CHAPTER THREE The quest for higher energies: Experimental nuclear physics to 1932

Oliphant’s activities at the Cavendish and in Birmingham in the decade before World War II and during that conflict had immense implications for his major post-war project, the Birmingham proton synchrotron. From the early 1930s he became deeply involved in the technology of experimental nuclear physics and built significant “capital” to support that endeavour. To place his activities into context, we need to grasp both the “ecology” of nuclear physics in the UK in the years immediately before (who was doing what where, interacting with whom and with whose support), and the chronology.

In particular, the development of the technology that underpinned experimentation in nuclear physics needs to be examined. While Oliphant undertook some important scientific investigations through the early to mid 1930s, leading to significant discoveries, including the nature of the interaction between ions of heavy hydrogen, the existence of nuclei of mass three of both hydrogen and , and the refinement of the masses of the light elements, the main implication for his post-war work lay in the role he played in technological development, notably in new methods to generate beams of high-energy particles for experimentation.

It follows then that the key contextual issues are those dealing with the development of accelerator technology; the stimuli to its development, the significant advances and outcomes, and the leading players. The key date toward which the following overview moves is 1932, the year in 39

which Oliphant, at the age of 31, first became involved in nuclear physics research.

Prehistory

There could be no “nuclear physics” until the existence of the had been demonstrated or until it had been shown possible to transmute one atomic nucleus into another. Those dates lie in the first and second decades of the 20th century. Some commentators have not seen nuclear physics as a definable field of study until the seminal discoveries of the early 1930s, such as the identification of the and the (at which time it began to metamorphose into “particle physics”)47. Thereafter, the field grew vigorously, with the discoveries of artificially-induced radioactivity, and the first transuranic elements following before the end of the decade.

On the other hand, key elements of the technology of investigation have a much longer history, and can be traced back to the mid 19th century, if not earlier. One definable starting point is the first work on “discharge tubes”, involving workers such as the Germans Heinrich Geissler and Johann Hittorf, and the Englishmen and William Crookes. These investigations involved the passing of high-tension electric currents between electrodes sealed into glass bulbs from which most of the air had been evacuated, electrical conductivity rising as the gas pressure fell.

47 This point of view was argued in a discussion paper prepared for a conference on the history of nuclear physics organised by the American in 1967 and 1969, and discussed by participants who included nuclear physics veterans such as Merle Tuve and Stanley Livingstone. Among the points made was that the number of workers and institutions active in the field grew rapidly after 1932, indicating at the least a major transition in the development of the field. See Weiner (1972). 40

These studies led directly to the phenomenon of “cathode rays”, and over the next few decades to both the discovery of X-rays and the identification of “electrons” as the negatively-charged constituents of cathode rays. They also led to the identification of “canal rays”, the streams of positively-charged particles (‘ions”), atoms from which the electrons had been stripped and which moved in a direction opposite to that taken by the cathode rays. Better designs of discharge tubes, resulting in higher concentrations of canal rays, were a key element of particle accelerators in the 1930s. From such studies also arose the vital technique of “mass spectroscopy”.

Progress in this field therefore depended on advances both in electrical technology, such as the invention of the transformer, and in vacuum techniques, which have their origins 200 years earlier in the work of Robert Boyle and Robert Hooke. As we shall see, early advances in nuclear physics likewise depended on better ways to generate both high voltages and high vacua, supplemented by advances in the new technology of electronics.

The central role of the Cavendish

For nearly two decades from 1919, the leading centre of research in nuclear physics in the UK (for most of the time the only one) was the Cavendish Laboratory.48. In consequence the dominant figure in the field was Sir Ernest (later Lord) Rutherford, Director of the Laboratory and Cavendish Professor from 1919 until his death in 1937.49

48 Indeed for at least the first decade the Cavendish was essentially dominant in Europe as well, with only a relatively brief period of challenge coming from researchers in Vienna in the mid 1920s who disputed some Cavendish results. See footnote 72. 49 Ernest (Lord) Rutherford (1871- 1937), New Zealand-born experimental physicist and Nobel Prize winner (1908) for investigations into the nature of radioactivity. Previously Professor of Physics at McGill (Canada) and . Active in defence research during the Great War. President of the Royal Society (1925-1930). For background see Eve (1936-1939), Eve (1939), Wilson (1983). 41

In 1919, the year of his demonstration of artificially-induced nuclear transformation (soon dubbed the “new alchemy”), Rutherford moved from Manchester to Cambridge, becoming both the Cavendish Professor of Experimental Physics and the Director of the Cavendish Laboratory. He began to assemble the powerful team of researchers who for the next decade and a half would make the Cavendish pre- eminent in this new field. Already in residence were Frederick Aston, master of the technique of mass spectroscopy, which had been originated by Rutherford’s predecessor JJ Thomson (who still maintained some research within the Laboratory), and the pioneer and Jacksonian Professor CTR Wilson. James Chadwick50 came with Rutherford from Manchester. Others to arrive over the next decade were Charles Ellis, , Thomas Allibone, John Cockcroft51, Norman Feather, Phillip Moon, Phillip Dee, Harry Massey, WB Lewis, and Mark Oliphant. Others like the Russian Peter Kapitza and radio-physicist Edward Appleton added to the lustre of the Laboratory, if not specifically to its reputation in nuclear physics.

The intellectual power of the Cavendish at its zenith was reflected by the often-reproduced group photograph taken in 193252. Of the eleven men sitting in the front row, eight were already, or would become, winners of

50 (1891-1974), British physicist and Nobel Prize winner (1935) for the . A life-long friend and colleague of Oliphant. Interned during the Great War, he was a leading participant in the Manhattan Project during World War II. At the time of Oliphant’s arrival at the Cavendish in 1927, Chadwick was Rutherford’s deputy. Later Professor of Physics in Liverpool and Master of Caius College, Cambridge. For background see Massey and Feather (1976). 51 (1897-1967), British physicist and engineer and Nobel Prize winner (1951 with Ernest Walton) for first nuclear transformations using artificially-accelerated particles. A life-long friend and colleague of Oliphant. Originally trained in electrical engineering and with a continuing association with the leading firm Metropolitan Vickers, he had arrived in Cambridge in 1922. Later first director of Harwell and Master of Churchill College, Cambridge. For background, see Oliphant and Penny (1968), Hartcup and Allibone (1984). 52 For example, see Hendry (1984), p. 70. 42

the Nobel Prize for Physics or for (in the case of Rutherford and Aston). Another (Walton) was in the second row, standing next to Oliphant.

As Burcham53 and others have pointed out, the slow erosion of the standing of the Cavendish in the field in the mid 1930s was in part due to the diaspora of those same researchers as they took up senior posts in universities around the country. Allibone returned to his previous employment in industry in 1930. In 1933 Blackett moved to Birkbeck College in London, the theorist Neville Mott to Bristol and Massey to Belfast. Walton moved to Dublin in 1934; in the same year Kapitza was detained by Russian authorities and prevented from returning to Cambridge after his summer holidays. Chadwick (and Feather briefly) went to Liverpool in 1935; in 1936 Ellis migrated to Kings College, London and Oliphant to Birmingham (though the latter did not take up his post till 1937). Of the leading players, only Lewis, Cockcroft, Dee54 and Moon55 were still at the Cavendish when Rutherford died in 1937. Rutherford’s successor, Nobel Laureate William , was not a nuclear physicist.

The loss sustained by the Cavendish was clearly the gain of the laboratories to which these researchers moved, often as heads of department. It also resulted in the spread of the techniques for investigating nuclear reactions and transformations, so that the study of nuclear physics in the UK moved beyond the Cavendish where it had been pioneered. The first significant item of nuclear physics apparatus built outside the Cavendish was the cyclotron at the , following Chadwick’s move there. It was not until later in the

53 Burcham (1989), p. 832. 54 In 1943, Dee was appointed Professor of Natural Philosophy at the University of Glasgow, a post made famous by the long tenure there by William Thompson (Lord Kelvin) in the 19th century. 55 Moon followed Oliphant to Birmingham in 1938. 43

1930s that GP Thomson, son of JJ, and Professor at Kings College, London since 1930, took a serious interest in experimental nuclear physics, in time for the discovery in of nuclear fission late in 1938.

The origins of nuclear physics

It is appropriate, and perhaps inevitable, that the Cavendish under Rutherford came to symbolise experimental nuclear physics from around 1920. Rutherford and his coworkers at McGill University in Canada (1898-1907) and at the University of Manchester (1907-1919) created this field of physics through a series of key discoveries; in particular:

o the existence and nature of alpha particles, emitted by various radioactive elements such as radium, and found by Rutherford to be doubly charged helium ions, that is, helium atoms stripped of their electrons; 56

o the use of alpha particles from 1908 to 1911 to bombard thin films of gold and other heavy elements, thereby demonstrating the existence of the atomic nucleus as the location of all of an atom’s positive charge and most of its mass;

o the demonstration, announced in 1919, that an striking an atomic nucleus (in this case a nitrogen nucleus), can cause it to transmute into the nucleus of a different element (in this case oxygen), with the liberation of other subatomic particles (a proton in this case).

56 It was for this discovery that Rutherford was awarded the Nobel Prize for Chemistry in 1908. 44

All of these discoveries involved alpha particles, and these successes cemented the use of natural emissions, which included beta-particles (electrons) and gamma rays as well as alphas, to bombard other elements as the key experimental technique in the new discipline, together with the gathering and identification of the products of any interaction.57 The technique had two limitations, the energy of the bombarding particles (the highest available being about 8 million electron volts for alphas from radium) and their numbers (limited by the available amount of the radioactive material, such as radium). In the experiments to date, only one alpha particle in a million hit a nucleus, so that productivity was low. The need to increase both the energy of bombarding particles and their numbers was the major motivation for the technological advances in this field in the late 1920s and early 1930s, at the Cavendish and elsewhere.

While more than a decade elapsed between the first artificial transmutation and the great discoveries of the early 1930s, those years were not entirely barren, if frustrating for the small number of workers active. The concentration, at the Cavendish at least, was on the more accurate measurement of nuclear masses, on alpha particle scattering from a variety of nuclei, on the hunt for the predicted neutron, and on measuring the spectra of the alpha, beta and gamma radiation emitted by radioactive materials.

Nuclear physics at the Cavendish

When Oliphant entered the Cavendish in 1927, his initial research plans had nothing to do with nuclear physics. He proposed to Rutherford that he undertake some work on the effect on metals of bombardment by

57 The vital role played by natural radioactive emissions in nuclear physics research from the 1910s to the 1930s has been well chronicled by Hughes (1993) who has dubbed the researchers at this time (at the Cavendish and elsewhere) “The Radioactivitists”. 45

positive ions, if Rutherford thought it fit within the program of the laboratory. Such research was in fact likely to be of more interest to Thomson or Aston, though it would continue lines of enquiry that Oliphant had begun in Adelaide.

Yet it was perhaps inevitable that he would in the long run become caught up in nuclear physics, so increasingly large did it figure in the research program of the laboratory. Cockcroft later ascribed this dominance to Rutherford’s influence, recalling that “Rutherford, in general, was not very interested in other branches of physics than nuclear physics”58. He went on to comment that, apart from the work carried out by Peter Kapitza on low temperatures and high magnetic fields, the only other significant side line was the study of the ionosphere initiated by Edward Appleton and carried on after his departure in 1924 by Jack Ratcliffe and others.

One piece of evidence for the pre-eminence of nuclear physics at the Cavendish is the program for the opening of the Royal Society Mond Laboratory in February 193359. This listed a number of demonstrations able to be viewed by visitors. Of a total of 12 demonstrations (other than those to be viewed in the Mond Laboratory), only one (on observations of the Heaviside Layer using reflections of radio waves) did not deal with nuclear physics.

This emphasis is perhaps not surprising (even if its extent is), given Rutherford’s own substantial achievements in the field, combined with what was thought of at the time, by American researchers at least, as the “European way”, in which the head of a research school set the tone and direction of research. They saw American laboratories as more

58 Cockcroft (1984). 59 Program for the opening of the Royal Society Mond Laboratory, 3 February 1933. EBA. 46

democratic. Rutherford, as is well known, controlled the Cavendish as a benevolent (and well-loved) despot, for example forbidding anyone to work there after six in the evening60. At the same time, he was such a fount of ideas and insights and expounded them with such enthusiasm that, as was generally conceded, virtually all research projects at the Cavendish originated with him, and most people followed him willingly61. On the other hand, it may well be that the breakup of the team in the early to mid 1930s was, at least in part, the result of the senior men feeling that, after a decade or more under Rutherford’s eye, it was time to have “a show of their own.”

Other influences have been cited, most notably the laboratory’s lack of resources. The pioneering discoveries had been made with inexpensive equipment, suggesting that the cash-strapped Cavendish could continue to be at the centre of activity if it concentrated on the nucleus.

Rutherford’s 1927 Presidential Address to the Royal Society

The first public canvassing of the potential afforded for progress in physics by the availability of very high voltages “for general scientific purposes” was that by Rutherford during his anniversary address as President of the Royal Society on 30 November 192762. Dealing also with the production of intense magnetic fields, he began by noting the limitations of existing technology and the need for new methods, whether “for purely scientific or for technical uses”.

60 See Weiner (1972), p. 37. 61 See for example Oliphant’s comment on working closely with Rutherford, “he drove us mercilessly, but we loved him for it”. Oliphant (1972), p. 108. 62 Rutherford (1928). This address was given a few months after Oliphant arrived at the Cavendish. Rutherford served as President of the Royal Society from 1925 to 1930. 47

Reviewing the existing technology, Rutherford referred to electrostatic machines able to produce weak currents at potentials from 200 000 to 300 000 volts, large induction coils which could give momentary voltages of the same magnitude, and light-weight transformers developed to supply X-ray tubes with the 300 000 to 500 000 volts needed to generate intense radiation for deep therapy. Tesla coils were able to supply a million volts or more through their secondary windings, though with such low currents and high rates of oscillation as to make them of limited value in the laboratory.

Rutherford went on to note the interest of the electrical supply industry in high voltage research, given the trend to higher transmission voltages to reduce losses over long distances. To that end, very high voltage plants had been built on a “cascade” principle for the purpose of testing insulation. These ran as high as 2.8 million volts in apparatus constructed at the Company at Schenectady in the United States, according to the latest information available to Rutherford, with the prospect of reaching 6 million volts.

While no doubt the development of such high voltages serves a useful technical purpose, from the purely scientific point of view interest is mainly centred on the application of these high potentials to vacuum tubes in order to obtain a copious supply of high-speed electrons and high-speed atoms. So far we have not yet succeeded in approaching, much less surpassing, the success of the radioactive elements in providing us with high- speed alpha-particles and swift electrons. The alpha-particle from radium-C is liberated with an energy of 7.6 million electron volts, i.e. it has the energy acquired by an electron in a vacuum which has fallen through this difference in potential. The swiftest beta-

48

rays from radium have an energy of more than 3 million electron volts63 ….

The best efforts to match such performance had come, according to Rutherford’s best information, from the work of William Coolidge64 of the US in relation to X-ray tubes65. Coolidge had succeeded in accelerating electrons to 35 000 volts in a single tube, the highest possible potential before electrical breakdown. By joining three tubes in series, he had forced a stream of electrons energized to 900 000 volts through a thin window in the last tube, so constituting a current of one or two milliamps.

While the energy acquired by the individual electrons in falling through 900 000 volts is smaller than that possessed by the swifter beta-particles expelled by radium, the number emitted by the vacuum tube is very much greater; for example, the number of electrons per second corresponding to a current of 2 milli- amperes is equivalent to the number of beta-rays emitted per second from 150 000 grammes of radium in equilibrium66.

Rutherford was in fact prescient in recognizing that beam current was a factor of significance, perhaps even of comparable significance to voltage in the acceleration of particles. It would prove so in the accelerator project that he and Oliphant jointly developed following events in 1932.

63 Rutherford op. cit. p. 309. 64 Coolidge was in the audience on the occasion of this address, as he was to be the recipient of the Society’s for his work on X-ray tubes. 65 As Rutherford reported, Coolidge was building on work done by the German Phillip Lenard in 1894, who had succeeded in extracting a beam of “cathode rays” at 80,000 volts from a discharge tube. 66 Rutherford op. cit. p. 310. 49

Noting the important discoveries that had been made with naturally- emitted alphas, to be specific, the existence of the atomic nucleus and of its potential for transformation, Rutherford concluded this section of his comments as follows.

It would be of great scientific interest if it were possible in laboratory experiments to have a supply of electrons and atoms of matter in general, in which the individual energy of motion is greater even than of the alpha-particle. This would open up an extraordinarily interesting field of investigation which could not fail to give us information of great value, not only on the constitution and stability of atomic nuclei but in many other directions.

It has long been my ambition to have available for study a copious supply of atoms and electrons which have an individual energy far transcending that of the alpha- and beta-particles from radioactive bodies. I am hopeful that I may yet have my wish fulfilled, but it is obvious that many experimental difficulties will have to be surmounted before this can be realised, even on a laboratory scale.67

Here Rutherford had unequivocally laid down the challenge to his colleagues, to produce the techniques needed to deliver his vision. Two immediate questions arise; what was the consequence of his call, and what was its motivation, or its antecedent? The first answer involves Ernest Walton, though he was not initially aware of the challenge, and soon after John Cockcroft; the second brings forward the name of Thomas Allibone. We will return to these names shortly.

67 Rutherford (1928), p. 310. We may note here in anticipation the similarity between Rutherford’s words and the arguments advanced by Oliphant nearly two decades later to justify the building of the proton synchrotron. 50

Rutherford concluded his address with the following words, stressing the importance of new investigative methods in the growth of knowledge.

The advance of science depends to a large extent on the development of new technical methods and their application to scientific problems. The recent work to which I have referred, on the development of methods of producing high voltages and intense magnetic fields, is not only of great interest to scientific men in itself, but promises to provide us with more powerful methods of attack on a number of fundamental problems.68

Dahl69 noted that the address represented a significant departure from Rutherford’s more pessimistic attitudes expressed over the previous five or so years. He quoted several statements which portrayed Rutherford as thinking that the nucleus would not further yield any of its secrets to current methods, or even to any likely to arise. This was in some measure a consequence of his failure to make any significant new discovery since the first artificial transmutation, and perhaps a sense of his own declining powers as an experimental scientist.

And what was the cause of this substantial change in attitude? To Dahl at least, the answer lay in the pioneering work of Thomas Allibone, which is discussed below.

Physics and industry

As the scale and complexity of physics research apparatus grew, most notably in nuclear physics, so did, inevitably the involvement of industry. It was traditional, at the Cavendish and elsewhere, for the first

68 Rutherford op. cit. p. 312. 69 Dahl (2002), p. 14. 51

generation of new apparatus to be hand-crafted by the experimenters themselves, using whatever “string and sealing wax” was available, with some support from the workshop staff. But technical demands began to exceed the in-house capacity of the laboratories, and larger pieces of equipment were soon being sourced from elsewhere.

Hughes70 has pointed out that the physics laboratory of the 1930s was “crammed with the products of industry”, such as the great range of radio valves, originally developed for wireless but well able to be adapted to serve in scientific instrumentation. Indeed there remains a view that in large measure the availability of materials and technologies arising from industry determined the pace of progress in physics. For example, Hughes quoted Patrick Blackett, in conversation with Julian Huxley in 1934, as follows.

In fact it may be said that the limits of knowledge at any time is set by the technical means available. I believe that the reasons for the rapidity of advance of modern physics is not the superiority of modern physicists, or even their number, but that it is to a considerable extent due to the technical aids made available by industry71.

As one example of this interaction, among the many that could be listed, Hughes cited the consequences of dealings between the Cavendish and the General Electric Company of Schenectady, NY72. It began with the presentation by AW Hull of GE of a 30 000 volt DC power supply in 1927. Contact having been established, Hull visited the Cavendish two years later, bringing news of a new type of radio valve. vapour-filled “thyratron” was able to serve as a relay or switch. It was

70 Hughes (1998). 71 Ibid. p. 88. 72 Ibid. p. 70. 52

with this and similar innovations that the genius of Eyl Wynn-Williams, WB Lewis and their Cambridge colleagues, many of them drawing on backgrounds as radio “hobbyists’ or “hams”, created a series of electronic devices to count particles automatically. This consigned to history the venerable scintillation screen (and the controversies which had arisen from its use73) and made possible the major discoveries in nuclear physics in the 1930s.

In their overview of the development of accelerator technology, Livingstone and Blewett commented on the particular closeness of physics, engineering and industry in this field.

The development of accelerators has paralleled and sometimes paced progress in the electronics industry and in several branches of electrical and mechanical engineering. In few scientific fields has there been such effective and happy collaboration with the engineering profession as in the accelerator field. 74

They continued later

The first ion accelerators were simple applications of direct high voltage to evacuated discharge tubes. Nuclear physicists leaned on experience in the X-ray industry, and as the accelerator art developed, it stimulated further development of higher energy X- ray machines. Focusing requirements forced intensive studies of ion and electron optics. Some of these developments have since

73 Dahl (p. 14) devoted a deal of space to such a controversy in 1926 between the Cavendish and the Radium Institute in Vienna, one of the few laboratories following up on Rutherford and Chadwick’s work on alpha particles. This dispute highlighted the problem of subjectivity in counting particles by watching flashes of light on a screen. See also Wilson (1983) p. 473, Hughes (1993), Stuewer (1985).. 74 Livingstone and Blewett (1962). 53

been applied to the electronics industry. New techniques for the production of X-rays, such as the resonance transformer, were stimulated by accelerator developments75.

For the physicist seeking to respond to Rutherford’s challenge, the focus was on generating and controlling very high voltages. The interest of industry in such voltages stemmed from a desire to pursue two commercial opportunities. The first has already been noted here; the production of X-rays, particularly for medical diagnosis and therapy, the latter in part as an alternative to gamma rays from costly radium. As a result nuclear physics researchers would in time be able to buy high- tension sources “off the shelf”, as Oliphant was able to do in 1936 to energise the new Cavendish HT laboratory.

The other relevant concern was the long range transmission of electricity, since for a given power transferred; higher voltages mean lower currents and therefore fewer resistive losses. As a result, high tension transformers and switch gear had to be designed, built and tested. As Livingstone and Blewett noted76, the use of 220 kV lines for electrical transmission in the western United States stimulated the establishment in the early 1920s of a high voltage laboratory at the California Institute of Technology77, sponsored by the Southern California Edison Company. As noted above, Rutherford referred to this development in his 1927 Royal Society address. There was also a need to devise methods to protect transmission assets against lightning strikes.

75 Livingstone and Blewett (1962), p. 7. 76 Livingstone and Blewett (1962), p. 22. 77 The high voltage technique used here was the “cascade transformer”, in which a number of high voltage transformers were connected in series, a few turns laid over the secondary of each being used as the primary of the next in the series. In this way, AC voltages of 750 kV were being produced by around 1928. 54

The Cavendish and “Metro-Vick”

In the case of the Cavendish, the preferred supplier of large scale apparatus derived from electrical engineering was the Metropolitan Vickers Electrical Company, based at Trafford Park, Manchester. The company had been founded in 1899 as the British arm of the Westinghouse Corporation, itself created by the American engineer and innovator George Westinghouse, and concerned with large-scale electricity generation and transmission. During the Great War, the Westinghouse assets had been purchased by Vickers, a large Sheffield- based armaments manufacturer, and the wholly British-owned Metropolitan Vickers (“Metro-Vick”) formed.

Its association with the Cavendish had its origins in Rutherford’s time in Manchester. It was cemented over the years by the strong mutual regard between Rutherford and senior staff at Metro-Vick, most notably Arthur Fleming78, initially head of the transformer testing laboratory, and later effectively head of research, his right-hand man, George McKerrow, and Brian Goodlet, who led the high-tension laboratory. Over the same time, two researchers with strong Metro-Vick connections, John Cockcroft and Thomas Allibone became part of the Cavendish, Allibone for only a few years before returning to Metro-Vick, Cockcroft for nearly two decades.

In 1928, Vickers sold its stake in the firm to the US General Electric Company79, but the linkage to the Cavendish continued and strengthened. Metro-Vick designed and built equipment for the Cavendish, and supplied it on favourable terms, even donating it. This

78 Dahl noted that Rutherford and Fleming had served together on a committee researching anti- submarine warfare early in the Great War. See Dahl (2002), p. 19. 79 GE already owned Metro-Vick’s great commercial rival, British Thomson- Houston of Rugby, but despite common ownership the two firms remained fierce competitors. 55

was not altruism; the connection effectively expanded the company’s limited research resources. The team there were kept in touch with the latest developments in nuclear physics, as they wished to be, and felt part of it. There was also the prestige of being associated with so famous a centre of research, and for many of the senior men, a love of Cambridge and all it stood for80.

In terms of equipment procurement, the first substantial interaction involved the very large (1500 KW) alternator required to energise the coils which Kapitza planned would generate magnetic fields of unprecedented strength. This was in place in the Cavendish by 1926, funded by the Department of Scientific and Industrial Research. With his strong electrical engineering background and Metro-Vick connections, Cockcroft was much involved in bringing the unit into service.

Thomas Allibone

If we look to determine who was first to propose a solution to the limitations of naturally-occurring bombarding particles, so answering Rutherford’s challenge by developing techniques for the artificial acceleration of particles to promote nuclear transmutations, the answer appears to be TE (Thomas) Allibone, with ETS (Ernest) Walton, not far behind. Interestingly, Allibone’s “response” came before Rutherford had publicly asked the question.

Allibone was not the first to try to accelerate sub-atomic particles by electrical or electromagnetic means. Some years earlier James Chadwick had tried to energise protons to 200 000 volts using a Tesla

80 Allibone noted that nine men associated with Metro-Vick became Fellows of the Royal Society. 56

coil81, though the goal was not the one that Allibone later sought. Chadwick was trying to create the predicted but elusive neutron by forcing protons into close proximity with the tightly-bound outer electrons of heavy elements82. Of this unsuccessful effort, Chadwick wrote

… I had considered the possibility that the neutron might be formed, or exist, only in a strong electric field; and that perhaps one might find some evidence by firing fast protons into atoms, especially those higher atomic numbers where some electrons were tightly bound …. I thought that at least 200 000 volts would be necessary for the acceleration of the protons. No suitable transformer was available, and though Rutherford was mildly interested, there was no money to spend on such a wild scheme. The research grant of the Cavendish was £2000 a year, little enough even in those days for the amount of work that had to be supported. I persisted with the idea for a year or two, and in the intervals of other work I tried to find a way of applying Tesla voltages to the acceleration of ions in a discharge tube. I had quite inadequate facilities and no experience in such matters. I wasted my time but not the Laboratory’s money83.

Dahl commented

Chadwick would indeed discover the neutron by a different procedure, in the disintegration of atoms under alpha particle bombardment, a decade later. For our purpose, however, the thing to note is that the need for higher voltages across gaseous

81 A Tesla coil, named after its inventor Nicola Tesla, was a form of transformer with very few turns in the primary windings and many thousand in the secondary, resulting in a very high output voltage. The voltage in the secondary was maximised by tuning the primary and secondary circuits so that they resonated at the same frequency. 82 At the time, it was thought that the uncharged neutron might be formed by the combination of the positive proton and the negative electron. 83 Chadwick (1984), p. 41. 57

discharge tubes was driven, not so much by the need for higher voltages in disintegration experiments as by the somewhat misguided effort of producing Rutherford and Chadwick’s pet construct, the neutron, by electrical means84.

Thomas Allibone arrived at the Cavendish late in the autumn of 192685, inspired to come to work on the disintegration of atoms by a lecture on the subject at the Sheffield University Physical Society (he was then the Secretary of the Society) given by Cavendish researcher Charles Ellis in November 1925. With the support of the firm of Metropolitan Vickers, for whom he had been working for several years, and a scholarship to Gonville and Caius College, he had secured entrance to the Cavendish, with the intention of working on the acceleration of electrons86 to very high energies using a Tesla coil.

It is not clear, even from Allibone’s own recollections87, how the notion of artificially accelerating particles to try to bring about transmutations occurred to him. We only know that it did, stimulated by Ellis’ lecture and supported by conversations with Brian Goodlet, head of the High Tension Laboratory at Metropolitan Vickers in which Allibone had worked. Allibone made a passing reference to the work of William Coolidge, working at the General Electric labs in Schenectady in the US, which he said “inspired me”. As Allibone recalls, Coolidge had developed X-rays which required an electron beam to be given an energy of 350 000 eV.

84 Dahl (2002), p. 14. 85 Cockcroft recalled Allibone as arriving at the Cavendish after the Rutherford address. “The first response to [Rutherford’s comments] was the arrival of TE Allibone from Metropolitan Vickers.…” See Cockcroft (1984), p. 77. Cockcroft was apparently in error here. 86 Allibone recorded that Rutherford had encouraged a focus on electrons rather than on positive ions since the latter were not readily produced and would have no chance of penetrating the nucleus unless they carried as much energy as natural alpha particles (around 8 million eV). This line of argument was soon to be countered by Gamov. 87 Allibone (1984). 58

Goodlet had suggested that the cheapest, lightest and smallest source of high voltages was a Tesla transformer, operating at around 50 000 hertz with no iron core. Through 1927, Allibone worked to assemble his equipment. Compared with Chadwick, Allibone had considerable relevant experience but the lack of resources typical of the Cavendish at the time made progress very slow, though the windings for the Tesla coil had been provided by Metropolitan Vickers.

Cockcroft later recalled that Allibone had arrived at the Cavendish

… bringing with him a 500 KV Tesla coil with rotating spark gap excitation which upset radio sets within a quarter of a mile around. He installed this in our already overcrowded laboratory. To this he connected a home-built 300 KV electron tube and he produced intense beams of electrons, directing them onto fluorescent minerals for the edification of visitors like Rutherford. …. Allibone stuck to electron work but his success in operating accelerating tubes at 300 KV was an important link in the chain of events which led me to take up Rutherford’s challenge88.

Allibone said little about the outcomes of his work in his recollections of the Cavendish but Dahl89 reported that by the autumn of 1927, Allibone had his system working, generating an almost monochromatic beam of electrons of several hundred kilovolts for scattering experiments. According to Dahl the results were impressive enough to inspire Rutherford and to encourage him in making his comments to the Royal

88 Cockcroft (1984), p. 76. 89 Dahl (2002), p. 20. 59

Society at the end of the year, though he does not mention Allibone by name90.

Likewise, it appears that the outcomes were enough to encourage Rutherford to seek even more. Allibone recalled him in 1928 seeking his views on the possibility of much higher voltages than those being attained “something over 450 kV albeit with many breakdowns”. Allibone consulted Goodlet and together they designed a two million volt Tesla transformer, to operate under oil and under pressure, in a two metre diameter tank. Financial constraints and other distractions prevented the machine being built91. Tesla coils were soon to fall out of favour as sources of high voltage for particle acceleration.

Ernest Walton

ETS Walton had arrived at the Cavendish from Dublin as an 1851 Scholar in October 1927, at the same time as Oliphant (by Oliphant’s recollection, it was the same day)92. In late 1927, having completed the necessary time in the Nursery93, he discussed with Rutherford the line of research he would follow.

As an undergraduate, when reading about [Rutherford’s] own work on the disintegration of atoms by the use of natural alpha- particles, I wondered why other projectiles had not been tried. By now I knew that alpha-particles were the only fast projectiles

90 We should not read too much into this omission. When Rutherford moved on in his address to discuss the generation of large magnetic fields, he did not name Kapitza, who had been at the Cavendish far longer than Allibone. 91 Allibone (1984), p. 159. 92 Oliphant (1972), p. 19. 93 Upon arrival at the Cavendish, new research students were required to spend a term under the supervision of Chadwick learning basic techniques of experimentation in nuclear physics in an attic room dubbed “The Nursery”. Because of his strong technical background, and the fact that his proposed research work would not be in nuclear physics, Oliphant appears to have been spared this experience. 60

available to him. To produce other particles would require the generation and application of electrical potentials of several millions of volts. This was much higher than anyone had previously been able to use in a laboratory. It was no use suggesting to Rutherford that I should attempt to bridge such a large gap and so I put forward an indirect method not requiring the use of very high voltages94.

Walton proposed accelerating electrons to be the bombarding particles by driving them many times around a circular electric field, picking up energy in each orbit, as in a modern betatron. He went on

When I suggested this method to Rutherford I was unaware that just two weeks previously in his presidential address to the Royal Society, he had stressed the importance of generating particles of energies higher than alpha and beta particles. So perhaps he was glad to find someone anxious to work in this difficult new field. I was lucky my suggestion was made at an opportune time95.

The betatron approach proved unproductive, and Walton then attempted, with no greater success, to build a linear accelerator, in which a high-frequency alternating voltage was applied to alternate members of a line of cylinders. It appeared that the way forward in the hunt for highly-accelerated particles must lie elsewhere.

The influence of Gamow

94 Walton (1984), p. 51. 95 Walton (1984), p. 51. 61

A vital clue to that way forward came from the insights of someone outside the Cavendish, the Russian theoretical physicist George Gamow. In 1928 he was working with Neils Bohr at his institute in Copenhagen. John Cockcroft’s account of this intervention, as recorded in the Hendry volume96, was brief.

Fortunately I was saved from attempting to rival the energies of the alpha particles from radium compounds by the timely visit of Gamow in November 1928, to expound his theory of the quantum theory of atomic disintegration which he had developed at Bohr’s laboratory at Copenhagen…

Other sources, including Cockcroft himself97, enlarged and corrected the story. Gamow did not in fact come to Cambridge until January 1929, at which time he addressed the Kapitza Club98. A copy in manuscript of his paper on the quantum theory of the radioactive nucleus, later to be published in Zeitshcrift für Physik99, had arrived at the Cavendish some three months earlier, which may account for Cockcroft’s confusion over dates. It was, however, Cockcroft’s mathematical training that enabled him to keep up with developments in the new world of quantum physics, which had its capital in Copenhagen, and to see the implications for the Cavendish of Gamow’s insight.

According to Gamow, the emission of alpha particles from a radioactive nucleus defied the strong attraction between the particles in the nucleus,

96 Cockcroft (1984), p. 74. 97 For example, Cockcroft JC The Cavendish Laboratory in the 1920s and 1930s. Typescript of illustrated talk in EBA. Hereafter to be cited as “Cockcroft (EBA)”. These recollections are amplified in Cockcroft (1964) and Cockcroft (1984). The three accounts contain much common material but differ in detail. 98 The Kapitza Club, a discussion group on trends and discoveries in contemporary physics, had been established in October 1922 by the Russian physicist Peter Kapitza. The Minute Book of the Society, which records, inter alia, Gamow’s presentation on the theory of the radioactive nucleus on 29 January 1929, is held in the Cockcroft papers at Churchill College. 99 Gamow (1928, 1928a). 62

which could be conceived as a potential well from which the particles had to escape. Yet, since quantum theory denied to any particle a precise definition of its location, a significant probability existed that any given alpha particle originating in the nuclei was actually outside the well, and so able to escape and be detected100 .

It is commonly conceded that Cockcroft’s stroke of genius was to turn this argument on its head101. If particles with less energy than represented by the potential barrier could escape from the nucleus, effectively by “tunnelling” though the barrier, then bombarding particles with insufficient energy to breach the potential barrier directly might likewise “tunnel” through it and cause transformations. In a memorandum to Rutherford102, Cockcroft presented calculations that incident energies as low as a few hundred thousand electron volts (eV) could initiate abundant transformations in light elements such as boron, and proposed that he build a particle accelerator which would create a proton beam of some 300 000 eV to investigate the matter.

Rutherford concurred. By the end of 1928, Cockcroft, soon joined by Walton, whose two efforts at accelerating particles had come to nothing, was assembling equipment for an accelerator, combining high voltage and high vacuum techniques. The initial thought was to use a Tesla coil to generate the high voltages to be applied to the tube, as Allibone had done and as teams in the US, unknown to the Cavendish, were doing.

100 Gamow was not the only one with this insight. See Condon and Gurney (1929). 101 Allibone argued that there were two papers, and in the second Gamow anticipates Cockcroft by suggesting that, in a reversal of the argument, a high probability existed of an alpha particle penetrating the potential barrier surrounding a light nucleus. 102 The later provenance of this crucial memorandum is debated. In Cockcroft (1964), Cockcroft maintained that it was recovered from Rutherford’s papers after the latter’s death. In his contribution to Hendry (1984) p. 161, Allibone claimed it lay unnoted in Rutherford’s papers until after Cockcroft’s death in 1967, when it was uncovered by Oliphant while compiling the Biographical Memoir on Cockcroft for the Royal Society. This memorandum is now among Cockcroft’s papers at Churchill College. Allibone maintained that neither he nor Walton, not being members of the Kapitza Club, knew of either the Gamow paper or the Cockcroft memorandum. 63

However there were two drawbacks with the Tesla coil. First it was very noisy, in the radio sense. The alternating current in the primary was generated by an oscillating or rotating spark gap which let loose an immense amount of radio-frequency static, enough to blanket radios and disturb experiments for many metres round.

The second problem was more serious. The output, like the input, was alternating. The peak voltage was attained for only a fraction of a second, and the particles being accelerated in the tube would end up with a whole range of velocities up to some maximum value set by the peak voltage. This was very different from the steady stream of particles of predictable energy produced by radioactive elements, and it would make much experimentation difficult.

What was needed was a technique to produce a high tension direct voltage, as steady as possible, such as with a transformer and a rectifier. To help Cockcroft and Walton move beyond primitive methods, Rutherford was able to bankroll them to an unprecedented level.

…. I started work on the program… using first an ancient spark coil as the source of my high voltages. However Rutherford was able to obtain a grant of £1000103 to purchase a High Voltage transformer costing £500, leaving £500 to build a rectifier and accelerating tube and other equipment. This was a very large sum for the Cavendish in those days104.

The link to Metro-Vick, represented in person by Cockcroft and Allibone, was strongly maintained at the technical level in this endeavour and, it

103 The source of this funding is presently unknown though, as with the support gained for Kapitza’s researches, and indeed the payment of Chadwick’s salary as Deputy Director of the laboratory from 1924, it is likely to have come from the Department of Scientific and Industrial Research. See Wilson (1983) p. 473. 104 Cockcroft (1964), p. 5. 64

may be argued, made it possible. The transformer came from Metropolitan Vickers105, as did oil diffusion pumps newly developed by CR Burch, replacing long-employed pumps using mercury, to produce the needed vacuum. These pumps were rated at 1000 litres per second, making the experiment, in Cockcroft’s words, “well in advance of the techniques in the field”. For this purpose, Burch had developed an oil with a very low vapour pressure, given the name Apezion. Burch also developed very low vapour pressure plasticine to seal joints in the vacuum system, the use of which took hours of painstaking effort to knead the plasticine into place many times over. The accelerating tube and rectifier stack were constructed from 30 cm bore glass tubing originally manufactured for use in petrol “bowsers”.

In May 1931 when forced to move to a new research room, Cockcroft and Walton rebuilt their equipment to deliver a beam at the much higher voltage of 500 000 eV106. To achieve this, Cockcroft and Walton invented (or, more correctly, re-invented) a “voltage doubling” circuit. The system, which used two rectifiers and two capacitors in each doubling stage, had been pioneered in the early 1920s by the German Heinrich Greinacher107 but it became generally known as the Cockcroft- Walton apparatus. The voltage was initially measured by seeing how far a spark would jump between two aluminium spheres.

The end result of this endeavour was the first successful operation of a particle accelerator for an investigation into nuclear physics108, and the

105 Allibone recorded that the design of this transformer was the work of Goodlet, who made it small enough to fit through both the Free School Lane entrance into the Cavendish and, even more importantly, through the door into the research room. See Allibone (1984). 106 Allibone later mused on this decision which seemed to reflect some uncertainly about the insights of Gamow which had given such impetus to the enterprise. That, and other evidence, led him to wonder how strong was the Cavendish belief in Gamow. See Allibone (1984), p. 162. 107 Greinacher (1921). 108 Cockcroft and Walton were not the first to successfully accelerate protons. The American Tuve and his colleagues published a paper in early 1931 describing the acceleration of protons 65

first “splitting of the atom”, achieved on 13 April 1932109. A beam of 2 micro-amps of protons110, fired at less than 200 000 volts into a target, produced a blizzard of scintillations on a screen coated with zinc sulphide, the detection method long favoured by Rutherford and only just beginning to be supplanted by electronic methods. The bright flashes were from pairs of alpha particles released when the lithium nuclei disintegrated. Soon after, nuclei of boron were broken into triplets of alphas.

It was for this pioneering work that Cockcroft and Walton would be awarded the Nobel Prize for Physics in 1951. The discovery was one of the three that illuminated the annus mirabilis of 1932 at the Cavendish, the others being the discoveries of the neutron by James Chadwick and of “” in cosmic rays by Patrick Blackett, with the help of GPS Occhialini. These also secured Nobel Prizes, for Chadwick in 1935 and for Blackett in 1948. Of these three key discoveries, only one depended on accelerator technology. The other two used naturally occurring radiation: alpha particles for Chadwick, cosmic radiation for Blackett.

The Cockcroft/Walton machine continued in use for several years, generating a steady flow of discoveries. Two significant developments added to its impact; the first being the building of a second accelerator tube to deliver particles into a cloud chamber, allowing the nuclear transmutations to be photographed by Phillip Dee and others, the second the availability from 1933 of bombarding beams composed of

to 1 MV using a Tesla coil. While the energy of the particles was measured, the beam was not used to try to cause nuclear transformation. In Weiner (1972), p. 29, Tuve admitted that they did not take the Gamow/Condon/Gurney findings seriously enough to attempt such experiments. 109 The technology of this investigation and the results obtained were reported in Cockcroft and Walton (1932/1, 1932/2, 1934). 110 Allibone recorded that the source of these protons used a design by Oliphant. See Allibone (1984), p. 161. 66

ions of “heavy hydrogen” or deuterium, which had been isolated as a component of “” by at Berkeley111.

In both of these, we find overlaps with the work Oliphant was soon to begin in partnership with Rutherford. An understanding of the reactions between deuterium ions (“diplons” or “deuterons”) was to be a major outcome of that research, leading to the discovery of two previously unknown nuclei of mass three: hydrogen three (“’) and helium three. The proof of the existence of the second of these was only made conclusive when it was caught in some of Dee’s cloud chamber photographs.

Rutherford at Metro-Vick

On 28 February 1930, between the inauguration of the Cockcroft-Walton venture and its brilliantly successful conclusion, Metropolitan Vickers initiated a new high tension laboratory at its Trafford Park headquarters, upgrading its facilities to handle tensions of a million volts. As described by Allibone112 , the official opening was a grand occasion, intended to strengthen the ties between industry and academia, particularly the Cavendish.

A huge gathering of some fifty pure and applied scientists was invited at the company expense to Trafford Park; over ten came from Cambridge alone, all members of the Cavendish. The photograph taken in the laboratory….includes four Nobel Laureates and three others who later won Nobel Prizes113

111 Cockcroft (1984), p. 77. 112 Allibone (1984), p. 163. 113 The photograph was reproduced in Allibone (1984). The caption showed that the Cavendish attendees included Wilson, Aston, Fowler, Cockcroft, Walton, Blackett and Kapitza, as well as Rutherford. 67

Rutherford was the guest of honour. His speech on the occasion was quoted at some length by Eve114. Rutherford began by reviewing the history of technology to generate high voltages, including friction machines and induction coils, leading to high voltage transformers. He recalled seeing a million volt discharge demonstrated at the 1904 St Louis Exposition.

However up until a few years ago, potentials higher than 200 000 volts were seldom available in the Laboratory and it is only by much improved technique in producing high vacua that we have been able to maintain such a potential on the terminals of a highly exhausted tube….

Rutherford then reprised much of his thought as expressed before the Royal Society more than two years earlier; the potential use of such technology to generate “a copious supply of swiftly moving electrons and charged atoms”, the limited energies available from such techniques when compared to the emanations of radioactive atoms. He was still looking for voltages of a very high magnitude.

I must point out that the ordinary university laboratory with its exiguous finance cannot hope to erect a cathedral-like structure such as we see today to house the high potential installation115. What we require is an apparatus to give us a potential of the order of 10 million volts, which can be safely accommodated in a reasonably sized room and operated by a few kilowatts of power. We require an exhausted tube capable of withstanding this voltage. I recommend this problem to the attention of my

114 Eve (1939), p. 337. 115 This sentence, which is contiguous with what follows, is not quoted by Eve. The source is Allibone (1984). 68

technical friends. I see no reason why such a requirement cannot be made practicable by the use of oil or air under high pressure, but these are problems for the future.

Allibone pondered116 if this statement represented a lingering doubt in Rutherford’s mind as to whether the puny 350 kilovolts he had approved for the Cockcroft-Walton venture would actually deliver any nuclear transmutations, whatever Gamow might have prophesied, and that much larger voltages would be needed. Certainly no reference was made to the enterprise then underway. As noted earlier, the decision to upgrade the apparatus to 500 000 volts or more may have reflected similar uncertainly by Cockcroft.

Whatever the expectations, Cockcroft and Walton did deliver. That, underlined by the success of the even lower-voltage machine that Oliphant commenced soon afterwards, seems to have influenced Rutherford’s attitude toward the quest for higher voltages. Elsewhere, especially in the United States though not in the UK other than at the Cavendish and at Metro-Vick) research teams were using a variety of techniques to generate the high energies seen as necessary for nuclear physics studies. As Dahl noted117, these included ambitious efforts in Germany to capture the potentials involved in lightning. When those experiments proved fatal to one of the researchers, an attempt was made to generate artificial lightning through the discharge of banks of capacitors in parallel (a “surge generator”). Another, the “cascade transformer” being refined at the California Institute of Technology, has already been mentioned. Dahl might have added to the list the early development of the Van de Graaff generator, begun in 1929. Whether

116 Allibone (1984), p. 163. 117 Dahl (2002), p. 24. 69

any of these efforts had been stimulated by Rutherford’s 1927 Royal Society address, as Allibone proposed118, is yet to be assessed.

Regarding these other endeavours, and their impact on him, Cockcroft recalled

I made a grand tour of the newly established nuclear physics facilities in the United States in the summer of 1933119; visiting Van de Graaff120 in Cambridge, Massachusetts, Tuve121 in Washington, Lauritsen122 in Cal. Tech. and Lawrence123 in Berkeley. I became convinced that we ought to build a cyclotron and on my return I tried to persuade Rutherford to authorise this. However, owing to the success of the experiment with 200 KV

118 Allibone commented “ It can also be said, I think, that the Professor’s words at his Presidential Address to the Royal Society alerted scientists in other countries to think about artificial disintegration and Cockcroft and Walton only just “made it” in time. Others were building apparatus and were ready to repeat the results achieved in Cambridge almost immediately afterwards…” Allibone (1984), p. 162. 119 Background to the various facilities visited by Cockcroft and described below can be found in Weiner (1972). 120 Robert Van de Graaff at the Massachusetts Institute of Technology had since 1929 been developing technology that accumulated static electric charge, carried on insulating belts, to achieve potentials exceeding 1 million volts. Cockcroft saw a new twin generator installation producing 2 million volts, though at the time the accelerating tube was not in place. 121 Merle Tuve and his colleagues at the Carnegie Institute in Washington had first used a Tesla coil, reaching 5 MV with oil under pressure as insulation. They later moved to a Van De Graaff generator to accelerate protons to 600 KV and confirm Cockcroft and Walton’s findings on the disintegration of lithium. In Weiner (1972) p. 29. Tuve recalled having an ambition around 1926 of accelerating protons to 1 MV in order to explore whether the Maxwell laws of held with the confined space of the atomic nucleus. He had planned to come to the Cavendish on a one-year fellowship to work with Rutherford and undertake this endeavour but was convinced to change his mind on the grounds that the work could not be done in a year. It took five years. 122 Charles Lauritsen and his colleagues at the California Institute of Technology were energising a double-ended x-ray tube, similar to those used in Cambridge, with a 1 MV cascade transformer to accelerate electrons. From 1934, the system was used for nuclear physics studies though the acceleration of deuterons but later abandoned in favour of a Van de Graaff system. 123 Ernest Lawrence and Stanley Livingstone at the University of California Berkeley had been developing a new type of accelerator called the cyclotron since 1930. It did not require the generation of high voltages, since the particles were subjected to repeated accelerations. See Seidel (1992). At the time of Cockcroft’s visit, protons were being accelerated to 1 MeV and used to repeat and extend Cockcroft and Walton’s work. A larger machine was nearing completion. Of the various particle-accelerating methods he observed in his trip, this was clearly the one that impressed him most and the path he wanted to follow. 70

protons and deuterons, he did not believe it was necessary to go to higher energies and I was unsuccessful. Later when he received a benefaction of £250 000 for the Cavendish124, and the cyclotron was further developed, he agreed to our building one.…125.

Summing Up

This then was the broad context of experimental nuclear physics around 1932, at the time Oliphant began to design and build particle accelerators, the activity which was to largely define his career thereafter. We have canvassed here the major elements of this context: the dominance of the Cavendish in the field in the UK and, until the early 30s, globally; the dominance of nuclear physics within the Cavendish; the growing realisation that the way forward lay in larger and more complex apparatus, particularly in the harnessing of high energies; the already significant interaction between researchers and industry in the provision of that equipment; and the first efforts at shaking loose significant amounts of public and private money to support this research.

Some significant trends into the future were already evident. These included the inauguration by Ernest Lawrence and his Berkeley colleagues of the cyclotron, which would soon take over from high- tension machines as the dominant method of accelerating particles for nuclear physics work, at least at the high-energy end. Before the decade was out, Cockcroft would have begun to build one at the Cavendish, as would Chadwick in Liverpool and Oliphant in Birmingham, all based on Lawrence’s designs. Nonetheless, as we shall see, Oliphant would seek to extend the life of the existing (Cockcroft-Walton) approach with his

124 This was the Austin bequest, announced in May 1936. 125 Cockcroft (EBA). These recollections are amplified in Cockcroft (1964) and Cockcroft (1984) 71

“basement accelerator” and then to inaugurate a new high-tension laboratory bought mostly complete from a commercial supplier off the shelf”.

Oliphant would be much involved in the use of deuterium as a tool of research and make significant discoveries with it, and as a consequence begin to build an important relationship with Lawrence. On the other hand, Oliphant would have little to do with the newly-identified . Others would, leading to the discovery of artificial radioactivity in France and Italy and then of nuclear fission in Germany in late 1938, soon confirmed elsewhere. The drive to harness that phenomenon for military use would give rise to an immense endeavour which would involve Oliphant and most of his colleagues, and greatly shape his postwar activities. Oliphant’s own endeavours through the 1930s up to the outbreak of war in 1939 will be the focus of the next chapter.

72

CHAPTER FOUR Building “capital”: Oliphant in the 1930s

In seeking to assess Oliphant’s career prior to the initiation of his synchrotron project in 1946, we can choose from a number of perspectives. We could for instance concentrate on his contribution to fundamental knowledge, especially in nuclear physics. While some of those contributions are of great significance, most particularly his illumination of the nature of the reaction between particles of heavy hydrogen (deuterium), the research was undertaken over a relatively short period of time, spanning half a decade in the early to mid-1930s, and generated barely a dozen scientific papers. Compared with many of his contemporaries, Oliphant’s list of scientific publications is noticeably short.

Another line to follow would be the manner in which Oliphant’s growing reputation was marked by a series of appointments and honours. We could follow his upward path through the hierarchy of the Cavendish Laboratory, his securing of the Fellowship of the Royal Society at a relatively young age (36), his appointment to the Chair of Physics at the University of Birmingham at almost the same time, the award of the Hughes Medal of the Royal Society during the war and so on. We can link that to his heightening profile among his scientific peers, and to the beginnings of enduring friendships with leading figures such as Cockcroft, Chadwick and Lawrence.

A third approach, and one on which we will place significant emphasis, deals with his growing technical expertise and his burgeoning experience in the design and construction of high-energy nuclear

73

physics equipment, a skill and insight which has been described as “mastery”. Throughout this period Oliphant was deeply involved in the construction of three substantial pieces of apparatus for experimental nuclear physics, characterized by increasing technical complexity, achieved energy, size and cost. That course also engaged him increasingly with industry.

Elements of these three approaches can be brought under a fourth heading, “the building of capital”. To undertake so substantial and innovative a project as the proton synchrotron, Oliphant needed to draw on a broad range of strengths beyond his own undoubted scientific insight and technical expertise. He had to leverage influence based around reputation and contacts, to exercise the leadership that can motivate and unite a large team, and to carry funders, bureaucrats and industrial suppliers with him as he moved into uncharted territory.

That he was able to do all that depended not only on his personal qualities but also on the “capital” which he had built up over more than a decade. That concept of “capital”, in its many facets, will serve as an organising principle and an analytical tool as we chronicle Oliphant’s progress between 1932 and 1939 in the current chapter, and during the war years to 1945 in the chapter to follow.

Getting started in nuclear physics

It was the success of Cockcroft and Walton that led Oliphant into the design and construction of particle accelerators, a line of work that was to largely define his career thereafter. Following the first splitting of the atom, Rutherford enlisted the assistance of Oliphant to build a similar accelerator, with lower voltage and higher current, to continue the investigations. Built in a basement room with a low ceiling, the

74

accelerator had a horizontal accelerating tube. The apparatus went through two generations. It was from these machines that Oliphant gained vital expertise in the design, construction, maintenance and operation of particle accelerators, expertise that was to support him through five similar endeavours over the next 30 years, including the Birmingham proton synchrotron. He also honed his native skills in innovative problem solving.

Cockcroft briefly recalled Oliphant’s involvement as follows:

Immediately after our discovery, Rutherford with his great strategic sense, diverted Mark Oliphant from his experiments on positive ion impact to build a 200 KeV accelerator for the joint use of Rutherford and himself, and great improvements in the ion source were incorporated, resulting in a 1000-fold increase in the number of disintegration particles of energies between 20 000 and 200 000 volts. One of their first results was to confirm that boron was split up by proton bombardment into three alpha particles. Shortly after this they were able to extend their work by the use of heavy hydrogen nuclei, having been presented with a cubic centimeter or so of heavy water by GN Lewis. They were then able to observe that the D-D reaction led to the production either of tritium or a proton, or alternatively, to helium 3 and a neutron. This discovery has tremendous political and scientific consequences in the 1950s, and provides the basis of most of the research on controlled thermonuclear reactions which are taking place today126.

Oliphant’s own recollections of his initiation were more detailed.

126 Cockcroft (1964), p. 7. 75

Immediately after the first observations of Cockcroft and Walton, while my wife and I were spending the weekend with the Rutherfords at their cottage Celyn, at Nant Gwyant in North Wales, Rutherford told me he wanted to exploit the new technique as fully as possible at the Cavendish. He suggested that I give up my work on positive ions and use my experience in collaborating with him to set up a second accelerating system. Naturally I accepted eagerly such an arrangement. We decided to aim for greater accuracy in measurement of the energies of the products of these transformations. So I designed and constructed a simple version of the Cockcroft-Walton apparatus, for a maximum energy of 200 KeV, and with an improved form of canal-ray tube, giving 100 micro-amperes or more of protons127.

In view of Oliphant’s later quest for the highest obtainable energies of bombarding particles, it is interesting to see that at this time beam current, that is the highest possible number of bombarding particles, was the goal.

An insight into the burgeoning reputation of Oliphant as an equipment builder, and therefore Rutherford’s choice as a collaborator, was provided by his Cavendish colleague AE Kempton in a letter to the author dated 22 December 1980128. Kempton worked with Oliphant in the basement between 1933 and 1936.

I think that one reason, perhaps not (sic) the most important, why Rutherford should respond to Oliphant would be one of the reasons he so strongly supported Kapitza. Both Oliphant and Kapitza possessed, to an unusual degree, a facility with

127 Oliphant (1972), p. 104. 128 Kempton to Ellyard, 22 December 1980. EBA. 76

apparatus which Rutherford almost certainly lacked. All Rutherford’s work had been done with extremely simple apparatus. Both Oliphant and Kapitza had the ability to dominate apparatus which, for the time, was very complicated and sophisticated: in gardening terms, they both had “green fingers”.

Oliphant’s extraordinary ability to design, make and operate apparatus is shown in miniature in the plate facing page 180 of Feather’s biography of Rutherford. In 1934, Rutherford gave a series of lectures at the Royal Institution, and the apparatus used is shown in this plate; this was designed by Oliphant, much of the work being done by Rutherford’s assistant George Crowe.

It was only two years earlier (1932) that Cockcroft and Walton had effected the first artificial disintegration, and here was a portable apparatus, self-contained and complete with counting equipment which would fit into quite a small van. This was a technical tour de force.

There was probably another less-generalised attraction for Rutherford. Though Oliphant had not been working in nuclear physics up to this time, he had some experience relevant to the new task. Much of his research over the previous five years had involved working with discharge tubes and the production of positive ions which were used to bombard metal surfaces and liberate electrons129. Positive ions (hydrogen ions to begin with) would be the bombarding particle in the new work too, and Oliphant had developed technology to produce them in abundance.

129 See for example Oliphant (1930). 77

The accelerator in the basement 130

The site chosen for the new accelerator was in the basement, the stone- flagged lowest floor of the Rayleigh wing of the Cavendish. It was a prestigious site; in the room next door, used by Oliphant as a darkroom, office and store, Lord Rayleigh, second director of the Cavendish, had made the first definitive measurement of the ohm, the unit of electric resistance. A brass plate on a pillar attested the fact. In the same room, JJ Thomson had identified the electron, and Rutherford and Chadwick had transmuted matter artificially for the first time. Access to the accelerator room was through a space occupied by one of the most brilliant men at the Cavendish, Geoffrey Taylor, his research assistants and students.

The room housing the accelerator had a low ceiling; as a result, the beam tube had to be placed horizontally, rather than vertically as in the Cockcroft-Walton machine. This was by no means the only difference, even allowing for differences in design to provide a much higher beam current at a noticeably lower voltage.

A major advance was the protection provided against X-rays created by the discharge tube. Oliphant had been shocked to find the room housing the Cockcroft-Walton accelerator was bathed in X-rays when the machine was running. The radiation was so intense that the bones of the hand could be revealed on a barium platinocyanate screen, which fluoresced yellow-green under X-ray bombardment, held at any orientation almost anywhere in the room. The Cockcroft-Walton machine was shielded only by a head-high wall of loose bricks. Oliphant, on the

130 The chief source for the following account is an interview with Oliphant conducted by the author on 21 May 1980 (transcript in EBA) , supplemented by material from Oliphant (1972) p. 104-106, Oliphant’s correspondence with Rutherford, Lawrence and others and the published papers resulting from the research. See also Cockburn and Ellyard (1981), Chapter Five. 78

other hand, had a local builder construct a solid brick wall across the room, with the discharge tube behind it and the beam tube emerging through a hole in the wall131.

The discharge tube was energised by 60 kilovolts DC supplied from a rectifier, built inside from large-diameter “petrol bowser” glass tubing. Current to the rectifier came, via a high-tension transformer from an old X-ray machine, from an AC motor run in reverse so that it acted as a generator. All this stood on a pine laboratory bench resting on hollow porcelain insulators found among junk near Lincoln’s stores132. The motor/generator in turn was activated by a belt drive from a similar motor on the floor.

It is not on record when construction began, but progress can be traced in outline through the few surviving letters between Oliphant and Rutherford. The fragmentary nature of this record is a consequence of most of the letters being written when Rutherford was absent from the Cavendish, for example on holiday at his retreat at Celyn in Wales.

Writing from Celyn on 23 June 1932133 (apparently in reply to a letter from Oliphant which has not survived) Rutherford reported that he had placed an order worth £52 with Metro-Vick for the pumping system the apparatus needed, and that he would be inquiring from Allibone as to the whereabouts of the “original 200 000 volt transformer” which he thought would be “temporarily sufficient for the purpose” (presumably to energise the beam tube). Rutherford’s concern was for good control

131 It is interesting to note that no such precautions were taken with the flux of neutrons which emerged from the target in later experiments with the accelerator. Oliphant and his co-workers often worked within a metre of the target so exposing themselves to radiation since shown to have significant biological effects. 132 “Fred” Lincoln was the head technician and keeper of the stores in the Cavendish at the time. Builders of equipment made much use of discards from the stores, and other accumulated odds and ends, since new materials were in short supply. 133 Rutherford to Oliphant, 23 June 1932. EBA. 79

over the discharge and accelerating voltages. He noted that “difficulties increase with voltage”. That the “original transformer” did not suffice is indicated by a letter from McKerrow at Metro-Vick to Rutherford dated 19 July134, reporting the dispatch of a 100 000 volt transformer with a primary winding “made suitable for a 210 volt 50 cycle supply according to Mr Oliphant’s instructions”.

Early in August, Rutherford on summer holiday wrote to see how things were going. “I expect there will be a good deal of time required to get everything working alright. I hear from Cockcroft that they got some results with Li+ but wait for information of the range of the alpha particles. Let me know any results”135. The letter crossed one from Oliphant. “I have little to report other than slight hitches in the preparation of the apparatus”.136 The “slight hitches” were mostly electrical. At the time the high tension apparatus stood on a pine table with its legs immersed in oil. This failed to withstand 100 000 volts, with one leg conducting enough current to allow it to catch fire. An oak framework, previously used elsewhere, was substituted and withstood 200 000 volts without problem.

A second difficulty involved the belt driving the generator in the high- tension system. A belt of rubberized canvas had been prepared but this proved too conductive, with the belt becoming warm enough to melt the rubber compound. An oil-soaked leather belt was used in its place. That being fixed, attention turned to the transformer powering the discharge tube. This was wound for 95 cycle current but fed from a 50 cycle generator. This mismatch wasted considerable power but there was little that could be done with the equipment at hand. The rectifier was working well, but an occasional spark-over on the outside of the glass

134 McKerrow to Rutherford, 19 July 1932. EBA. 135 Rutherford to Oliphant, 8 August 1932. EBA. 136 Oliphant to Rutherford, 7 August 1932. EBA. 80

tube when the potential exceeded 200 000 volts resulted in the burnout of one or other of the heaters in the oil pump. That had to be rewound.

This litany of minor troubles could not cloak the progress being made. Oliphant was able to report that the high tension system was complete and in good working order, supplying up to 200 000 volts and 50 milliamps (for short periods even 100 milliamps), the much more efficient discharge tube was ready to generate positive ions and the accelerating system was complete. The in-house electronics guru Wyn- Williams had the particle counter with its electronics in hand. “We are only waiting for the pump connection to try out the whole thing. With luck I hope to be able to report preliminary measurements by the end of the week.” Barely three months had passed since the invitation from Rutherford to participate in this enterprise. Oliphant would never build an accelerator so quickly again.

Another few months were needed to fine-tune the machine and gather the first results. Rutherford and Oliphant did not produce a paper until the spring of 1933137, and that was mostly a confirmation of the findings of Cockcroft and Walton, such as the breakup of lithium and boron under proton bombardment. They were however ready to move into an exciting new field, as did Cockcroft and Walton, sparked off by the first minute samples of “heavy water” donated by GN Lewis from Berkeley in the summer of 1933.

Oliphant’s own recollections of the onset of the new research were as follows.

In 1933, GN Lewis from Berkeley visited the laboratory. He presented Rutherford with about 0.5 cc of almost pure heavy

137 Rutherford and Oliphant (1933). 81

water138, which he had concentrated electrolytically. It was sealed in three tiny glass ampoules. After much discussion I reacted one of these with a film of potassium on the walls of an evacuated glass bulb [so liberating the heavy hydrogen]. Meanwhile I asked my colleagues to try the effect of mixing hydrogen with a large excess of helium, to see if such a mixture gave a reasonable beam of protons when used in the canal tube of our apparatus. We were pleased to find that a mixture of five parts of helium with one part of hydrogen gave precisely the same proton beam as pure hydrogen. We also arranged to collect the gas from our pumping system, and to purify it by freezing out all components other than helium and hydrogen, in a glass trap immersed in liquid nitrogen boiling under reduced pressure, to obtain as low a temperature as possible. We were surprised when the [impurities] solidified to a white crystalline substance, but the method worked and we were able to use our limited supply of deuterium gas over and over again139.

Notwithstanding Oliphant’s efforts to conserve the deuterium, more would soon be needed.

It was clear from our initial experiments with the sample of heavy water from GN Lewis that deuterons were of great interest for experimental work on nuclear disintegration. I cabled Cockcroft, who was on a visit to America, to ask him, while with Lawrence in Berkeley, to try to obtain some more heavy water from GN Lewis.

138 “Heavy water” is the name given to water in which some hydrogen atoms have been replaced by atoms of the naturally-occurring “heavy hydrogen”, which has twice the atomic weight. With the new understanding following Chadwick’s discovery of the neutron, it could be stated that the nucleus of heavy hydrogen contained a neutron as well as a proton. Heavy water makes up about one part in 6000 of naturally-occurring water. It had been discovered by American Harold Urey in 1932. 139 Oliphant (1972), p. 108. Oliphant’s ingenuity and technical skill is evident from this account. 82

Lewis was kind enough to let him have two gallons of water, with two per cent deuterium, for $10. He had trouble convincing the customs on his return that the can contained only water. When he asked Rutherford to refund the ten dollars, Rutherford was angry with him for spending that amount without permission. Dr Harteck concentrated the deuterium content further by electrolysis. In this way we obtained enough deuterium for all users until heavy water became available commercially from Norway.140

This story also helps set a timeframe for the commencement of Oliphant’s research with deuterium. As we noted in the previous chapter, Cockcroft had been in the US in June 1933, so the work must have begun by then. In preparing for the experiments with heavy hydrogen, Oliphant had redesigned and rebuilt his machine to such an extent that it may be said to constitute a second generation. As he reported in a new paper

By the addition of another 100 000 volt transformer in tandem and the use of appropriate condensers, the DC voltage available was raised from 200 000 to 400 000 volts. The main change however consists in the use of a horizontal rather than a vertical discharge tube. In place of glass, a corrugated porcelain wall bushing capable of withstanding high voltages has been used to insulate the positive electrode, while the earthed metal casing forming the negative electrode projects through a brick wall …. The oil cooling circulation has been improved as the electrodes now cannot be cooled by radiation alone. As before a magnetic field is used to sort out the various types of ions generated in the discharge tube. The use of the horizontal [beam] tube has many advantages not only for assembling controls at convenient points

140 Oliphant (1972), p. 114. 83

but also in the ease of handling the counting apparatus and absorbing screens.

The new installation has worked smoothly and satisfactorily and we have been able to increase the number of disintegration particles by a factor of 10 to 50. The thick brick wall acts as a complete screen for the X-radiation generated by the apparatus141.

The quoted increase in particle flux represented an enhancement factor of 10 000 to 50 000 over the output of the original Cockcroft-Walton apparatus of a mere two years earlier. Such was the pace of technical advance.

Heavy hydrogen and Ernest Lawrence

Given our overall focus on Oliphant as a machine builder, our interest here is predominantly with techniques rather with results, and so it is not appropriate to dwell at length on the new understandings that arose from experiments with the basement accelerator. We can briefly state that from several years of experimentation undertaken by Oliphant and a number of co-workers under the inspiration of Rutherford arose three significant findings: (a) the nature of the interaction when particles of heavy hydrogen were made to collide; (b) the existence of two previously unknown of mass three, namely hydrogen three or tritium (containing one proton and two neutrons) and helium three (two protons and one neutron); (c) a refinement of the masses of the light elements. These insights constitute Oliphant’s only truly important contributions to scientific knowledge.

141 Oliphant, Harteck and Rutherford (1934). It is worth noting that Oliphant’s name came first on this paper (and on those that followed), unlike the first paper in 1933. Clearly he was becoming the driving force in this research. 84

It was also in this context that the names of Oliphant and Lawrence142 first became known, each to the other. This began a life-long association and friendship of such importance that Oliphant later named Lawrence as the second of the “Two Ernests” who so profoundly influenced his career 143.

Lawrence and his team at Berkeley had already used “deutons”144 to bombard targets, accelerating them in a cyclotron as they had previously done with protons. A dozen different light element targets all gave the same results, copious numbers of protons with a range of about 18 cm in air. Cleaning the targets to make them free of contamination did not affect these uniform results, so the only explanation appeared to be that the deuteron was a fragile or unstable particle that broke up on impact or in the powerful force fields surrounding the nucleus. Embedding deuterons in the targets (for example replacing some atoms of hydrogen in, say, lithium hydroxide with deuterons) produced the same protons, though much more copiously. As late as December 1933, Lawrence was sticking to his “fragile deuteron” thesis, which had apparently been first suggested by Robert Oppenheimer145.

By now, the physicists at the Cavendish, with Oliphant playing a leading role, had accumulated a lot of data, and were ready to promote an

142 Ernest Orlando Lawrence (1901-1958) American physicist. At this time Lawrence was Director of the Radiation Laboratory at the University of California at Berkeley. Winner of the Nobel Prize for Physics in 1939 for the invention of the cyclotron. For background in Lawrence, see Childs (1968), Heilbron and Seidel (1989). 143 Oliphant (1966). 144 The name the Berkeley team had given to particles of heavy hydrogen or “deuterium”. There was considerable debate over terminology in this research. Rutherford had named the new isotope diplogen and its particles diplons. Berkeley preferred deuterium and deutons. Ultimately, all settled on deuterium and deuterons. We will use that name from now on, except where “diplons” is used in a quotation. One wag reportedly suggested that Rutherford had been happy to accept the Berkeley nomenclature once his own initials (“er”) had been inserted in it. 145 Oppenheimer was of course later to play a leading role in the quest for the atomic bomb. 85

explanation of their own. The source of the protons, they argued, was an interaction between deuterons in the beam and deuterons in the target, either because the target contained deuterons in place of some hydrogen atoms, or it had become contaminated with deuterons from the beam.

The key observation underlying the explanation was the presence of a second group of particles that had previously not been noted. These were singly charged and therefore were hydrogen ions, but their range was short, only 1.6 cm. Within the limits of measurement, they were equal in number to the protons. Writing up the results for publishing a month later, Oliphant and Rutherford wrote

On these data, it is natural to assume that the particles are emitted in pairs opposite to each other, and that the difference in range arises from a difference in mass, and therefore in velocity and energy. The simplest reaction which we can assume is

2 2 1 3 146 1D + 1D gives 2He4 gives 1H + 1H

So the most obvious way to interpret the result was as evidence of the existence of a third isotope of hydrogen, previously undetected. Each such particle, together with a proton, resulted from the breakup of an unstable (since excited) helium nucleus, itself produced by the union of two deuterons. The particle was soon given the title of a “triton” and the element called “tritium”.

146 In this equation and the one following, D represents deuterium, H stands for hydrogen, He is helium. In each symbol, the leading subscript is the “atomic number” (the number of protons in the particle). The following superscript is the ”atomic mass” (the total number of protons plus neutrons). 86

It was soon found that the deuteron on deuteron impacts produced something else as well: a large amount of radiation so penetrating it passed through 20 cm of lead. This proved to be a flux of neutrons, roughly equal in number to the long range protons, and suggested a second mode of breakup of the excited helium nucleus, roughly equal in probability with the first mode. If a neutron was one product, the other must be carrying away two protons and a neutron; in other words it was a nucleus of a previously unknown isotope, helium three. In shorthand

2 2 4 1 3 1D + 1D gives 2He gives 0n + 2He

The existence of helium three could be implied from the experimental data, and had been previously proposed by Rutherford from other observations. But energy considerations suggested that its range would be very short, less than 1 cm, and therefore all but impossible to detect directly by the methods being used, which required the particle to pass through a thin mica window and into the counting chamber. As late as April 1935, Oliphant was still admitting the particle had not been seen, though he remained convinced it existed.147 That the particle did indeed exist was finally proven when Phillip Dee photographed its fleeting track in a cloud chamber.

This matter is worth some further examination in view of the light it throws on the reliability of Oliphant’s recollections of this and other events. It is not always possible to check his often vivid accounts against other evidence. Here it is possible, and, perhaps significantly, the account fails the test. In Recollections, Oliphant devoted two pages

147 Undated typed lecture notes by Oliphant entitled Conference in Leiden 1935. EBA. The dating of this conference as being in April 1935 comes from contemporary correspondence, also in EBA. 87

to a colourful anecdote devoted to the “discovery” of helium three148. According to this account Oliphant, the very short-range particles had already been found (in addition to the protons and the tritons), but their meaning was puzzling. Oliphant recalls being woken in the middle of the night by Rutherford who proclaimed the short-range particles to be helium particles of mass three. When asked by the surprised Oliphant as to the reasons behind the statement, Rutherford roared “Reasons? Reasons? I feel it in my water!”. “He then told me that he believed the helium particle of mass three to be the companion of a neutron, produced in an alternative reaction that just happened to occur with the same probability as the reaction producing protons and tritons”.

Oliphant concluded by recounting that the next morning additional work was undertaken to accurately measure the energies of the particles involved and thereby to prove the Rutherford hypothesis. Oliphant drafted a brief paper to Nature, describing the work. This was “scribbled all over” by Rutherford, retyped and sent off. Dated 9 March 1934, it was published on the 17th149.

Despite its wealth of detail, this anecdote is of limited value as a source of reliable information about the actual course of events described. It recounts observations which were not in fact made, and reverses the direction in which the relationship between the neutrons and the helium particles was established. It is valuable as an account of the sense of excitement that commonly accompanies scientific discovery, and of the impact that Rutherford had on those around him, but counsels caution in accepting at face value Oliphant’s accounts of events, however apparently plausible and embellished with detail.

148 Oliphant (1972). This account was obviously well rehearsed. The author can attest to hearing it in almost identical terms on several occasions over a decade or more. 149 See also Rutherford and Oliphant (1933). 88

Given the key role Oliphant had played in accumulating the evidence, it was appropriate that Rutherford had charged him with communicating the Cavendish findings to Lawrence and his team at Berkeley. Oliphant’s pithy note to Lawrence was included in a letter from Rutherford dated 13 March 1934150. As Oliphant wrote in The Two Ernests151:

My note went as follows

You may have heard of the experiments which we have carried out during the last week or two on the effects observed when heavy hydrogen is used to bombard heavy hydrogen. As I believe that these are intimately related to your own work I would like to tell you what we have found.

The letter went on to give details of the results and of their interpretation as due to two competing reactions, the first leading to the production of hydrogen of mass three and a proton, with ranges of 1.6 cm and 14.3 cm respectively, and the second to helium of mass of three and a neutron.

We suggest that your results may be explained as due to the bombardment of films of [deuterium] and [deuterium] compounds. Our results with [carbon], [beryllium] etc could all be accounted for by the presence of less than one monomolecular layer of [deuterium]…

150 Rutherford to Lawrence , 13 March 1934. LPBL. 151 Oliphant (1966) (1), pp. 48-9. 89

On 4 June Lawrence replied to my note, saying that the late answer was due to his desire to be able to send some news of interest.

Your experiments with diplons152, together with Cockcroft and Walton’s recent work, have certainly cleared things up in beautiful fashion. There can be can no longer be any doubt that our observations which we ascribed to diplon breakup, in fact are the results of reactions of diplons with each other.

Lawrence had already replied to Rutherford’s letter153, confessing “It is difficult for me to understand how we could have failed to detect the effect of diplons on each other”. There was also an acknowledgment of Oliphant’s yet-to-be answered note. “Please tell Dr Oliphant that I appreciated his letter very much and that I will be writing to him directly before very long.”

Separating isotopes

One of the issues tackled by Oliphant and his colleagues tending the basement accelerator was the behaviour of lithium under proton bombardment. Results to date had not been conclusive, largely the consequence of naturally-occurring lithium being a mixture of two isotopes, mass six and mass seven, each with its distinctive pattern of disintegration. This matter is relevant to later events because of the manner in which Oliphant solved the puzzle. With young research student ES Shire doing most of the work, equipment was assembled to

152 As previously noted, this was the name the name the Cavendish researchers had initially settled on for particles of heavy hydrogen. 153 Lawrence to Rutherford, 10 May 1934. LPBL. 90

separate the isotopes, so each could be subjected to bombardment by itself. In this way, the modes of disintegration tentatively ascribed to each isotope could be confirmed. The first account of the work appeared in Nature one week before the deuterium on deuterium results154.

While Shire came up with his own technique for the separation, Oliphant already had some experience with such matters. At the time Rutherford had called him from his work on positive ions, he had been in the middle of setting up a semi-circular mass spectrograph, designed to separate the isotopes of potassium in order to show that the relatively rare K40 was the source of the element’s weak radioactivity. It is likely that in this work, Oliphant had used a “mushroom magnet” which generated an annular magnetic field. This work had two probable consequences in later years. It clearly fed into his interest in using such magnetic fields to separate the isotopes of , once that became of military significance after the outbreak of war, and, as we shall see, there is some evidence that it influenced his thinking when devising the “synchrotron principle” in the closing months of World War II.

The influence of the Royal Society Mond Laboratory

The building of the Royal Society Mond Laboratory in Cambridge in the early 1930s was arguably the first event to make Oliphant aware that substantial funds could be shaken loose to support new equipment. It was not inevitable that the Cavendish physicist had to rely on the meagre annual grant from the University. The development also showed what changes a charismatic individual could bring about, the establishment of the Mond being the direct consequence of the

154 Oliphant, Shire and Crowther 1934/1, Oliphant, Shire and Crowther 1934/2. 91

presence at the Cavendish of the ebullient Russian Peter Kapitza155. It also may be said to represent one of the earliest examples, at least in Britain, of the trends that later, when more fully developed, were dubbed Big Physics.

John Cockcroft was involved in Kapitza’s work almost from the start.

In the late 20s and early 30s, low-temperature physics was being actively developed in Cambridge by Kapitza, who had joined the Laboratory in 1921. When I joined the Laboratory in 1924, I was asked by Rutherford to help Kapitza, and having been previously partly trained as an electrical engineer156, helped with the installation of the twenty megawatt pulsed electrical generator which was used to produce magnetic fields of up to 320 kilo- gauss in small coils. I used my mathematical training to work out the theory of the strong forces in the small coils, which often burst under the stress, and after that, I took part in the design of the first hydrogen liquefier in Cambridge, and Kapitza went on to design his well known expansion-engine type of helium liquefier….

Kapitza’s helium liquefier was installed in the Royal Society Mond Laboratories newly built by the Royal Society to house his equipment157.

155 Oliphant recalled Kapitza “as the most colourful figure in the Cavendish when I arrived….” Kapitza had lost his first wife and family during the Revolution. He had advanced rapidly following his admission as a research student under Rutherford in 1921, becoming a Fellow of Trinity College in 1925 and a Fellow of the Royal Society in 1929, while still a Soviet citizen. See Oliphant (1972), p 90. For a broader perspective on Kapitza’s life and work, see Boag et al (1990). 156 Cockcroft had worked at Metropolitan Vickers prior to joining the Cavendish. 157 Cockcroft (1964), p 12. 92

It has been suggested that the building of the Mond Laboratory was motivated by more than the need to house Kapitza’s equipment.

By the 1930s conditions in the [Cavendish] Lab had become too much for many researchers, who were abandoning the Lab to work in more comfortable conditions elsewhere. Kapitza persuaded Rutherford to use £15,000 from the Royal Society to build the Mond Laboratory. C.H. Hughes designed the building with large communal areas rather than long corridors, a design which promoted chance meetings between research students as they moved about the building. The ideas that were exchanged during such chance meetings were found highly beneficial to research, and the layout of the Mond inspired the design of the new laboratory in West Cambridge, 40 years later. In February 1933 the Mond was opened.158

Oliphant recalled the origins of the Mond Laboratory in the following terms;

Kapitza’s tremendous energy, drive and self confidence appealed to Rutherford, as did his voracious appetite for understanding of physical science. This insistent drive to know more led to his founding of a discussion club which met regularly to discuss the major advances and problems in physics. At first Kapitza received his income from the Russian purchasing office in London. Later through Rutherford, Kapitza’s work was financed by temporary grants from D.S.I.R.

In April 1930, Kapitza wrote to Rutherford pointing out that he had been working on strong magnetic fields in the Cavendish for eight years. He asked that his position be made more secure through

158 http://www-outreach.phy.cam.ac.uk/camphy/laboratory/laboratory12_1.htm 93

some definite, continuing appointment, which will enable him to expand his work and go ahead without continual worry about the future. In a letter to Rutherford he added that he had received offers from another place, but he did not specify which, that was prepared to take over the whole of the equipment of his magnetic laboratory at a fair price, but that he would rather stay in Cambridge, with Rutherford, to whom he owed so much, if satisfactory arrangement be made. It seems that this letter precipitated the moves which led to Kapitza’s appointment to a Royal Society Professorship in 1933, and to the provision of a grant to build the Mond Laboratory for his work. The new laboratory was built beside the Cavendish, on the site of the old electrical substation.

Oliphant is not quite accurate here. According to Wood159, Kapitza’s appointment to a Messell Professorship by the Royal Society was in 1930. Wood also explains the origin of the name of the Laboratory. The £15 000 needed to build and equip the facility came from a bequest made to the Royal Society by the industrialist Ludwig Mond160.

We have noted previously161 that a copy of the program for the opening of the Mond Laboratory on 3 February 1933 survives among the Oliphant papers. After formalities at the nearby Arts School, which featured an address by the Chancellor (Prime Minister Baldwin) and “short addresses” by Rutherford, the President of the Royal Society and Sir Robert Mond, tea was served in the Cavendish and visitors invited to observe “demonstrations”. In the Mond they could see liquid air and

159 Wood (1946), p. 52. See also Crowther (1974), p. 226. 160 That Rutherford was able to secure this funding has been taken by some observers as an example of the (perhaps undue) influence of the “Royal Society-Cambridge” axis, and of Rutherford’s manipulation of philanthropic sources channeled through the Society. This was a matter of some political sensitivity in British science in the mid 1930s, and a contributing factor to the Royal Society “revolt” of 1935. See Hughes (2010). For some further comment on the role of the Royal Society in funding the Mond Laboratory see Hughes (2010a). 161 See footnote 59. 94

liquid nitrogen prepared, and an exhibit of “coils and apparatus used in research with intense magnetic fields”.

Also on show in the Cavendish was a cross-section of current research, or at least that suitable for demonstration; FW Aston’s mass spectrograph; JA Ratcliffe’s sounding of the ionosphere; PMS Blackett’s photographs of cosmic ray tracks; Chadwick and Feather’s studies of neutrons; the “ring magnet” used to measure the velocity of alpha particles; as well as the Cockcroft/Walton and Oliphant/Rutherford accelerators.

Plans for the High-Tension laboratory

The building of an advanced high-tension laboratory at the Cavendish Laboratory, with target energy above a million volts, was one of Oliphant’s most significant achievements but its origins are not clear. Oliphant himself makes no reference to the development in his own memoir of his time at the Cavendish. The building of the facility was at least two years in the planning. The earliest surviving documentary reference is in a letter from Oliphant to Rutherford162 in January 1934, following a visit by Oliphant and Cockcroft to the plant run by the Dutch electronics firm Philips in Eindhoven. The letter is here quoted in full.

Dear Professor

We have had a very successful trip to Holland, though the sea was rough enough to make me sick on the way back. The Philips labs were an eye-opener to me, and I think when the Cavendish is rebuilt their design must be borne in mind. Of course they are very expensive, and in the opinion of Cockcroft are far superior to

162 Oliphant to Rutherford, 7 January 1934. EBA. 95

any in America. The people who work there are obliging and impress me very much with their knowledge and their efficiency.

The million volt-set was demonstrated to us and worked very well indeed. I feel now I could go straight ahead and build a two million volt set at once. They have overcome the whole of their corona troubles and their design is very compact and emits no brush or ozone. I am certain that a set for 2 million volts can be built into a room 60 x 40 x 40 feet, and allow for two experimental tubes, one of which could be used at two million volts and both of which could be used simultaneously at one million volts. The design of the building is quite clear.

If we are to have higher voltages I think we should set about preparing the parts for the rectifiers and tubes straight away, and make some provision for space as soon as possible, for with all we have to do in our experiments at present, I think there are dozens imminent for which the number of particles observable at our present voltages is much too low. In any case, it will take the best part of a year to set things up, and in the meantime we can continue on with our present apparatus. If we wait until we are forced to use higher voltages to prepare for them, it will still take a long time to get ready. Perhaps you will feel that I am over- enthusiastic and inspired too much by my trip, but I have given a lot of careful thought to the question and think I have learnt a great deal about it of late. Of course the voltage is necessary for gamma-ray work. By splitting it into two one-million volt units which can be used separately or together, I think it can be made very flexible and that breakdowns will be minimised. The question of rectifier design is quite clear in my mind.

96

This letter is full of fascinating detail. It appears that the visit was made so that Oliphant and Cockcroft could become familiar with the technological advances Philips had made in controlling high voltages, presumably for use in generating X-rays, an area in which Philips was active commercially163. These techniques would form the basis of a machine to be built in-house at the Cavendish. There was no suggestion at this time that apparatus might be bought ready-made, as was in fact later done.

A target voltage of 2 million volts had already been set, twice what the Philips equipment could deliver, three times higher than that used by Cockcroft and Walton, and five times greater than Oliphant and Rutherford had employed in the basement. It was also perhaps typical of Oliphant that having seen a one million volt unit demonstrated he at once began to think of a two million volt machine. Oliphant seemed always ready to “push the envelope”. It was a challenging goal, but Oliphant was ready to take it on. He saw the need to get started as soon as possible, aware of how long the design and construction would take. He already was planning the new building in which it would be housed.

Most importantly, he knew that higher voltages were the way forward, generating more energetic beams to bring about transformations in heavier atoms than could be explored with existing equipment. In Berkeley, Ernest Lawrence was similarly motivated in the development of the early cyclotron.

Though the surviving documentary record makes no further reference to the project until 18 months later164, the development of the facility has

163 For a perspective on the Philips Laboratory during this period, see Boersma (2003). 164 Documentation in the Cavendish archives shows that a High Tension Laboratory Building Committee was formed around April 1935. Members were Cockcroft, Oliphant and several university officials, including those with responsibility for finances and buildings. Rutherford 97

been comprehensively chronicled by Oliphant’s Cavendish student WE (William) Burcham, later a colleague at Birmingham165. Burcham entered the Cavendish in October 1934, keen to work with Cockcroft but assigned by Rutherford to “learn the trade” under Chadwick until the departure of the latter to Liverpool in mid 1935. Knowing of Burcham’s interest in high-tension technology, Chadwick assigned Burcham to work with another research student BC (Ben) Brown to re-assemble and make improvements on an existing HT set, no longer used166.

At the same time, Oliphant was working with Rutherford on the “basement accelerator”, a relatively low-voltage, high-current machine. However, he had thoughts on how to replace the very tall rectifier columns used in the original Cockcroft/Walton equipment with rectifiers using mercury vapour. These would have been smaller and more compact, able to work at higher voltages before sparking over. Burcham and Brown were set to work under Oliphant to develop equipment along these lines, making use, inter alia, of “glass tubing and plasticine”. The rectifiers were intended for use in Oliphant’s HT machine but not needed.

Burcham recalled that after working on this task “for about a year”, and achieving some success, the team was overtaken by industry developments. “The Dutch firm Philips had begun to market some well- made and efficient mercury vapour rectifiers for very high voltage work. attended a number of these meetings, as did architect Charles Holden once he was appointed. The first meeting on record was on 15 May 1935. See CAV 3/2, University of Cambridge Library [hereafter cited as “UCL”]. At the meeting, Cockcroft and Oliphant were asked to provide detailed specifications for the facility. 165 Burcham (1947), Burcham (1999). In the latter, Burcham drew comprehensively on surviving documentation and correspondence In the 1999 paper, he omments.”[The paper] is a shortened version of a longer account which has been placed in the Churchill Archives Centre together with a number of relevant documents held since 1939 by the author. In writing, use has been made of the author's personal notebooks and diary kept during the period covered.” Additional material comes from an interview with Burcham, conducted by the author in Birmingham in 1980 [transcript in EBA]. 166 It is likely this equipment is that first put together by Allibone (see Chapter 3). 98

These could have been the basis of a new equipment but were overtaken by events”.167 Burcham seemed to imply that Oliphant and his team had begun to plan to use the new rectifiers in the apparatus they were building, but were offered another option.

By mid 1935, the project was moving forward, if Cockcroft’s report to Lawrence is taken at face value168. “With regard to our own work, we are putting up a laboratory about 45 feet high and 45 feet broad, in which we hope to have two experimental tubes. We think it will be quite easy to get two million volts with one end grounded and possibly to get a little higher than this by the use of oil immersion for the rectifier unit.” Writing to Lawrence the same day, Rutherford was more cautious. “We have in view the construction of a High Tension Laboratory to give us about 2 million volts DC by multiplication of our old system, and Oliphant and Cockcroft are busy designing the apparatus. We hope that a suitable laboratory will be built some time next year”169.

Rutherford was right to be cautious. The challenge of cost had still to be met. On 25 August 1935, we find Oliphant writing in some agitation to Rutherford, then on holiday, mostly in response to a quote he had received for the building of the new facility. Having initially referred to some urgent official correspondence which he had enclosed170, Oliphant went on

A representative of Holden171 was down yesterday about the new High-Voltage Laboratory. The approximate estimate of £15 000

167 Burcham (1947). 168 Cockcroft to Lawrence, 18 July 1935. LPBL. This statement was in fact one of intention rather than actuality. Construction did not begin until early 1936. 169 Rutherford to Lawrence, 18 July 1935. LPBL. 170 Oliphant was now officially Rutherford’s deputy, having been appointed Assistant Director/Research in Physics on 15 June 1935, following the departure of Chadwick to take up the Chair of Physics at the University of Liverpool. Relevant papers are in EBA. 171 Charles Holden was the architect for the HT Laboratory as well as for the Austin Wing. 99

staggered me, as I imagine it will you. Cockcroft’s original estimate was £6000, and allowing for his well-known optimism one might have guessed £9000. I feel the plan has been made too elaborate for a laboratory that must of necessity be prepared for changes, to follow fresh requirements for future work. Too much attention has been paid to the requirements of a pleasant exterior, and to the screening of X-rays all over, rather than at the source. I would like to see a simple steel frame covered with a light roof and made weather-tight by filling in with the minimum of brick, or even iron or wood. It would then be a real workshop, and recognised as a place liable to change and alteration long before the more stable parts of the new laboratory. Something like an airship hangar is required in my opinion. I have discussed some of these points with Cockcroft and the architect, and they will give the question of cost their attention at once.172

I can see the new lab receding into the distance if we are not careful. Of course if we had the money, or if you thought we could raise it, all would be well. I do not know much about such things as yet, but I feel that the building committee has not quite done its duty in not having met since the architect was appointed. Many changes have been by individuals in the general plan approved by the committee, and Cockcroft has been too busy with College buildings173 to give it the full attention it required. Some of the

172 It now appears that Oliphant was out of step here. Cavendish records show that t the HT Laboratory Building Committee (of which he was a member) had decided six months before that the project would be seen as the first stage of a redevelopment of the Cavendish as a whole, and that therefore all the buildings had to be uniform in style. See Minutes of Committee 30 April 1935 (CAV 3/2/3, UCL). Records show that the £15 000 was based initially on the findings of quantity surveyors. All six quotations received in October were in the range £14 000 to £1 6000. See CAV 3/2/1. UCL. 173 Cockcroft was at this time Junior Bursar at St Johns College, with responsibility for upkeep of the buildings. This may be seen as an unfair judgement. Records (for example CAV 4/3/iii. UCL) show that Cockcroft, whatever his other responsibilities, became very deeply immersed in all details of design and construction of both the HT facility and the Austin Wing to follow. 100

changes I had never heard about, and I doubt very much whether you are familiar with them all.

I am sorry to write so fully about this matter of the new lab, but I think we need it urgently, and not in some distant future when all the cream has been scooped off by folks whose results we dare not trust too deeply174.

By early 1936, things were beginning to move, as Rutherford reported to Lawrence, though there is no reference as to how the facility was to be paid for175.

At present we are just beginning the new building for our high tension D.C. plant, and we hope with luck to reach two million volts positive and negative, and possibly higher, but no doubt will find plenty of trouble before it is in working operation. We shall, of course build up the component parts of the apparatus ourselves so as to keep down the expense176.

Rutherford’s ambivalence and scepticism about large and complex apparatus was once again on show in this letter, as was the need to contain costs. Yet within a few months, the situation would change completely, with more than the HT laboratory impacted.

The Austin bequest

174 This can be read as a reference to Lawrence’s team at Berkeley, following the controversy over the interpretation of the experiments with deuterium, discussed above. 175 It should be noted that moves were underway to secure additional funds for the Cavendish. Rutherford had prepared a plan for the redevelopment of the laboratory in 1935, and that had been used by the University as the basis for a public appeal, with the appeal document written by Eddington, the Cambridge-based astronomer. It is likely that it was that document that stimulated the response. For example see Wilson (1983), pp. 587-8. 176 Rutherford to Lawrence, 22 February 1936. LPBL. 101

In 1936, the substantial funds necessary to bring the proposed high- tension facility into existence unexpectedly became available. The motor car magnate Lord Austin wrote to Prime Minister Stanley Baldwin on 29 April 1936 in the latter’s capacity as Chancellor of the University.

I have for several years been watching the valuable work done by Lord Rutherford and his colleagues at Cambridge in the realm of scientific research. Knowing that as Chancellor you are keenly interested in obtaining sufficient funds with which to build and endow a much-needed addition to the present resources, I shall be very pleased indeed to present securities to a value of approximately £250,000.177

According to Wilson in his biography of Rutherford178, Cockcroft many years later described the genesis of the Austin largesse as follows:

We went to the Prime Minister [Stanley Baldwin] because he was Chancellor of the University, to get him to find a benefactor or to sign an appeal for money. And Mr Baldwin was so lazy he decided he would do it all in one go by going to [Herbert] Austin who in turn got a peerage for the benefaction. So he produced £250,000 and in return he was made a Lord179.

Wilson noted that Cockcroft does not say what evidence he had for this statement, but the official history of the Cavendish180 appears to support him by implication. “In 1936, the motor manufacturer Sir Herbert Austin had given a quarter of a million pounds for the laboratory’s development. Shortly after, he was raised to the House of Lords.”

177 Quoted in Wood (1946), p. 53. 178 Wilson (1983), p. 588. 179 The source of this quotation is an interview with Cockcroft on 2 May 1963. The Neils Bohr Library, American Institute of Physics, Oral History Archive, OH 89. 180 Crowther (1974). 102

In his short history of the Cavendish, Wood181 noted that of the total Austin bequest of a quarter of a million pounds, £37 000 was used to defray the cost of the High Tension Facility, though it is not clear that this constituted the entire cost. Even if that were so, the expenditure on the new laboratory would have been more than twice as great as was required to build and equip the Mond Laboratory, taking the project very much into the realm of Big Science for its day182. The remainder of the funds was to be used to construct a new research wing of four storeys, connected to the old building by a bridge, and to undertake an extensive renewal of the old building. According to Wood, work did not begin on site on these elements of the new facility until October 1938, and a foundation stone was not laid by Lord Austin until May 1939.

Commenting on the impact of the Austin bequest, Wood said “The offer was gratefully accepted by the University, and Rutherford at once threw himself with his usual enthusiasm into the planning of the new building. Unfortunately he was not to see its completion”183. The impression given by Eve was similar. “This generous donation came as a complete surprise to Rutherford, and gave him great satisfaction... After the initial pleasurable satisfaction came thought and planning; some money was to be spent on building and some reserved for endowment”184. Eve also cites a letter from Rutherford to the German physicist Otto Hahn late in 1936, which does not suggest any adverse feeling about the situation. “You will have heard of the substantial gift from Lord Austin to the Cavendish Laboratory, and we are now in the midst of considering plans

181 Wood (1946), p. 57. 182 While the basis of the Wood figure is not clear, it now appears overstated. The construction of the building cost £14 000, the purchase of the Philips high-tension source ultimately cost £4 000. 183 Wood (1946), p. 53. 184 Eve (1939), p. 408. 103

for a new research block and for possible reconstruction of the old Cavendish”.185

Even before this time, Rutherford appeared to be looking forward to, or at least had been reconciled to, the new age in physics. Eve cites a lecture given to the NW Centre of the Institution of Electrical Engineers on 22 January 1935. Rutherford said, as Eve reported it

….whereas in the old days apparatus was simple and fairly inexpensive, today we must have high voltage devices of a far more costly nature. He hoped to be handling two million volts for this purpose in the near future. This was necessary for bombardment – for firing very swift particles at certain substances under investigation. He most gratefully acknowledged the help and advice of Metropolitan-Vickers, for they had put their research resources at his disposal and been of enormous assistance.

Wilson in his Rutherford biography had quite a different view of the great man’s response. “Rutherford himself took little pleasure in the gift or in the spending of it. He knew he was building for his successor, and would say so loudly, jingling the loose change in his pocket as he did so”,186

The attitude of Rutherford to large expenditures on equipment and facilities is worthy of further investigation, since the signals that have reached the present day are confused. The popular image of Rutherford is one of an experimenter who regarded the thinking that went into the design of an experiment and into the analysis of the results more

185 Rutherford to Hahn, 5 November 1936. Quoted in Eve (1939), p. 409. 186 Wilson (1983), p. 588. 104

important than the experimental equipment itself, which could be cobbled together from what ever was at hand; the much-quoted “string and sealing wax” tradition. He almost relished having to make do with very little. The following is widely quoted, and perhaps apocryphal, but reflects this point of view: “We don’t have much money so we have to think”187.

Yet the longer record suggests some ambivalence. Crowther pointed out in his official centenary history of the Cavendish188 that at the time of his appointment as Cavendish Professor in 1919 Rutherford had thought the laboratory was in need of a major upgrade in facilities, driven at least in part by the big rise in enrolments following the end of the Great War. In a memorandum to the University authorities, he maintained that significant expenditure was warranted, particularly to promote teaching, through the appointment of “three additional lecturers of high standing”, and to advance research in areas like applied physics, optics and properties of matter by the “provision of new, well-equipped laboratories”.

Crowther wrote

Rutherford’s attention to, and conception of, financial needs, were no less remarkable than his emphasis on industrial, national and political considerations in the appeal for the extension of the Cavendish. He estimated that the new buildings required would cost not more than £75 000, and that a further £12 5000 would be necessary as an additional endowment.

187 Quoted by Jones (1962), p. 102. 188 Crowther (1974), p. 185. 105

In 1919 £200 000 was a very large sum for a university science laboratory. The University was not entirely convinced by the necessity of the requirements enumerated by Rutherford, and in any case, the money was not available.

These plans and proposals were in striking contrast with Rutherford’s personal methods of research. The apparatus with which he had disintegrated the atoms was fantastically simple, and could be operated by one man alone.

This raises the question of how sincere Rutherford was in his 1919 proposal. It seemed he was in no hurry to find the funds he thought were needed. Not until 1935 did he undertake any concerted effort to find sponsors. Indeed, as Chadwick recalled in a letter to Oliphant much later, Rutherford did not avail himself of resources even when they were offered. The relevant section of Chadwick’s letter, as transcribed by Oliphant, is worth quoting in full189.

One evening after dinner I was alone in the Combination Room with the Master, Sir Hugh Anderson190, when he suddenly said to me “why wouldn’t Rutherford accept the grant I had arranged for him?”

I could only reply that I knew nothing whatever about the matter; Rutherford had not mentioned it to me. (He never did; nor to anyone, as far as I know.)

The Master then told me the story. As I knew, Anderson had a great admiration for Rutherford. He thought that the grant made

189 Chadwick to Oliphant, 2 April 1967. EBA. 190 Anderson was Master of Caius College where Chadwick was a Fellow (and later Master). 106

by the university to the Cavendish for research expenses was not sufficient to enable Rutherford to pursue his work properly, but it was as much as the university could afford and there was no possibility of increasing it. But Anderson’s regard for Rutherford’s work was such that he felt something must be done to help him. Accordingly, he approached a number of his wealthy friends and persuaded them to offer Rutherford a private grant of £2000 a year for some years to use at his discretion. He told Rutherford what he had arranged. The position was that Rutherford had only to say he needed money for research at the Cavendish and it would be forthcoming.

But Rutherford did nothing, to Anderson’s astonishment and indeed exasperation. I could offer no explanation for Rutherford’s refusal to say he needed more money for his research. I knew so well how hampered and restricted Rutherford was for lack of equipment and technical assistance, and his attitude was quite incomprehensible to me.

This conversation took place at sometime between 1922 and 1925. It was only some years later that the explanation dawned on me. It came about this way. One day I had taken Rutherford up to the radium room so that he could assure himself that all was in order. We had at that time about 400mg radium in solution for the preparation of radon and active deposits sources. I remember well how, as we were coming down the stairs, I said that we did not have enough radium, so that I had to allocate sources very carefully to meet demands. And I said it was a pity that somebody or other had not made a gift to him of a gram of radium, as the women of the United States had made to Madame Curie. His reply astonished me. It was “Well, my boy I am very

107

glad that nobody did. Just think; at the end of every year I should have had to say what I had done with it. How on earth could I justify the use of a whole gram of radium? This was said quite seriously.

I had, of course, long realized that Rutherford was modest about his achievements, notwithstanding his eager enjoyment of his reputation, his almost boyish delight in any laudatory references to his work, his susceptibility to flattery -- aspects of his character with which you must be familiar. But I had not realized how deeply ingrained his modesty was until I pondered over this remark. And it threw a light on some arguments concerning the spending of money on research, especially on his own work. It was then that I thought I had an explanation for his refusal to ask for the research subsidy which Sir Hugh Anderson had arranged for him; he did not feel he could justify spending so much money - - which was a great deal in those days -- on himself and his research students. And this in spite of his unique position in the scientific world, his extraordinary achievements in the past, and the urgent need to press on with the establishment of nuclear physics as a new branch of inquiry.

He was, in truth, deeply modest.

In his response to Chadwick191, Oliphant recounted a story of his own, but with quite a different message. Rutherford’s attitude seemed to be the result of more than modesty.

Your story of Rutherford and the offer of a grant which he turned down is most interesting. In some ways it does not surprise me.

191 Oliphant to Chadwick, 20 April 1967. EBA. 108

Just after the Austin gift we were spending the weekend with the Rutherfords at their holiday cottage. He was gloomy and when asked about the cause of his preoccupation let out at me about all the goddam worry and trouble that the bequest meant for him. He said he was damned if he was going to spend his time planning a new Cavendish, that it would upset the whole lab and that Cockcroft would not do a stroke of work for a year or two. He went on to say that a new laboratory was unnecessary, anyway, if the number of people was kept down to about that present, and that thinking was more necessary than grand surroundings or elaborate and expensive equipment.

I had a letter from Lawrence as a result of which I was brave enough to argue back, telling him of the results being obtained at Berkeley and that we must not be left behind. He then really blew up, becoming red in the face, and shaking his pipe at me. He declared that luxury was not good for anyone, least of all a physicist, and the amount of physics done per pound of expenditure was in inverse proportion to the total expenditure.

When he was exceedingly busy as P.R.S. 192 and we were working together in the basement room, he really enjoyed himself, coming down almost everyday to look at records and discuss results. After the installation of the Philips equipment, he suddenly came to see us. He complained that we spend more time keeping all the gadgetry in operation than in making observations, which was perfectly true, and he thoroughly disliked apparatus that he did not, and would not, understand.

192 President of the Royal Society. Rutherford held this post between 1925 and 1930. Oliphant’s recollection is therefore faulty on this detail, since the work in the basement room did not commence until 1932. 109

Rutherford was modest, as you say, but he also hated the idea of having to report to anyone on what he had done, was doing, or intended to do, though he would hold forth readily enough if he were speaking voluntarily. Perhaps the potential patrons rounded up by Sir Hugh Anderson required too much information. I remember that when Lawrence was trying to obtain funds for his 184 inch cyclotron, one of the donor institutions wrote to Rutherford, and to me also, requesting a statement that the proposal was the most important in physics at the time, and that it would undoubtedly lead to results of great importance. Rutherford was indignant when he spoke to me about it. And he did write but I never saw his reply. Perhaps it was in longhand, as was so much that he wrote, and hence there is no copy.

Wilson put some context around the Austin gift and why it might have affected Rutherford in the way Oliphant recalled.

The colossal munificence of the gift must be recognised, however. In 1930, the university contributed just over £15 000 toward running the Cavendish, and much of this money came from students’ fees. In addition, the DSIR193 contributed just over £2600 in grants, mostly towards Kapitza’s High Magnetic Field Laboratory. At almost the same time Cockcroft and Walton were building their first accelerator which cost, in round figures, £1000. In 1935, the year before Austin’s benefaction, the university’s bill for the Cavendish was £16 200 and the sum spent by Rutherford on new apparatus that year was slightly over £1000. Compared with these sums, £250 000 seems almost out of proportion194.

193 Department of Scientific and Industrial Research, which apart from running a number of laboratories provided public money to support some university research. 194 Wilson (1983), p. 588. 110

Here and there in Rutherford’s correspondence we can find intimations that he regretted the impact of major organisational tasks that might cut heavily into his time and energy. For example writing to Lawrence before the Austin gift was announced195, he commented

I have been kept very busy in a multitude of ways and have not had much time for my own research during the last few months. It was a difficult business making the arrangements with the Russian government. I had to keep the University, the Royal Society, the Department of Scientific and Industrial Research, the U.S.S.R. Government, not to mention my own Laboratory Committee, all working in unison. Fortunately the Soviet people were very anxious to arrange a reasonable concordat. In the course of a year or so, Kapitza should have the greater part of the apparatus available in his new laboratory in Moscow.

In mid 1937, a few months before his death, Rutherford prepared the following word picture of the completed apparatus in the high-tension laboratory, contrasting it with the simplicity of the equipment with which he had first transmuted matter in 1919.

At Cambridge, a great hall contains massive and elaborate machinery, rising tier upon tier, to give a steady potential of about two million volts. Near by is the tall accelerating column with a power station on the top, protected by great corona shields – reminding one of a photograph in the film of Wells’ The Shape of Things to Come196.

195 Rutherford to Lawrence, 22 February 1936. LPBL. 196 Quoted in Eve (1939), p 423. The address containing this quote was intended for the Jubilee meeting of the Indian Science Congress of which Rutherford was President. He died before the meeting was held. 111

Whatever Rutherford’s attitude to big spending on physics had been earlier, it was clear that a watershed had been crossed in experimental nuclear physics, and Rutherford knew it. On this matter at least, Oliphant had always been ahead of his master. Awareness of the importance of the new technology, and the need to mobilise much larger resources to procure such equipment, were vital elements of the ‘capital’ Oliphant was building.

Cockcroft and his cyclotron

In parallel with the building of the high-tension laboratory, Cockcroft commenced the construction of a cyclotron. A year earlier he had no public plans to do so and indeed had argued that the high-tension route was the way to go. Writing to Lawrence in July 1935, he commented “Our main reason for going on this method is that there seem to be plenty of laboratories who would be capable of using the cyclotron method and we are anxious to be able to use our set both for X-rays and positive ion work with counters and Wilson chambers”197.

It can be argued that Cockcroft was dissembling; financial constraints were the main impediment, rather than fear that there would be too many cyclotrons. He had wanted to build one since he had visited Lawrence in 1933, and would move very quickly once the money was available. This is confirmed in a letter to Lawrence a few months later198.

There is now a probability of our getting funds for installing a large magnet in the Cavendish and one of the purposes for which we should wish to use it would be for producing high velocity

197 Cockcroft to Lawrence, 18 July 1935. LPBL. 198 Cockcroft to Lawrence, 16 October 1935. LPBL. 112

ions199. I should therefore be grateful for any recommendations you could make about such questions as gap-spacing, gap- diameters and the best value of magnetic induction to work at.

In The Two Ernests, Oliphant clarified the source of the funds for the now-proposed cyclotron. It came, in the main, not from the Austin bequest but from the sale to Russia of equipment from the Mond Laboratory.

The reference [in a letter from Rutherford to Lawrence dated 22 February 1936] to the Royal Society Mond Laboratory concerns equipment which has been provided for the work of P. [Peter] Kapitza, the Russian engineer-physicist who had joined the Cavendish in 1921. He was in the habit of visiting Russia in the summer to visit his old mother. In 1934, the Soviet government refused to allow him to return to Cambridge, but offered to buy his equipment from the University so he might continue his researches in Russia. With the able help of Cockcroft and others, Rutherford proved himself a better man of business than expected, and negotiated a good price for the equipment. Meanwhile Rutherford’s resistance to the idea of as complex a piece of apparatus as a cyclotron in the Cavendish Laboratory, had been worn down, and he was willing to devote part of the sum received from Russia to the acquisition of a large magnet which could be used inter alia for a cyclotron200.

Rutherford added to this, writing to Lawrence201.

199 That is, in a cyclotron. 200 Oliphant (1966), (2) p. 43. 201 Rutherford to Lawrence, 22 February 1936. LPBL. 113

At present we are very busy transferring the apparatus from the Royal Society Mond Laboratory, and getting duplicates, and keeping the cryogenic work going as usual. We do not intend to get a duplicate for the big generator for producing very strong magnetic fields, but have in view instead the installation of a large magnet for general purposes, and also probably for use as a cyclotron. We have not had time as yet to go into the matter, but I think Cockcroft will be writing to you soon to see if you can give him any information of the best design of magnet to be used for the latter purpose.

For some time, the notion that the new magnet would be for multiple purposes persisted. Three days later Cockcroft wrote to Lawrence.

We are now definitely going ahead with the construction of a large magnet, I should be very grateful to receive your drawings showing details of your latest construction. My present ideas are to build a magnet with a pole face of 100 cm diameter, a gap width of 3.5 inch and a maximum field 17 500 gauss.

Since we may wish to use the magnet for other problems in addition to nuclear work, I would prefer to build this with the pole faces vertical. I do not see that this will introduce any special difficulty in the support of the electrodes etc202.

By the end of 1936, and in the wake of the Austin gift, the machine had become purely a cyclotron and the degree of in-house construction had been significantly reduced, as Cockcroft indicated to Lawrence203. “I ought to let you know what progress we have made toward getting our

202 Cockcroft to Lawrence, 25 February 1936. LPBL. 203 Cockcroft to Lawrence, 16 November 1936. LPBL. 114

cyclotron. We have in the first place got our magnet under construction by Brown-Firth of Sheffield. The steel weighs 40 tonnes and the copper about 8 tonnes and the final pole face diameter is 90 cm. We are using air cooling for the coils with a water jacket around the poles… The oscillator will probably be presented to us by Metro-Vickers.”

June 1937 saw the foundations for the cyclotron being prepared with the magnet expected to arrive from the manufacturers the following month.204 It took another year for the machine to be assembled and made ready for use, including chasing down the leaks. By late July 1938, Cockcroft could report that he was preparing for “the excitement of looking for a beam”205.

Progress on the High-Tension Laboratory

The impact of the Austin bequest on the high-tension laboratory was just as profound. Now that money was no longer a problem, the fruits of two years of planning could now be harvested. As Burcham recalled

On the third (I think) of May 1935206, the bequest was announced from Sir Herbert Austin which among other things made possible the buying of a one million volt equipment from the Philips company. A laboratory was built for this, though actually the laboratory had been designed for equipment that Oliphant was going to build, using enormous Bakelite columns immersed in oil and he planned to push to 2 million volts. But again we were overtaken by events. First of all the benefaction of money was available to buy equipment off the shelf. Secondly; the shelf

204 Rutherford to Lawrence, 10 June 1937. LPBL. 205 Cockcroft to Lawrence, 26 July 1938. LPBL. 206 Clearly a slip in Burcham’s memory (or of his pen) at this point. The year was 1936. 115

equipment was available from Philips. They had developed sealed-off mercury vapour rectifiers and were very anxious for reasons of prestige to sell the equipment to the Cavendish and gave very favourable terms207.

Now Oliphant could tell Rutherford “I had a ring from Philips today and they offer to install their H.T. set for 1250 KV and to maintain it free for two years at the price quoted i.e. £5700208. This is subject to our cooperation in securing the entry of the apparatus free of duty. It appears therefore that they are prepared to back their valves very fully”209. The pace of construction could now pick up. In November Rutherford reported to Hahn. “Our new High-Tension Laboratory is nearly completed, and we hope to install our new apparatus to obtain two million volts D.C. early in January. Oliphant is busy preparing this new installation, which is experimental, but which I hope will work alright. As 2 million volts will probably give a spark of more than 30 ft. long, you will appreciate that plenty of headroom is required”210 .

Two weeks later Cockcroft could tell Lawrence “The new High-Voltage Laboratory is nearing completion and the various bits and pieces are ready to go inside after Christmas. We are putting in a ready-made Phillips 1.2 MV DC. Generator in addition to Oliphant’s oil set which it is hoped to get 2 MV with”.211 Rutherford backed this up a month later212; “The New High Tension Laboratory is nearly completed, and we hope to get a D.C. potential of 2 million volts going. We are also making

207 Burcham (1947). 208 Oliphant to Rutherford, 21 July 1936. EBA. Cavendish records show that the final price was £4000. The reduction was due partly to bargaining, but more to the keenness of Philips to make the sale to so prestigious a labotatory. 209 The valves Oliphant refers to are the mercury-filled rectifiers. 210 Rutherford to Hahn, 5 November 1936. Quoted in Eve (1939), p. 409. 211 Cockcroft to Lawrence, 16 November 1938. LPBL. This letter was on the letterhead of the Royal Society Mond Laboratory, which Cockcroft was running following the return of Kapitza to the USSR. 212 Rutherford to Lawrence, 21 December 1936. LPBL. 116

arrangements to run one of your cyclotrons in due course.” By early 1937, the Philips set was operational. In April Rutherford wrote to Oliphant from Chantry Cottage; “Glad to hear you have reached one million volts on the tube [?]213 and hope that you will soon succeed in getting a good beam for work. The finding of a pinhole must have been a task of the first magnitude – the proverbial needle in a haystack”214.

This documentary record supports Burcham’s recollections.

The one million volt set from Philips was run in by the end of 1937 and ran very successfully through the war. During the war it was used by Halban and Kowarski to measure atomic cross- sections of importance in the release of nuclear energy. It was under the control of Feather and Bretscher. The machine continued use after the war at least until 1951 when Burcham moved to Birmingham and later went to South Africa and was finally decommissioned in 1971.

The second equipment was just beginning to work at the start of the war. This would reach 1.4 million volts before spark-over. It possibly did not operate during the war but was used by Devons doing angular correlation studies. Its use was discontinued after 1951, but the gear probably not (?) sold. 215

In The Two Ernests216, Oliphant helped clarify the chronology of the high-tension development. Oliphant stated that the original 1.25 MV equipment “has now been further developed” so that it delivers 2 million volts, but he does not say by whom. The original intent, going back to

213 It should be remembered that Rutherford usually wrote his personal letters by hand with a blunt pencil and his handwriting is not easy to decipher at times. 214 Rutherford to Oliphant, 2 April 1937. EBA. 215 Burcham (1947). 216 Oliphant (1977), p. ?? 117

early 1934, was for Oliphant himself to build the 2 million set, spring- boarded from some of the Philips insights, but that does not now appear to be the case. Oliphant here referred to two deliveries of equipment from Philips; it seems reasonable to think that the second delivery was of a 2 million volt set, and that as a corollary the Oliphant equipment was not proceeded with.

This seems to fit with Burcham’s recollections.

The HT lab held two very large tanks filled with oil, probably in preparation for Oliphant’s new accelerator. There was also a column of porcelain cylinders for the accelerator tube and Bakelite cylinders for the HT. But the HT apparatus was never built, having bought the Philips sets. The ion sources still needed to be built but that was to a standard design using generators driven by insulated belts217.

It was late in 1938 before the two million volt apparatus was up and running. Writing to Lawrence in July that year, Cockcroft reported; “We are just completing the installation of our two million volt DC generator, and hope to have it in operation by the end of August. Meanwhile the 1.2 million volt set is keeping about a dozen different investigations going very satisfactorily, and a great many new results are coming out”. 218 Cockcroft here makes a reference to the “productivity” of the 1.2 million volt set, that is, its capacity to support more than one research team. As we shall see in Chapter Nine, such productivity can be seen as a key characteristic of Big Science.

217 Burcham (1947). 218 Cockcroft to Lawrence, 26 July 1938. LPBL. 118

We have gone at some length into the saga of the HT Laboratory for a number of reasons, all of which can illuminate aspects of Oliphant’s growing “capital”, of the new experiences and understandings that he would bring to bear on the Birmingham project in due course. It was, for example, the first project in his experience that required resources beyond those the Cavendish could provide internally. Without the Austin bequest it may never have gone ahead at all, or it would have been on a notably different scale or time frame.

For the first time, Oliphant interacted significantly with industry, to the point that much of the plant was bought in finished form from an outside supplier. Given that Oliphant had initially planned to build the apparatus in-house, in the well established tradition, that outcome marked a major transition. It reflected in his own work the way the technical environment for experimental nuclear physics was shifting and would continue to shift. We have already noted the divide that had appeared between Oliphant’s new approach and that of Rutherford. The technically-adept Australian was very comfortable in the new and complex environment, welcoming the opportunities it brought; the New Zealander was never really at home, accepting it with gritted teeth (or so it seemed), always hankering after the simpler ways through which he had built his reputation.

It fell to Oliphant to give the first complete account of the new high- voltage laboratory. This he presented in October 1937 at a congress in Bologna marking the 200th anniversary of the birth of Luigi Galvani219. His paper was later published in Nuovo Cimento220. Oliphant began

219 It was during this meeting that news reached Oliphant of the death of Rutherford. 220 Oliphant (1937). 119

During the past year or so, a number of changes and developments have been made in the technical methods employed at the Cavendish Laboratory for obtaining high energy beams of ions for work on nuclear transformation problems. At the same time, some interesting results have been obtained. As a number of the members of this Congress are interested in the practical side of nuclear physics, it seemed to me that it would be interesting if I spoke about these newer developments.

The new high-tension Laboratory at Cambridge has been built at a time when the technique of producing and controlling high voltages has shown remarkable development, so that the whole represents a compromise between what appeared to be best when the building was begun in 1936 and what appears to be the most advanced procedure at the present time. The Laboratory was designed to a maximum tension of 2 000 000 volts positive with respect to earth, and we still feel this represents the maximum with which it is practicable to work at the present day When the Laboratory was first designed, it was intended to build the new high-tension equipment, and indeed an apparatus for producing 2 000 000 volts is under construction. However in the middle of 1936, Philips of Eindhoven, Holland, produced an apparatus of an engineering character which delivers a potential of 1.25 million volts d.c with a minimum of trouble, and it was decided that the commercially-made apparatus had many advantages over our own221. . As Oliphant reported, designing a building for high-tension experiments had posed a number of challenges. Wire gauze and netting, connected both to earth and to the copper roof, covered the walls and ceilings of

221 Ibid. 120

both the large apparatus hall and the smaller experimental rooms, forming a complete Faraday cage to exclude stray electrical fields. The walls of the control rooms were formed of half-metre thick concrete to protect the operators from X-rays generated by the accelerating tubes. Care was taken to minimise the entry of dust, and surfaces were made as smooth and flat as possible to prevent dust from building up.

“The Contribution of Engineering to Physics”

As the building of the high-tension laboratory and then the Birmingham cyclotron demonstrated, Oliphant through the 1930s had become increasingly engaged in the application of the skills and insights of engineering, notably electrical engineering, to research in physics. A summation of his developing views on this interaction comes from the talk he gave on the topic at the meeting of the British Association at Cambridge in the autumn of 1938222. Oliphant’s original typescript, with handwritten amendments, survives among his papers223. Reflecting as it does the impact on Oliphant’s outlook of the previous five years of work, it is worth examining in some detail.

In the margin of the first page, we find the following added comment, “It is true to say that the various branches of engineering are the children of physics. It is equally true to say these various progeny have attained majority and that a very substantial proportion of the family tree of common knowledge has been contributed by them.224” The first part of this positive opening was reinforced by a quote from Rutherford: “I think I can speak from some knowledge of the effect of science on the growth

222 It had been at such a gathering in 1933 that Rutherford had declared that thoughts of getting useful energy from atoms were “moonshine”. His comments were reported inter alia on the front page of the New York Herald Tribune on 12 September 1933. 223 The Contribution of Engineering to Physics. A paper delivered at the annual meeting of the British Association for the Advancement of Science, in Cambridge in Autumn 1938. Typescript in EBA. 224 Ibid, p.1 121

of the great electrical industry, for it was born in the scientific laboratory, was nurtured by science and now wholly depends on science for its growth and expansion to meet the needs of man.”

What of the reverse half of the interaction? Noting but not really lamenting the passing of an era in physics research, Oliphant wrote

Apparatus used in modern physical experiments is of engineering size. No longer is it possible to do first class work with sealing wax and string. The tradition of the Cavendish -- that the substances were essential, and practically the only essentials, of a piece of apparatus for physical research -- worked wonderful things in the days when the Master of Trinity225 presided over its fortunes. Many of us will regret the passing of that direct simplicity of approach, but not one of us can ignore it. We are forced to recognize that the physicist cannot now be self- sufficient and find among the flotsam and jetsam of Lincoln’s stores or among the junk stored in boxes of scrap metal, the necessary equipment for work of the highest quality. Hence we may feel that the time is ripe for a review of the situation, to see how the interactions of engineering and physics may best be employed to the advancement of either subject.

Electrical engineering grew directly from physics and it has made the greatest contribution to physics, both in the provision of technical equipment and facilities and by nurturing physical research in its own laboratories.226

225 A reference to JJ Thomson, who had become the Master of Trinity College in 1919, after he had stepped down as Cavendish Professor. 226 Oliphant (1938), p. 2-3 122

As an example, Oliphant recalled the labours of Faraday, who in his seminal researches into electromagnetism had to wind his coils by wrapping together twine, wire and calico, insulated wire not then being available. “Nowadays the branch of engineering which arose from this experiment provides the physicist with countless sizes and types of insulated wire with which a similar apparatus can be constructed in a few minutes. The mere act of providing for its own needs has made the electrical industry contribute this important article of laboratory equipment.”

Other powerful examples Oliphant chose to quote included the “provision of a steady and reliable supply of electricity for laboratory purposes” (recalling the horror stories recounted by Rutherford and Thomson of the times when such was not so), cheap and reliable meters for current and voltage (essential in the electrical and radio industries), and copper oxide rectifiers able to provide direct current when the supply authorities changed over to alternating current.

Oliphant made much of the contribution of John Cockcroft, trained in electrical engineering (at Metropolitan Vickers) before coming to the Cavendish. His skills and background impacted not only on his own research, but on those around him.

In the past ten or twenty years, the apparatus and methods of physics have tended to assume engineering proportions, and this change is largely due to the contributions of such workers as Cockcroft. When Kapitza and Lawrence and Cockcroft and others began their work, physical apparatus, even for the most complex experiments, could, for the most part, be placed on a few laboratory benches. Now as the result of the revolution initiated by them, we need electrical generators producing thousands of

123

kilowatts and great halls into which to put apparatus producing voltages not dreamed of in the past.227

The last statement was a clear reference to the high-tension laboratory which he had played such a role in creating.

Oliphant then referred to the work of Kapitza in more detail. Kapitza had increased five-fold the magnetic fields available to researchers by driving very heavy currents through coils, firstly by short-circuiting secondary batteries with low internal resistance, and then a 50 cycle AC generator, built with help from Metropolitan Vickers. This was nominally rated at 1000 KW, but could generate 55 000 KW over one hundredth of a second. The resulting magnetic fields reached an unprecedented 50 000 gauss. “This work opened up an entirely new sphere of investigation in physics. It was only possible by an application of engineering principles, and by the cooperation of skilled electrical engineers. Experiments of this type are costly, and cannot be borne by the ordinary funds of a laboratory. In this case, the cost of the machine was borne by the D.S.I.R.” Here is an early salute, not only to the inputs available from engineers, but also to the question of funding which was to loom so large in the Big Science era to come.

Oliphant now moved on to discuss developments more akin to his own interests, including developments we have already noted. He referred to the work of Allibone228. As we have already noted, Allibone worked for some years at the Cavendish applying high voltages to evacuated insulated vessels to accelerate electrons and ions to simulate radiation from radioactive sources like radium. Much greater fluxes had been obtained than from natural sources. For example, a current of one

227 Ibid, p. 6. 228 In contrast to the career of Cockcroft, Allibone had left the Cavendish to rejoin Metropolitan Vickers. 124

microampere was equivalent to 10 million million electrons, the beta ray flux from a kilogram of radium when usually only a few grams were available. Before he left to rejoin Metropolitan Vickers, he had produced considerable beams of “artificial” beta rays.

This work had been taken up by Cockcroft and Walton, firstly using a generator from Metropolitan Vickers with a peak potential of 350 000 volts. They then decided that direct current was more useful and developed “a very ingenious cascade system of rectifiers and capacitors which could provide extremely high potentials….The first apparatus built by them was very crude in appearance, for as in all physical development work, it was built of whatever materials were at hand, and which could be forced into service.”

Here again we see the sealing wax and string tradition, making use of whatever was available, but this was beginning to break down. Even at the Cavendish, outside help was needed. Oliphant recalled the vital role again played by Metropolitan Vickers.

This [the Cockcroft-Walton equipment] was the first large scale apparatus to use the new methods of pumping gas from exhausted vessels which had been developed by Burch of Metropolitan Vickers research department. Burch had found that it was possible to produce, by vacuum distillation of mineral oils, residual oils with very low vapour pressure. [These] could be used in place of mercury in the diffusion pump….

The great advantage of oil lies in its very low vapour pressure, which makes it possible to dispense with the usual liquid air trap. These “Apiezon” products have revolutionised the technique of continuously-evacuated high voltage apparatus and have

125

contributed as much as any one technique to the present day attack on the nucleus229.

Acknowledgment of this vital role played in the endeavour by Cockcroft and Walton in splitting the atom appears to be a strong endorsement of the part that industry can play in research in the physical sciences.

At this point, Oliphant moves very close to home.

The greatest technical difficulty faced by any laboratory which decided to follow this new line of work was that of providing a steady and copious source of accelerated particles. This problem was solved in a number of ways but the apparatus was invariably largely improvised and was somewhat capricious.

At this point, and greatly to their credit, the Philips Company of Eindhoven, Holland , through Dr Bouwers,230 decided to attack the problem, and within a few months they were able to sell a commercial apparatus at reasonable costs and of reasonable size, which would develop a steady potential of 1.24 million volts and delivered 4 milli-amperes of direct current. This beautiful piece of apparatus, a real triumph of engineering applied to physics, has now been further developed, and can be made to deliver 2 million volts and the same current. There is not time now to consider the many ingenious points of this equipment. Some of you will have taken the opportunities of seeing this beautiful apparatus in the high voltage laboratory of the Cavendish Laboratory.231

229 Ibid, p. 8. 230 Albert Bouwers was the head of the X-ray Department at Philips in the 1930s. 231 Oliphant (1938), p. 9-10. 126

With the aid of this apparatus it is now possible to carry out as much work in a few hours as it was possible to do in several days with the older equipment232.

Despite this strong endorsement of the contribution that engineering and industry could make to fundamental research in physics, not all was as it should be.

I would like to make one point which is of importance. When called upon to make this apparatus the Philips Co. quoted a delivery time of three months, and they adhered rigidly to this in both cases. They realised that important new equipment of physical research is needed promptly, and that if delivery is delayed there is great waste of time and opportunity, beside the fact that serious delay may make the apparatus out of date even before delivery. So often, apparatus of this type, which is for research purposes, is given scant consideration by the contractors, and [the] delivery date may be so postponed that the workers for whom the equipment was designed may have died or moved on. The research and development departments of English companies are fully alive to the needs of science, but those responsible for the production side relegate any research equipment to the very end of the production list. This is a very short-sighted policy, and does not help to keep English research and development in the forefront, but tends to make English scientists dependent more and more on foreign sources for research equipment.233

232 This can be seen as an example of the increased “productivity” to be gained from such early examples of the incipient era of Big Science. This issue will be considered in Chapter 9. 233 Ibid, p. 10. 127

Some other mixed messages are evident. After discussing the origin and early development of the cyclotron by Lawrence and his team in Berkeley, and praising the innovation for its powerful impact on nuclear physics research, Oliphant made the following comment:

In order to obtain high energies and high yields from such an apparatus, electro-magnets of very great size are required. These must be built by engineering firms who can handle such large masses of metal. In this country, cyclotrons have been installed in the Cavendish Laboratory, and in the Physics Laboratory in Liverpool, while a third and larger apparatus is contemplated at Birmingham. The magnets and some other parts of the cyclotrons in Cambridge and Liverpool were made by Metropolitan-Vickers, the general design being again due to our friend Dr Cockcroft. The magnets weigh about 50 tonnes, of which 7 tonnes is copper, and they produce a field of 19 000 gauss 11.2 cm wide and 90 cm in diameter. The oscillators, which produce about 80 kilowatts of power at about 30 metres, are built on standard radio lines.234

Yet Oliphant’s attitude to industry was not always positive. In The Two Ernests, Oliphant quoted from letters between Rutherford and Lawrence showing the latter’s enthusiastic willingness to support and assist anyone planning to build a cyclotron. He went on to comment

Unfortunately, the design of the cyclotron, and its brother for Chadwick in Liverpool, did not follow the lines developed in Berkeley235. It was entrusted to a large electrical engineering firm,

234 Ibid, p.10a, 235 Chadwick, now in at Liverpool, had embarked on the building of a cyclotron, using the same design and manufacturer as Cockcroft was using for the Cavendish machine.

128

with no previous experience, while funds were too restricted to enable the magnets to be as large as desirable. Much trouble was experienced with them, and they never performed as efficiently as the virtual copy of the 60 inch Crocker cyclotron built by us in Birmingham. However, they did useful work, and established the resonance accelerator technique in Britain236.

It seems appropriate therefore to describe Oliphant’s attitude to the engagement of industry in academic physics as ambivalent, as we are able to cite both positive and negative examples. We shall see this attitude worked out in the great enterprise of the proton synchrotron in the decade to come.

Oliphant’s appointment to Birmingham

In January 1936, the University of Birmingham advertised that its Chair in Physics would soon fall vacant upon the retirement of the incumbent SW Johnston Smith, who had held the post for two decades. The Chair was named after the physicist John Poynting who had occupied it (in the University’s forerunner institution Mason Science College) from 1880 until his death in 1914237. The stipend offered was £1300 a year. Applicants were required to send no less than 24 copies both of their application and of three testimonials. Applications were to close on 2 March.

Oliphant described becoming aware of the vacancy in the following manner.

236 Oliphant (1966) (2), pp. 45-46. 237 Poynting is remembered, inter alia, as the developer and eponym of the Poynting vector, which describes the direction and magnitude of electromagnetic energy. 129

Early in 1936, the Dean of the Faculty of Science in the University of Birmingham, Professor N[eville] Moss, called on Rutherford to ask if he could assist in finding a man to fill the Poynting Chair of Physics at the University. Rutherford sent him to talk to me. I was reaching the stage where I felt I wanted a show of my own. Rutherford had told me he intended to retire at 70, and I could not envisage working at the Cavendish under any other professor. So I told Moss I was interested.

Although he had sent Moss to talk with me, Rutherford was very angry indeed when I told him I was tempted to go to Birmingham. He grew red in the face and shouted that he was fated to be surrounded by ungrateful colleagues, and much else, ending with “Go and be damned to you!”

I had never been on the receiving end of one of Rutherford’s choleric outbursts, though I had heard of them, so left his room greatly upset and worried. Shortly afterwards, he came to my room where I was sitting in despair, and asked hesitantly if I could spare time to talk to him. His apology for his reception of my news was complete, reinforcing my distress at having upset him. He went on to discuss how the work upon which I was engaged could be continued, and what he could do to help me if I did go to Birmingham238.

We can fill in some of the background to this account by drawing on the Minutes of the meetings of the Faculty of Science at the University of Birmingham. On 6 March 1936 these recorded “The Dean reported that the Committee appointed to review the applications for the Chair of Physics had given preliminary consideration to the matter. The

238 Oliphant (1972), p. 152-3. 130

Committee was not however yet in a position to select a candidate for interview and proposed to make further enquiries”239.

It was apparently this lack of suitable candidates for the post that drove Moss, perhaps inter alia, to seek Rutherford’s assistance. The outcome was reported to the Faculty on 11 May.

The Dean informed the Faculty that since the last meeting the Committee appointed for the purpose had given consideration of applications for the Chair of Physics and had made further enquiries with a view to securing further candidates who might be suitable. As a result he was able to state that an application had been received from Dr MLE Oliphant, Fellow of St John's Cambridge and the Director of Research at the Cavendish Laboratory. The Committee commended that Mr RR James of the University of Manchester and Dr Oliphant be invited to appear at interview240.

In the meantime, Oliphant had been getting to know his potential new place of work. On 1 April Moss had written to Oliphant (calling him “Mr”) to suggest that he make arrangements to come to Birmingham in the next few weeks to visit the Department and meet Johnson Smith. This was a natural consequence of Oliphant being a candidate241. Moss had other news. He had discovered, contrary to advice he had previously given Oliphant, that the physics department was getting by on only £400 a year, a sum that “shocked” him. This, he presumed, was purely Johnson Smith’s choice, since he could have asked for more. The

239 Minutes of the Faculty of Science at the University of Birmingham Volume 7, p. 211. Archives of the University of Birmingham [hereafter cited as “BUA”.] 240 Ibid. p 215. 241 Writing on 4 April, in response to a letter from Oliphant, Johnson Smith regretted that, suffering from a severe chill, he might not be able to meet him when he came to visit, “but that he would “arrange for some competent person to be available … to show you over the Physics Department”. Johnson Smith to Oliphant, 4 April 1936. EBA. 131

chemistry department had an annual grant of £1650, and Oliphant would be quite justified in demanding up to £1000 if he were in charge of physics242.

When Oliphant made his visit, he must have met a number of staff, since he thought “my colleagues in the Department of Physics would make a good team with which to work”, but, perhaps not surprisingly “the equipment and facilities were not very attractive”. He also must have meet Norman Haworth243, Professor in the relatively affluent Department of Chemistry. Oliphant later reported that Haworth was “most persuasive and reassuring about what could be done244.”

Yet doubts must have lingered. On 17 April, Chadwick wrote to Oliphant from Liverpool245, in response to a letter from Oliphant which I have not been able to locate246. Clearly, Oliphant had sought his advice on whether he should accept the appointment if offered. Chadwick’s reply ran to six pages, covering both sides of the issue. On balance, Chadwick thought, it would be a good position to hold; the salary was higher than usual, the conditions as good as could be expected, the department’s reputation for teaching, if not for research, quite sound. Based on his own experience, Chadwick pointed out some of the frustrations: time lost in getting things moving, time wasted in committees. The atmosphere would not be like Cambridge, not having “the same tremendous surge of life”, but there would be compensations, most particularly “a laboratory of your own”.

242 Moss to Oliphant, 1 April 1936. EBA. 243 Haworth was to share the Nobel Prize for Chemistry in 1937 for his work on Vitamin C. (add reference). For his Royal Society obituary see Hirst (1951). 244 Oliphant (1967), Part 1, p. 2. 245 Chadwick to Oliphant, 17 April 1936. EBA. 246 The original of this letter may be held in the Chadwick papers at Churchill College, but has not currently been located . 132

That, Chadwick maintained, was the key issue. “To me it all depends on whether this is the kind of post you want… if you want a university chair and a laboratory of your own. If you are only lukewarm about this, then it would be a mistake to go to Birmingham or any other university. If you do want a chair, then I think arrangements can and must be made which should satisfy your obligations to Cambridge and to yourself.”

The obligations to Cambridge centered, as Oliphant had said in his letter to Moss, on the high-tension laboratory, which he felt he ought to see through to completion. Chadwick agreed that he too would feel some moral compulsion not to leave Cambridge immediately, but suggested that Birmingham would look favourably on a request to defer the appointment for a year (it would not really be decent to ask for two). After all, as he had previously noted, they would be getting “the best man available” for the post and a chance to bring the physics department there “to life”, with strength in both research and teaching.

On 22 April, Oliphant wrote to Moss247, saying that, having discussed the matter with Lord Rutherford and “with the others who are concerned”, he would be glad to accept the appointment if it were offered but could not take up the position for some 18 months, that is, until the beginning of the 1937 academic year (October), for reasons which he had already discussed with Moss. The first, and most important, of these reasons was his obligation to the High Tension Laboratory.

Lord Rutherford has allowed me a very free hand in the present line of development of the high-voltage research in this laboratory, on the assumption that I will be available, at least until the new buildings and the apparatus housed in it are in working

247 Oliphant to Moss, 22 April 1936. EBA. 133

order. The building will be ready in about 6 months, so there would then remain about one year in which to develop the department as a going concern. There is no one available at the present time with the requisite physical and technical experience to carry out this work in my place. I have worked over the last five years in very close co-operation with Lord Rutherford, and naturally I feel that it is impossible to leave him under the conditions which would apply at the present time.

Oliphant was also keen to finish the researches which he had underway at the time, which could not be “easily transferred to a fresh institution without serious interruption at an important stage”. There would be a benefit for Birmingham here. Finishing that work, and getting the HT laboratory running, would “enable me to make a much more definite choice of the mode of approach which I will use in Birmingham”.

There were two other issues which Oliphant had already discussed with Moss. One was the need for a “full-time personal assistant of the mechanic-technical type… to help with the making and erection of research apparatus”; the other for an initial outlay of £1500 to £2000 for research apparatus, with an increase in the yearly grant for apparatus and research to at least £1000. Oliphant was hopeful that these requests would be thought reasonable “should you see fit to appoint me to the Chair”. Moss responded on the 24th, commenting that he admired Rutherford’s “generosity in being willing to release you from the Cavendish in order that you might be available to take up duties here from October 1937”248.

We should note that at the same time as Oliphant was securing a promise of more money for research from the Birmingham authorities,

248 Moss to Oliphant, 24 April 1936. EPA. 134

Herbert Austin was making his extraordinary gift to the Cavendish. That too was announced in the same month. At what stage Oliphant became aware of the bequest is not clear, nor how it might have influenced his thoughts about going to Birmingham. However, the new funds did allow Oliphant to advance more quickly on the equipping of the high-tension laboratory, the chief reason for the requested delay in taking up the Birmingham post. It also may have focused his attention on the possibility of financial assistance from an industrial sponsor, leading ultimately to the securing of support from Lord Nuffield.

Oliphant, of course, did not yet have the position, though he was most likely the favoured candidate. The Faculty met on 15 May to interview both Oliphant and RR James, Reader in Experimental Physics at the University of Manchester. James was interviewed at noon and Oliphant at 12.30. That James was given only half an hour to impress suggests that the appointment was a “done deal”. Certainly the 16 members present unanimously agreed to appoint Oliphant as of 1 October 1937, and to agree with the conditions he had set and which we have noted above. It was further resolved that since Oliphant could not take up duties for almost 18 months, Dr Ibbs be appointed acting Head of Department with an honorarium of £200, and invited to attend meetings of the Faculty249.

A recommendation that Oliphant be appointed went before the University Council on 3 June, and was adopted, with Oliphant getting all he had asked for in terms of funds and technical assistance for research250. By mid June, the appointment had been announced and Oliphant was receiving notes of congratulation from colleagues like

249 Minutes of the Faculty of Science at the University of Birmingham Volume 7, p. 228. BUA. 250 CG Burton (University Secretary) to Oliphant, 25 May and 5 June 1936. EBA. 135

Appleton (soon to return to the Cavendish as Jacksonian Professor) and CP Snow251.

Early days at Birmingham

The major source of information and perspective on the early days of Oliphant’s tenure at Birmingham is his own memoir252, written in 1967 (some 30 years after the event) at the request of the University. While this deals mostly with the project to build a large cyclotron, there are some details of the context. In some cases, Oliphant’s recollections can be checked against correspondence from the time or against the brief history of a century of physics at the University written by Moon and Ibbs253. However Moon and Ibbs said almost nothing about the cyclotron, referring the reader to Oliphant’s memoir.

When Oliphant took up the post, the facilities were far from impressive.

…. the Department of Physics was housed in the original Poynting Building and in some wooden huts, erected after the First World War, between the Poynting Building and the Library (Harding). With the exception of an L-shaped laboratory set aside for the professor’s use, a roomy but cluttered basement and a small room opening off a cloakroom, there was no space whatever that could be used exclusively for research purposes. The sum available for all purchases and maintenance of equipment for both teaching and research was £2000 a year. The small workshop was quite inadequately equipped for research purposes and the technical staff very small and occupied almost exclusively with the teaching laboratories and lecture theatres.

251 These notes are held in EBA. 252 Oliphant (1967). Part 1. 253 Moon and Ibbs (1980). 136

As Oliphant commented, this was a “grim period” for science in Britain. The impact of the depression on industry was only slowly being moderated by the first stirrings of rearmament, mounted in the face of the worsening climate in Europe, industrial production was down, money for all purposes was short. While Oliphant had been given an undertaking about recurrent expenditure, significant capital outlays, as would be needed for research into newer areas of physics, would be much harder to source, at least from within the University.

In April 1938, 6 months after he took up the appointment, the University Council received a report from Oliphant, commissioned by Moss and the Vice-Chancellor and cataloguing the deficiencies of the laboratory facilities.

The equipment is almost wholly out of date and in many cases useless, and the facilities required for modern work in physics are completely absent. By clearing a portion of our basement and dividing it into three rooms with temporary partitions, and by clearing two rooms used for storing instruments and apparatus, temporary accommodation has been provided for some of the research, but as a result of this I myself have nowhere to work.

The condition of the huts in which Course 1 practical classes are held is precarious and the time has come to review the temporary expediency, adopted 20 years ago, of housing a considerable proportion of the department in those huts254.

To address these deficiencies, Oliphant’s report proposed refitting the existing Poynting building, including his own office and laboratory, to

254 Quoted in the Birmingham Post, 7 April 1938. Clipping from BUA. 137

meet the needs of teaching, and building a new wing largely devoted to research. This would contain space for Oliphant himself, his secretary and administration, library and staff rooms, machine room, workshop and dark rooms, and most importantly a large research hall for work at high voltages and with large equipment, eight smaller research areas and a basement room for “high-energy X-ray work”. Oliphant wanted as much of this to be lightly-built and flexible, able to be reconfigured as research needs changed. This was the policy he had fought for with the high tension laboratory at the Cavendish.

Building and equipping this new facility was expected to cost £60 000. The Council was sympathetic to the project but felt unable to fund it. The Minutes of its meeting on 6 April read in part

The Council considered the recommendation of the Finance and General Purposes Committee with regard to the pressing needs of the Department of Physics, involving a capital outlay of about £60,000. It unanimously recognises the urgency of the requirements, but owing to the present state of University finances, regrets that it is not possible at present to put this work in hand. It is, however, most desirable in the interests of both teaching and research that there should be no delay, and the Council would be most grateful for any financial help from outside to enable the University to meet these essential needs255.

The situation was more difficult, as the Birmingham Post pointed out in its leader of 7 April, because the University was at the same time seeking to raise money for a new Medical School. The resolution was, the leader said,

255 Minutes of the Council of the University of Birmingham, 6 April 1938. BUA. 138

…in effect, a prayer that one or a few benefactors may open their purses in the cause of physics without prejudice to the appeal which must be made on behalf of medicine. The Council may reasonably hope that in Birmingham, of all places, its prayer may not go unanswered, for physics is a subject with which the industrial life of Birmingham and the West Midlands is intimately concerned256.

As an example, the leader cited research already underway in the department into the strength and plasticity of metals, “researches that have a direct bearing on the question of fatigue and fracture in metals used in industry”. The leader concluded, rousingly “If the University of Birmingham is to rise to its opportunity in physics, and the industrial community around it is to enjoy both the benefit and the prestige that might accrue therefrom, the money for these essential extensions must be found. There is a clear call to the bounty of the Midlands.”

Oliphant himself was already ploughing that furrow. His vision for the refurbishment of the Department was not unlike that which Rutherford had put forward in 1919, but unlike his mentor, Oliphant was prepared to go out and try to find the money he needed. In his memoir he recounted that he made visits to “every large firm in the area”, who received him well and were happy to show him over their works, but could offer no financial support. They could build equipment or forge components for him, but only on normal commercial terms, “even when the Chairman was a member of the Council”. Business was so bad that nobody could afford to do anything for nothing.

Having met Lord Austin at Cambridge at the time of the bequest to the Cavendish, he boldly sought a meeting to see if Austin, whose

256 Birmingham Post, 7 April 1938. 139

manufacturing works were in Birmingham, had more largesse to offer. The answer again was “no”257. It was time to look elsewhere.

These fruitless attempts to obtain the funds we needed in the Birmingham area led to explorations further afield. Professor Moss and I approached the then Prime Minister, Mr. , a Birmingham man whose family had virtually established the University. He was not of a warm disposition, and did not give the impression he could give much help, but he did indicate that he would look in the question. Sometime later, we were very pleased indeed to hear from Lord Nuffield258 that he was prepared to make a gift of £60 000 for our purpose and he wrote to the University accordingly.259.

We can flesh out this account by reference to correspondence held in the Chamberlain Papers at Nuffield College. This records that on 27 April, Moss wrote to the Prime Minister as follows

In view of the keen interest which you and your family have always taken in this University, I am taking the liberty of addressing this private and confidential letter to you in the hope of soliciting your advice and perhaps your help.

257 Oliphant (1967) Part 1, p. 4. Oliphant entertainingly recalled afternoon tea with Austin and his wife, “a formidable woman…with an abrupt but disarming frankness in conversation”. She told Oliphant almost at once that if he was after “the sugar”, they had already given away all they could afford. 258 William Morris (1887-1967). known as Lord Nuffield from 1934. British industrialist and philanthropist. Founder of the Morris Motor Works, and also of the Nuffield Foundation and Nuffield College, Oxford. For more information on the life of Nuffield, see Andrew and Brunner (1955), Hull (2008). 259 The Birmingham Gazette of 30 June 1938 quoted a University official as saying of Nuffield’s letter of offer “There were hardly five words in the letter, but what happiness those few words gave us”. Clipping in EBA. 140

The enclosed article taken from a recent issue of the Birmingham Post will, I think, enable you to acquaint yourself fully with the position. In addition to the very urgent needs of the School of Physics, the Science Faculty is badly in need of Research Fellowships and money to complete the new extension to the Chemistry Department260.

On receiving the letter, Chamberlain wrote by hand in the margin “I am taking this up personally”. Moss had sought a meeting with Chamberlain in London or Birmingham, which the Prime Minister was happy to agree to, as his Private Secretary informed Moss on 29 April, but it would have to be in London. The meeting must have taken place some time in May since by 29 May, Chamberlain was ready to act. Clearly he had been sufficiently impressed by meeting with Moss and Oliphant and hearing their story that he wanted to assist. His letter to Nuffield261, written on that day began

Before I entered national politics I was for many years intimately associated with the Birmingham University. It was, as you probably know, practically founded by my father, and I was one of the original members of the Council.

I suppose the family connection is the reason why they have recently come to me to ask if I could help them. Great effort has been made to complete the new combined hospital and medical school but they still lack another £60 000 to get this finished and that precludes them from making a public appeal for the other needs.

260 Moss to Chamberlain, 27 April 1938, NC 5/8/1/63. Chamberlain Papers, Nuffield College. 261 Chamberlain to Nuffield, 29 May 1938, NC 5/8/1/66. Chamberlain Papers, Nuffield College 141

I enclose a list of these. The biggest item is for Physics. They have got Oliphant, who was Rutherford's assistant and is continuing his work with great distinction, but their equipment is old and out of date and they want more room. If they could get the money, they have a good chance of leading the world in this department.

Clearly Oliphant, with his connection to Rutherford, had been one of the selling points for the appeal to the Prime Minister; this was appropriate since his needs had been the original motivation for the search for financial support. The letter was sufficient to elicit a response from Nuffield. On 23 June, Moss was able to inform the Prime Minister that Nuffield’s solicitors had been in touch with the University and that £60 000 would be made available. The gift would be reported to the Council on the 29th. There had been no public announcement as yet, though Nuffield had given permission for one. Moss thought that Chamberlain's name should be associated with any announcement, but Chamberlain did not want that, as a hand-written comment on the letter indicated.262

It is not clear why Chamberlain settled on Nuffield as the likely source of the needed support. There is no evidence of any particular personal or family connection with him, though Nuffield was a noted supporter of the Conservative Party and had donated £5000 to the Joseph Chamberlain Centenary Fund in 1936. Chamberlain may have been inspired by the knowledge that Nuffield’s fellow motor car magnate Austin had provided substantial support for research in physics at the Cavendish; Nuffield would presumably have been aware of this also. Interestingly, Nuffield was not noted for his support for the physical sciences. Much the greater part of his generosity was directed towards medical research and nursing, including a particular interest in the provision of "iron lungs"

262 Moss to Chamberlain, 23 June 1938, NC 5/8/1/67. Chamberlain Papers, Nuffield College. 142

for sufferers from polio and in the welfare of "cripples", that is, for orthopaedics263.

We can also consider whether Nuffield was rewarded for his generosity in Birmingham (as Cockcroft claimed Austin has been). He was already a baron (since 1934), he would become a viscount in early 1939, and a Fellow of the Royal Society in the same year. The University awarded him an honorary degree in 1938 but there is evidence that was already in train by the time the approach for support was being made.

Whatever the truth on these issues, events had moved remarkably quickly, with the gift being announced by the Council on 29 June, within 2 months of the Council’s “prayer”. On 4 July, the Birmingham Post carried comment from Oliphant on the impact of the gift, £40 000 of which would be used to build and equip the new “Nuffield Laboratory”, with the rest being used as an endowment to pay for maintenance and to found a research scholarship. Oliphant wrote at some length about the many ways industry can benefit from research in physics, but did not mention that the main focus of research in the new facilities would be in nuclear physics, a field whose impact on industry would be at a notably greater distance.

263 Nuffield kept a record of his donations in a simple hardcover exercise book, now held in the Nuffield Papers. The donations are listed alphabetically and then chronologically over time. The reference to Birmingham is very succinct (as indeed are all the entries): "Birmingham University. Physics Department covenant. £60 000. 1938”. The grant to Birmingham was relatively large; of the many dozens of entries since 1936 most are for donations of less than £10 000. The largest was for £2 000 000 to the Oxford University medical trust in 1936. In the same year he gave £200 000 to found the Nuffield Orthopaedic Centre. Nuffield’s earliest support was for hospitals in Birmingham and Coventry where he had factories. The only other large donation for research into the physical sciences was £25 000 to the Jodrell Bank radio telescope, and that was many years later. The biggest donations were in 1943 and 1944, a total of £10 million to establish the Nuffield Foundation. In the postwar years , as a trustee of the Foundation, was to help direct significant funds toward research in nuclear physics, including that undertaken by Oliphant. See Clark (1972), pp. 58-9. 143

Moon and Ibbs stated that from the start, Oliphant’s intention was to orient his new Department strongly towards research into nuclear physics, though he encouraged the continuation of existing work into areas like thermonics and meteorites. The first senior researcher to join him at Birmingham was JT (John) Randall264, whose expertise was in luminescence. As Moon and Ibbs reported

To get some nuclear physics going in the meantime, he obtained parts for an accelerator of the Cockcroft-Walton type, which HAH Boot and PB Moon assembled in the professor’s private laboratory. It attained as high a voltage as the room allowed (about 300 KV) but by the summer of 1939 war was approaching and the ion-accelerating tube was never completed. Some components were later used in the 500 KV injector of the proton synchrotron265.

Moon and Ibbs also maintained that a decision to build a cyclotron in Birmingham had already been made by the time Oliphant arrived in October 1937 and that had necessitated many trips by Oliphant between Cambridge and Birmingham over the previous year. This appears incorrect. Oliphant’s plans for the laboratory made no reference to a cyclotron, even if the goal of building one was in his mind. The basement of the new wing was to be used for “high-energy X-ray” work, technology unspecified. In an interview with Weiner266, Oliphant stated that if he had not been able to get enough money for a “decent-sized” cyclotron, he would have built a Van de Graaff generator instead. That was perhaps why his plans for the new laboratory included a large space for working with high voltages.

264 (1905-1984). English physicist and biophysicist. Previously employed (since 1926) in the General Electric Company (GEC) at Wembley. Appointed to Birmingham as a Royal Society scholar in 1937. For overview of life and work, see Wilkins (1987). 265 Moon and Ibbs (1981), p. 19. 266 Weiner (1971), p. 29. 144

This interpretation of events is supported by correspondence between Oliphant and his relatively new Birmingham colleague Rudolf Peierls267. In addition to offering Peierls accommodation in the still-to-be-built new laboratory for his Department of , Oliphant alerted him to his plans, expressed in terms that indicate they were relatively new.

I don't know if I have told you but I have decided to go for a large cyclotron, as it seems to me that this method, or modifications of it, offers the only real possibility of obtaining really high-energy particles. It would have been nice to obtain electrons with energy of the magnitude of 100 million volts, but in the absence of those particles, very useful information should be obtained from protons and other particles moving at energies of 50 million volts. The rest of the laboratory has, therefore, been changed to fit in with this new conception268.

Peierls replied “I note with interest that you have decided to go for a cyclotron, and that you think you can get to 50 million volts. If that can be done, it will of course lead to extremely important information.”

Two points are worth noting here. We see again Oliphant's drive for the largest available energies, though he was willing to settle for less on the grounds of feasibility. At the same time we see the theoretician Peierls endorsing such a quest, on the grounds that it will open up "extremely

267Rudolf Peierls (1907-1995). German-British physicist. Of Jewish extraction, he settled in Cambridge when Hitler came to power in 1934, Appointed Professor of Mathematical Physics at the University of Birmingham in 1937. Gained British citizenship in 1940. For overview of life and work, see Lee (2007). 268 Oliphant to Peierls, 27 July 1938. Box 22X/6. Special Collections, University of Birmingham Library [hereafter cited as “SCBU”]. 145

important information”. We have seen this line of argument before and will see it again.

Seeking help from Lawrence

For Oliphant, the Nuffield gift had totally transformed the picture, making possible the building of a cyclotron, in the same way that the Austin bequest had so greatly advanced the building of the Cavendish HT laboratory. Within a few weeks of the announcement Oliphant had revived his long-dormant correspondence with Ernest Lawrence, whose first cyclotron had been set running in 1932. His letter to Lawrence on 19 July 1938 requested some details of the large cyclotron which Lawrence was then building, Oliphant having plans to build “as large an accelerator as possible”. Was it possible, he asked, to go to even higher energies than Lawrence was then pursuing? He concluded by expressing the hope that he would be able to visit Berkeley in the New Year and that the two of them would have a chance to talk things over.

Lawrence replied on 2 August269, delighted to hear that Oliphant was planning to build a big cyclotron and assuring him a cordial welcome should he be able to come to California. They were, he felt, still well short of the practical upper limit of energy for such machines. Problems would arise but ways could be found around them. The cyclotron he was then building had 60 inch pole faces and would deliver deuterons at 25 MeV and alpha particles at 50 MeV.

In reply Oliphant indicated that he wanted to rely heavily on the Berkeley experience: ”I have decided that I could do no better than to build a magnet of about the same size as the one you mention in your letter. An increase above that size seems to mean a disproportionate increase in

269 Lawrence to Oliphant, 2 August 1938. LPBL. 146

the amount of steel and copper. I am concerned about the supports for the upper magnet, which will weigh 15 tons270.” Oliphant was here indicating a concern which would in time become one of the major drivers to the development of the synchrotron, the burgeoning masses of metal that seemed inevitable in the quest for higher cyclotron energies.

With regard to the apparatus that would be required to generate the alternating electric field that would accelerate the protons, Oliphant made a comment that seems atypical. “Perhaps we should buy a commercial high frequency generator as are now being made in the UK and in Holland for television.” Oliphant had by this time had some experience of buying large-scale equipment, rather than building it in- house. Lawrence was quick to seek to change his mind. “Wait until you get here before you decide on [high frequency] equipment. Building your own will be equally satisfactory and much cheaper271.”

In the meantime, Donald Cooksey, Lawrence’s deputy, was gathering together copies of the blueprints of the cyclotron they had just built, having been told by Lawrence that Oliphant was considering the construction of a magnet of roughly the same dimensions, and added to them what he termed “a flood of technical detail”. Oliphant acknowledged their arrival on 1 October. “It was good of you, and of Professor Lawrence, to pass on this information which must have cost you a great deal of time and considerable patience and ingenuity… I sincerely hope I am not making a nuisance of myself in asking for this information”272.

270 Oliphant to Lawrence, 30 August 1938. LPBL. 271 Lawrence to Oliphant, 22 September 1938. LPBL. 272 Oliphant to Cooksey, 1 October 1938. LPBL. 147

In his reply273, Cooksey commented that he was “…most impressed by your letter being dated October 1st. I would have supposed that it would be very difficult for a man in Birmingham to notice such things as the arrival of blueprints so shortly after the Munich pact.”274 Not necessarily so, Oliphant replied. “Though it may surprise you, I assure you that we in Birmingham are more interested in our own scientific work than in the possibility of war in Europe. However there is no doubt that the war atmosphere seriously interferes with what one hopes to do275.”

Oliphant of course had more to consider than just the cyclotron. For one thing, it would have to be housed. By late 1938 it was evident that the building would cost £24 000. It could not be done for less because of the insistence of the University that it would have to follow the style of surrounding buildings, which were faced with red brick and embellished with stone. The “Nuffield Laboratory” would be filling a gap in the existing semicircle of such buildings. Behind the façade, the building to house the research and teaching would be of a simpler character and devoid of any extravagance. The equipment already ordered would leave very little change from £20 000.

By this time the details for Oliphant's visit to Berkeley were in place. He would arrive in New York on 15 December, after a five day crossing by ship. A letter to Lawrence late in November276 proposed that he delay his visit to Berkeley, since both Cooksey and Lawrence were to be absent from the laboratory at times over the holiday period. However it appears that at Lawrence’s suggestion he made the acquaintance of

273 Cooksey to Oliphant, 17 October 1938. LPBL. 274 The meeting in question took place in Munich between Hitler, British Prime Minister Chamberlain and the leaders of France and Italy and had resulted in a pact, signed on 29 September, which allowed Germany to annex much of Czechoslovakia. It was a pointer to the worsening international situation which would so impact on Oliphant’s work and ambitions over the next decade. 275 Oliphant to Cooksey, 9 November 1938. LPBL. 276 Oliphant to Lawrence, 21 November 1938. LPBL 148

William Brobeck, who was in charge of much of the technical design for the cyclotron. Brobeck was to later play a major role in the genesis of the American synchrotrons277.

The dates of his visit to Berkeley are not clear. It is known that by 11 January Oliphant was back in New York, writing by hand to Lawrence from Columbia University to return the keys to the laboratory, which he had inadvertently taken away, and to repay the balance of his account at the Faculty Club.

I find it very difficult to thank you for the magnificent and instructive time that I had in Berkeley. It was truly fine of you to be so liberal with your time and thought on my behalf. I know of no laboratory in the world at the present time that has so fine a spirit or so grand a tradition of hard work. While there I seemed to feel again the spirit of the old Cavendish, and find in you those fine qualities of the combined camaraderie and leadership which endeared Rutherford to all who worked with him. The essence of the Cavendish is now in Berkeley. I am sincere in this, and for these reasons I will return to you some day, and I hope very soon.

I have been digesting some of my experiences while on the train. Many things about the cyclotron are now clear, which formerly were hazy, while the physics I learnt was of inestimable value. I fear that I neglected to say goodbye to all who had been kind to me. I hope you will assure them, if so trivial a matter should arise, that I do indeed feel grateful for all the help and encouragement which I received. I return with a greater

277 As we shall see, Brobeck was responsible for the design work that resulted in the building of the Bevatron at Berkeley and stimulated the building of the Cosmotron at Brookhaven. 149

confidence and a greater belief in the cyclotron, in physics and in mankind278.

The profound influence exerted by Lawrence on Oliphant was clearly expressed. Indeed, Oliphant was still sufficiently in awe to head this letter “Dear Professor Lawrence”. The more familiar “Dear Lawrence”, which in the convention of the time was an acknowledgment of face-to- face contact, would come later. Oliphant would return to the parallels between Lawrence and Rutherford thirty years later in The Two Ernests.

On 21 March, back in Birmingham, Oliphant wrote again to Lawrence279, commenting how much his usual work was being disrupted by matters in connection with defence, and reporting some progress on the cyclotron, made with the help of his old Cavendish colleague RR Nimmo280. With his background in the technology to generate high frequencies, Nimmo was to look after the oscillator system and related apparatus, while Oliphant dealt with the design of the “Dees”281, modifying slightly the system Lawrence had used. In general, however, Oliphant was sticking closely to the Berkeley model, and was keen to hear both of progress in Berkeley and of any comments Lawrence had to offer.

I am anxious to know how your cyclotron is progressing and in particular whether you have yet been successful in applying voltage to the Dees. If, as a result of your preliminary work, you have seriously modified your designs, it would help me

278 Oliphant to Lawrence, 11 January 1939. LPBL. 279 Oliphant to Lawrence. 21 March 1939. LPBL. 280 New Zealand-born RR Nimmo had studied at the Cavendish as an 1851 Scholar from 1926, securing his doctorate in 1929. From 1929 he was on the staff of the University of Western Australia, before being appointed in 1938 as the first Nuffield Fellow at the University of Birmingham. 281 The “Dees” were a pair of flat copper chambers, shaped like the letter D but with their straight sides facing each other. The particles being accelerated spiraled outwards from the centre of the circle defined by the Dees. The acceleration took place as the particles crossed the narrow gap between the Dees. 150

considerably if you could indicate the directions in which this has been done. I hope you will forgive for following the designs laid down by your people so slavishly, but I feel it would be dangerous to embark on anything novel after the unfortunate experiences of the folk in Cambridge and in Liverpool.

However, as Oliphant’s 1967 memoir282 recorded, some of the changes required were substantial. The Berkeley 60-inch “Crocker” cyclotron, on which Oliphant was modeling his own machine, was intended for medical outcomes as well as scientific ones. The coils of the American machine were to be cooled by circulating oil, a system judged both too expensive and too dangerous in the confined space being prepared for the Birmingham apparatus. Water cooling was likewise not economical, requiring a reliance on forced-air cooling, with inconveniently large air ducts.

Oliphant went on to detail a series of technical challenges, some due to manufacturers who did not (or could not) keep to specification. The vacuum chamber as delivered was a millimeter or more out of tolerance and had to be scraped by hand by the dedicated technician MP “Jimmy” Edwards. Other tasks, such as the threading of large copper tubes, required much ingenuity by the technical support team, men like SR “Mick” Cornick and AW Holland. The winding of the copper coils for the magnets was a similar triumph of improvisation, which Oliphant recounted with relish. He was also delighted when the manufacturers exceeded his expectations. “The copper Dees were made for us by an old-fashioned firm of coppersmiths in Birmingham. The work was done entirely by hand, but when finished they were beautiful examples of the work of true artisans, accurately to dimension and with a mirror-finish.”

282 Oliphant (1967) Part 1. 151

Picking up the tempo

It appears from Oliphant’s memoir that the pace accelerated through the early months of 1939, and that the long-sought support from local industry began to appear.

Fortune began to smile on us. We were able to acquire from the Manchester Corporation283 a 600 KW D.C. generator driven by a 11,000 volt A.C. motor at no cost beyond installation. Dr McCance, of the Glasgow steel firm Colvilles, agreed to supply the steel for the magnet, ready machined by Harland and Woolf, at a nominal cost. Plans for a very simple laboratory block were produced rapidly and the building commenced. The local branch of ICI Metals undertook to manufacture the copper strip required for the magnet coils for the cost of the raw copper required. The Birmingham Battery Co made seamless tubes of copper for the dees and resonant vacuum lines.

Oliphant’s letter of 30 May284, in reply to one (or several) that have apparently note survived, was the first to begin “Dear Lawrence”. Oliphant acknowledged receipt of the pleasing news that Lawrence’s big new cyclotron was in preliminary operation, before reporting on some technical difficulties his team was having in winding the copper coils. His next paragraph showed that he, like Lawrence285 and others, was already seeing uses for the cyclotron beyond nuclear physics research, even if he was cautious about the possibilities.

283 The Manchester Corporation was the governing body of the nearby city Manchester, responsible for, inter alia, the provision of tramway services, for which this generator may have been used. 284 Oliphant to Lawrence, 30 May 1939. LPBL. 285 Ernest Lawrence’s brother John, a physician, was an early experimenter in the use of cyclotron radiation in medical therapy, including using beams of neutrons to treat cancer. 152

I spoke at an informal meeting of the Association of Physicians here in Birmingham last week, concerning possible medical applications of the cyclotron. The more I think about these questions, the more convinced I am that for the moment these applications must be on the research side, and that any clinical applications must be introduced slowly and cautiously as our knowledge develops. This view was enthusiastically supported at the meeting, and I am sure I will have the full co-operation of the physicians of this town and country, in any medical work we may carry out with the cyclotron here286.

The letter carried less optimistic news as well. In reply to an invitation which Lawrence had apparently offered, Oliphant wrote

I am afraid that it is quite impossible for me to come to California during the coming summer, as the international situation is still extremely tense. It is difficult for those who are not engaged in the hurly-burly of defence preparations to realise the state of continual anxiety under which we work. Rumours come, and are multiplied, about the activities of other counties, and of their knowledge of our defence program, which make it necessary to be continually on the qui vive, and ready to modify any arrangements that may already be made.

It was in this uncertain atmosphere Oliphant and his team pushed ahead with their cyclotron. Letters to Lawrence reported a number of problems, “nothing very serious”, but requiring small modifications in design. Oliphant remained keen to hear any comments Lawrence might have on those modifications. The oscillator was to be built very much on

286 The possibility of medical benefits from the cyclotron would have helped garner local support. It would have pleased Nuffield as well. 153

Berkeley principles, so no problem was expected. The steel for the magnet had been due for delivery by road from Glasgow in May.

The project was beginning to attract the attention of the press, including in Glasgow, source of the machined steel plates from which the magnet would be built. Under the headline "Experts to test Scots-built atom machine", the Glasgow Express287 reported in June

The cyclotron which is designed to produce a substitute for radium for the treatment of cancer leaves for Birmingham University this week, where it will be tested by Prof ML Oliphant. Built at Colvilles Limited, and Harland and Wolff Glasgow, it is the largest in the world. It weighs nearly 300 tons.

Five million volts of electricity, carried by hundreds of miles of wire, will be played between the two magnets on the element chosen to be radioactive. To reduce the risk of accident to the experimenters, the cyclotron will be sunk 15 feet below ground in a room surrounded by 3 feet thick walls containing water.

Scientists all over the world will watch closely the experiments which will assist ridding the world of cancer. Smaller plants have been used with varying success in Japan, the United States and Germany.

An artist's impression accompanying the article showed a large rectangular frame. It appears to be about 4 metres by 5 metres, when compared with images of men drawn standing nearby wearing cloth caps. The pole pieces were clearly visible in the centre of the frame. The caption read “5 000 000 volts of electricity will be passed through the

287 Glasgow Express, 5 June 1939. Clipping in Box 22X/6, SCUB. 154

poles, through a foot-long spark. The element to be made radioactive will be placed in a special container between the poles and when bombarded by the spark it is hoped will produce the new element.” Clearly the writer had little grasp of physics.

The article noted several times that the machine was to be the largest of its kind in the world. The correspondent may not have understood how the machine would work, but he made several references to the role of the project in promoting the treatment of cancer. That, rather than advances in nuclear physics, was presented as the main justification for the cyclotron. We can assume that that point was made strongly by Oliphant or members of his team who were presumably the source of the technical information on which the article was based. To the non- technical readers of the newspaper, as for the physicians of Birmingham, that may have appeared a more appropriate reason for the expense, even if, as we have noted above research in physics was more important for Oliphant himself. 288 . Oliphant reported a setback in August289, with both Nimmo and a laboratory assistant suffering serious leg injuries during the assembly of the base of the magnet290. Keen to see the work completed, the pair had been working overtime on the job, though it usually required a team of four. Both had their legs broken when a steel plate weighing more than a ton toppled onto them. Though both had made a good recovery, work was held up, in part by the need to put in place more stringent safety rules. Still, the suppliers had done an excellent job of fabricating the

288 A briefer report appeared in the Glasgow Evening Times on 29 June 1939. A caption to a picture of Oliphant read” Prof Marcus LE Oliphant, 38 year old Director of Physical Science at Birmingham University, is to direct the work in a Birmingham laboratory built underground in secret with money provided by Lord Nuffield. The laboratory is to house a 280 ton machine to tap the atom. Apparatus is now being completed in Glasgow". Clipping in Box 22X/6, SCBU. 289 Oliphant to Lawrence, 24 August 1939. LPBL. 290 According to newspaper reports (e.g West Australian 28 July 1939), this accident occurred on, or a little before, 27 July 1939. The incident had been reported in Western Australia because of Nimmo’s previous association with the university there. 155

steel, which went together to form the magnet “with all the precision we had requested”291, despite a total weight of 200 tons. Oliphant was able to enclose a photo of the assembled magnet in his letter to Lawrence.

Other aspects of the project were coming together. The new building had been roofed. Within it the 5 metre-deep pit equipped with a 5 ton crane was ready to receive the machine. The dees were nearly ready; other difficulties had been largely overcome. These had included the need to build a small power station, with transformer and switch gear, to supply alternating current at 11 000 and 230 to 400 volts, only direct current being available to the Department at the time. Though most of the research and technical staff would be away during September to “carry out some work for the Air Ministry”, Oliphant saw the end of the project coming into view. “Given a clear run, I am confident that we could have the apparatus functioning in a preliminary manner by Xmas.”

He did not get that “clear run”. Less than two weeks later, Britain was at war with Germany. On 18 September Oliphant wrote to Lawrence “our own work on the cyclotron has come to an abrupt end. I cannot see any prospect of any work being done for the present and probably not until the end of the war. …In the meantime we are carrying on with certain Government research work which will occupy our time very fully”292.

Accumulated “capital”

As we approach the end of the decade, let us summarise the “capital” Oliphant had built up through the 1930s at the Cavendish and then in Birmingham. During this period he had constructed, or commenced to construct, at least three accelerators, and as many as five if the various

291 Oliphant (1967) Part 1, p. 10. 292 Oliphant to Lawrence, 18 September 1939. LPBL. 156

generations are counted separately. There is little doubt that the events of the 1930s had added significantly to Oliphant’s reputation as a builder of advanced apparatus. His success in the development of new apparatus, as exemplified by the Cambridge high-tension facility, is referred to in the citation for both his election as a Fellow of the Royal Society on 6 May 1937293, and the award of the Hughes Medal by the Royal Society in 1943, the latter being for “original discovery in the Physical Sciences, particularly electricity and magnetism and their applications”.

The first citation states that “[He] has been active in the design of high- voltage apparatus for the production of swift positive ions”. The Hughes Medal citation is more expansive. After describing his contribution to the growing understanding of nuclear reactions, particularly his work on the deuterium reactions and the masses of the light elements, this states “Oliphant has been particularly successful in the design of apparatus. Besides that required for the researches to which reference has been made, he has been responsible for the development of the two-million volt installation at Cambridge. During the war, he has successfully carried out work of great importance”294. The nature of that “work” could not, of course, be disclosed at the time.

The Hughes Medal citation also makes the following significant point. “In the published accounts of the work, Oliphant’s name is often associated with that of others, including Rutherford, but there is no doubt he was responsible for the very specialised experimental

293 Oliphant had an impressive list of supporters for his election, a number of whom were or would become Nobel Prize winners; Rutherford, Cockcroft, FW Aston, CTR Wilson, GP Thomson (son of JJ), and PMS Blackett (but not, for some reason, Chadwick), as well as CD Ellis, RH Fowler (Rutherford’s son-in-law) and the Australian TH Laby. 294. These comments were included in the record of the Anniversary Address by the Royal Society President, Sir Henry Dale. See Dale (1943), p. 232. 157

technique, as well as for other contributions to the collaborative enterprise.”

Beyond reputation, we can cite a growing network of influential contacts, most notably Lawrence, Cockcroft and Chadwick and a fount of insight and practical experience into the building of large research facilities. He could legitimately have been regarded as one of the most experienced “accelerator men” in the world, so diverse had been his work. The Mond, Austin and Nuffield bequests had raised his awareness of the availability of the funds needed for such enterprises, if potential patrons were appropriately handled. His interaction with industry had not produced a consistent attitude with, as we have noted above, both good and bad experiences to report.

He had also begun to exhibit the force of personality and a capacity to connect and persuade that would prove so important later. The Faculty had been unanimous in recommending him for the Poynting Chair, and had agreed with his quite stringent conditions, which might not have been so with a less impressive candidate. The University had agreed wholeheartedly with his plans to expand the Physics Department, even if circumstances prevented that enthusiasm being backed with money. He appears to have won over Chamberlain sufficiently to inspire the approach to Nuffield, and he had begun a relationship with the media which would prove productive.

He had also demonstrated a notable willingness to push the envelope, to go for the bigger outcome. This was shown in the construction of the basement accelerator, where (with Rutherford’s encouragement) he sought to outdo Cockcroft and Walton in the numbers of disintegrations that could be stimulated, even though bombarding energies were lower, in him stressing to Rutherford the importance of the highest possible

158

energies for bombarding particles, (and then being willing to shoot for two million volts when only one million had been demonstrated as feasible), in him asking Lawrence, with regard to cyclotron energy “how high can you go?”, in him deciding that his Birmingham cyclotron would be the biggest in Europe. The push for the highest attainable energies would characterise his next phase of machine building once the war was over.

159

CHAPTER FIVE Oliphant at War: 1939 to 1945

Though he could not have anticipated it at the time, the half-decade following the outbreak of war in late 1939 was to have a profound influence on Oliphant's post-war activities, and in particular on the genesis of his grand project to construct a new form of particle accelerator. The wartime years added greatly to his stock of “capital” he developed or deepened important contacts, his own status was substantially enhanced by the key role he played in defence-related research, he gained valuable experience in the construction of large- scale apparatus, he learnt to manage and direct large teams of colleagues and to interact with bureaucrats at the highest level. Furthermore, as we shall explore in detail in Chapter Six, his work in the United States on aspects of the development of the atomic bomb provided the context within which his proposals for the Birmingham proton synchrotron first took shape and were first committed to paper.

We shall also examine in the current chapter the extent to which the nature of his work within the Manhattan Project, when time pressures were extremely powerful and many problems had to be solved "on the run" as they arose, shaped his approach to his post-war endeavours. We shall also note how Oliphant displayed some of the characteristic traits and methods of operation which would manifest themselves in the great venture. Of course it was not only Oliphant who was strongly influenced in his post-war work by the wartime environment, and its immediate aftermath, even in the narrow field of particle physics. All three of the initial attempts to build proton synchrotrons had at least some of their roots in the war period. Certainly all three were funded by

160

agencies primarily concerned with the military and civilian applications of nuclear energy.

In summary, we will be examining the war period for evidence of three things: how it provided the context for the genesis of the proton synchrotron, how it added to Oliphant’s stock of “capital” and how it revealed key aspects of his character.

Within a few weeks of the outbreak of war in Europe on 3 September 1939, Oliphant found himself deeply, almost exclusively, involved in research with powerful implications for the defence of the nation, as indeed did a great many physicists and other researchers295. The completion of the cyclotron, over which he and his team had laboured for so many months, was set aside, and there was considerable uncertainty as to when, or even whether, it could be taken up again.

Over the next few years, indeed through to 1945, two lines of investigation would particularly concern him, though the degree of his involvement would vary with time. One of these involved attempts to detect the approach of enemy forces, particularly aircraft, by sending out pulsed beams of radio waves and sensing the returning echoes. Initially known as RDF ("radio direction finding")296, this technique was later universally known as "radar" (for "radio direction and ranging"), a term

295 For general background on the role of science, notably physics, in the course of World War II, an early reference is Crowther and Whiddingon (1947). A more personal account is provided by Jones (1978). More recent analyses include Moss and Hooijmaijers (2000), especially the chapter by Walker on the mobilization of science and science-based technology, Edgerton (2006), who argues a revisionist position that Britain from 1920 to 1970 is better described as a “warfare state” than the much more commonly used description as a “welfare state”, and the relevant entries in the Oxford Companion to World War Two (Foot (2004). For an overview of Oliphant’s role in research with military objectives, see Cockburn and Ellyard (1981), Chapters Seven to Nine. 296 It has been suggested that use of the term RDF was a cover for the real capability of the system, which was more to detect the presence of attacking aircraft and to estimate their distances and numbers. Early versions of the system were poor at delineating direction. 161

invented by the Americans. In the United Kingdom such research had been under way since 1935297.

The other line of work was nearer to Oliphant’s interests through the 1930s, namely nuclear physics. Could the energy latent within atomic nuclei be released on command, perhaps catastrophically, and therefore used as a weapon of war? Oliphant’s interests in both these matters predated the war by up to a year. The first evidence of the phenomenon of nuclear fission had emerged from Germany late in 1938, around the time that Oliphant had been in California to confer with Lawrence and his colleagues about building a cyclotron in Birmingham. Within a few weeks the reality of nuclear fission had been confirmed in laboratories in a number of countries, and speculation was beginning as to the circumstances under which the associated energy could be released. More than six years later, the practical outcome of those speculations would be strikingly demonstrated.

The hunt for

The outbreak of war in Europe in September 1939 caused a massive disruption to the research Oliphant had planned to do in Birmingham, but the break was not totally unexpected. It had been looming for more than a year, as the international situation deteriorated and demands began to grow on physicists like Oliphant to become involved in defence-related research. Late in 1938, Oliphant and many others, most of them with Cavendish connections, had been initiated into the secrets of RDF. This vital move appears to have followed from a meeting of a government committee in November of that year at which the possibility

297 For general background on the development of RDF in the UK see Crowther and Whittington (1947), Rowe (1948), Watson-Watt (1957). Watson-Watt (1959), Price (1979), Bowen (1987). A broader perspective is provided by Brown (1999). 162

of drawing university and other scientists into radar research was first canvassed298.

A significant technical challenge had been identified, namely the urgent need for generators of radio waves of much shorter wavelength than those then in use. Following early investigations, a network of radio transmitting and receiving stations (the “” or CH) was under construction around the south and east coasts of the UK. RDF would use the detection of reflected radio waves to give 20 minutes warning of the approach of aircraft, giving time for an appropriate response to be organised. In this way it was hoped to provide some defence against air attack, giving the lie to the widespread gloomy assertion that “the bomber will always get through”.

The limitations of the system were clear. Existing technology could provide high powered transmitters and adequately sensitive receivers only at long wavelengths, 10 metres or more299. This gave a good range but poor resolution, limiting measurement of the altitude of the attackers and their numbers. The long wavelengths also necessitated huge antennas, up to 100 metres high, easy to identify and attack; the associated equipment weighed tonnes. There was no way, for example, that such equipment could be carried in an aircraft, as was looking increasingly desirable. The answer appeared to lie in the development of radio transmitters and receivers that could deliver adequate power and sensitivity at wavelengths of a metre or even less. By early 1939, a target wavelength of 10 cm had been set, mostly to meet the needs of airborne radar. The Admiralty took on management of the hunt for this

298 The process whereby many hundreds of scientists and engineers would be recruited into the war effort over the following years was managed by Oliphant's former Cavendish colleague (and incipient novelist) C. P. Snow. 299 The shortest wavelength to be used with any success was 1.5 meters, employed in early air- borne radar sets, and making use of technology already developed for early television transmissions. See Bowen (1987). 163

desirable new technology, working with the other services through a joint “Committee on Valve Development” (CVD).

It was in this context that Oliphant wrote to Lawrence after his visit to Berkeley. In his hand-written letter of 11 January 1939, sent from Dunning’s laboratory at Columbia, we read

I find a letter from the Defence Department in England, asking whether I would inquire about the generation of large powers of very short radio-waves. As I think I told you, these are used to obtain reflected waves from aircraft and about 1 metre or less is suitable. I believe the “rhumbatron” has been developed for this purpose. I am familiar with the general principles only and cannot find how it is excited. I shall be grateful if you know this and can pass it on without betraying a confidence, if you will tell me how it is done. Some kilowatts of energy are required to obtain good “pictures” of enemy aircraft by the method of reflected waves.300 Oliphant later recalled the origins of his involvement in development in a letter301 to Rowe.

We started on centimetre wave development in Birmingham as a result of deep appreciation of the necessity of shorter wavelengths which came from our contacts with the “chain” stations and discussions with Tizard, Cockcroft and Bowen and with you at Bawdsey302. We felt that our efforts in the nuclear energy field were too long term to satisfy us in the crucial early years in the war. The scales were tipped by Tizard, who emphasised the importance of R.D.F., and asked me to

300 Oliphant to Lawrence, 11 January 1939. LPBL. Some might wonder why Oliphant would send a note on such a clearly sensitive matter through the ordinary mails. 301 Oliphant to Rowe, 4 April 1983. EBA. 302 A manor house on the east coast of England which served as the headquarters for RDF work from 1936 till the outbreak of the war. 164

undertake the work, and by Wright, who offered to provide the necessary funds and facilities303.

Over the next few months, Oliphant obtained information, from Lawrence and elsewhere, about a device called the “”, invented in California by the Varian brothers and which seemed to meet some of the requirements. With the important assistance of Sayers304, who was “the only one in my group who had any knowledge or experience of radio at all”, he started a program of research to improve the technology and to make it operational in a radar set, devoting a great amount of time and effort to the task. Other teams pursued rival possibilities. In early May he reported to the Faculty that he had been asked by the Air Ministry to carry out “a piece of research of a highly confidential character in association with the Electrical Engineering Department. The Air Ministry was prepared to make a grant towards the expenses of the work”305. This was a clear if veiled reference to the quest for microwaves. Oliphant reportedly made (unreported) suggestions how

303 This quotation introduces some of the leading figures of the British radar effort. Some biographical details follow. AP Rowe (1898-1976). British scientist and Air Ministry bureaucrat at the time of the first proposals for RDF. In the later years of the war he headed much of the radar R and D effort from a research centre at Swanage in SW England. After the war he served as Vice Chancellor of the University of Adelaide. See his entry by Hugh Streeton in the Australian Dictionary of Biography. http://adb.anu.edu.au/biography/rowe-albert-percival-11572. See also Rowe (1948). Sir Henry Tizard (1885-1958). British scientist and bureaucrat, highly influential in defence matters. He was an early supporter of research into RDF. At this time he was scientific advisor to the Air Ministry, and chaired a committee concerned with the Scientific Survey of Air Defence (Rowe was the first Secretary of the CSSAD). He was later of assistance to Oliphant in the early stages of the Birmingham proton synchrotron. See Farren and Jones (1961), Clarke (1965 ) EG “Taffy” Bowen (1901-1991). British physicist, one of the pioneers of radar development in the UK, especially for airborne applications. He had joined the fledgling RDF effort in 1935. See Hanbury Brown et al (1992), Bowen (1987). Sir Charles Wright (1887-1975). British physicist and Admiralty official who chaired the Committee on Valve Development. Wright had been a member of Scott’s expedition to Antarctica 1910-1913. See http://en.wikipedia.org/wiki/C._S._Wright 304 James Sayers (1912-1993) British physicist. An early appointee to the University of Birmingham physics department by Oliphant. He later became Professor of Electron Physics at the University. http://en.wikipedia.org/wiki/James_Sayers_(physicist) 305 Minutes of meeting of the Faculty of Science, Volume 7, p 143, 8 May 1939. BUA. 165

the funds might best be used. The Faculty agreed to approve this case only, with any others to be treated on their merits.

By the outbreak of the war, Birmingham had entered into a contract with the Admiralty covering this work. As the dismemberment of Poland proceeded, the almost complete Nuffield Laboratory became a secret area called the Admiralty Laboratory; staff members were given reduced teaching loads or released altogether for defence work306; Oliphant’s laboratory was now on a war footing. Normal business would not be resumed for more than five years.

A few months earlier, the takeover of Czechoslovakia by Germany being completed and war apparently drawing very near, teams of university physicists had been organised to spend time at the various stations of the Chain Home, to study their operation and perhaps to suggest improvements. Oliphant’s team went to Ventnor on the Isle of Wight307, only a few weeks before the war began. It was here that the physicist John Randall and the research student first began to work together, forming a team whose efforts would bear such spectacular fruit six months later.

The hunt for “the bomb”

RDF was not the only defence-related matter of concern to Oliphant and his laboratory through 1939, diverting their attention from the cyclotron. Less than a year earlier, in research that could trace its ancestry back to Chadwick’s 1932 identification of the neutron, the German researchers

306 Moon and Ibbs (1981), p. 20. 307 This was the task to which Oliphant had referred cryptically in his 24 August 1939 letter to Lawrence. Surviving correspondence shows that Oliphant had made strenuous efforts to secure the presence at this exercise of Moon, then “on loan” to GP Thomson at Imperial College to help with early (and unsuccessful) experiments on chain reactions in natural uranium. Oliphant believed that Moon’s skills and experience would be of great value. It is not clear from the record if he succeeded. 166

Otto Hahn and Hans Strassman had discovered the phenomenon of “nuclear fission”. When bombarded by neutrons, the nuclei of some heavy elements, most notably uranium, could be split in two, with the release of energy and of neutrons that could, under the right circumstances, cause other nuclei to split308.

The growth of Oliphant’s interest in this, and his understanding of the implications, can be traced through his correspondence with Lawrence. In a letter dated 21 March, we read

I have been very interested in the reports which have come from America and elsewhere concerning the fission of uranium under neutron bombardment. It is very likely that when really large energy particles are available from the cyclotron, such fission processes may be observed more frequently. I have hopes that there is a considerable future in this direction for work with our new apparatus309.

Within a few months the emphasis had changed from pure science to potential applications.

We are very much interested at the present time in the problem of nuclear fission, and the possibilities that it may at some time prove a source of power, or of explosion. I am quite sure these possibilities are very remote but the Defence authorities here feel that they must be absolutely certain that no possibility is

308 This discovery had arisen from work through the 1930s in which elements bombarded with neutrons were transmuted and made radioactive. For general background on the discovery of fission and its implications, See Dahl (2002), Yruna (2009), Kragh (1999) Chapter Eighteen. 309 Oliphant to Lawrence, 21 March 1939. LPBL. 167

overlooked in this direction, as there are rumours that great developments have taken place along this line in Germany310.

The matter had already reached beyond the scientific community and into the public arena. On 30 April the Sunday Express 311carried a report filed from “somewhere in England” under the blazing multiple headline Scientists Make an Amazing Discovery; Stumble on a Power “Too Great to Trust Humanity With”; A Whole Country Might be Wiped Out in a Second. The journalist briefly (and with a general accuracy) recounted the discovery of fission and the possibility of a that could liberate unprecedented amounts of energy. “All the other scientific discoveries ever made”, the article declared, “would be dwarfed by this discovery”.

Three British laboratories were named as sites of research into these possibilities; Cambridge, Liverpool and Birmingham, presumably because they had, or would soon have, cyclotrons. The local press was soon onto the story. Two days later the North Mail published in Newcastle312 carried a story headlined Students Make an Atom Splitter Which May Create Energy. Oliphant was quoted as asserting the cyclotron he was building was “not a practical machine from an economic point of view, but that it was of great value to science”, with a role in physical and medical research work, such as the treatment of cancer. The Birmingham Post at the same time313 quoted an unnamed source, apparently linked to the university, denying ”exaggerated and premature reports” of research into the splitting of atoms.

310 Oliphant to Lawrence, 30 May 1939. LPBL. 311 Sunday Express, 30 April 1939. Clipping in EBA. 312 North Mail, 2 May 1939. Clipping in EBA. 313 Birmingham Post, 3 May 1939. Clipping in EBA. 168

It is true that scientists in various parts of the world are today taking a great interest in experiments involving uranium, but reports that a discovery had been made “too great to trust humanity with” and capable of “wiping another nation off the face of the earth in a second” are described as gross exaggerations “without any real foundation”. Experiments that are being made are purely academic and no practical application is envisaged.

Clearly the source felt unable to admit to the true state of affairs, unlike Oliphant when writing to Lawrence shortly afterwards.

As regards uranium we haven’t been doing much in the past months but from what I read in the literature, there does seem to be evidence that more than one neutron on the average comes off in fission, and so it should be quite possible to produce a chain reaction with a large enough amount of uranium. There is some question as to whether the isotope involved is 235 or 238 etc, and it might be necessary to separate 235 in order to really bring this about. In any case it is the sort of thing that would be subject of wild rumours, and I can well imagine your defence authorities not being willing to overlook any possibilities in this direction. Indeed this sort of thing is another reason why the British government should come forward with generous support for nuclear physics in Great Britain314.

A few months later again, research on the matter was well underway in Birmingham

We have been carrying out some further work on the fission of uranium and other elements, but the results of our experiments

314 Lawrence to Oliphant, 15 June 1939. LPBL. 169

are such that I feel very great doubt now whether it can ever be made to produce a chain reaction, unless the isotopes of the element can be wholly or partially separated. To that end we are building in the laboratory an apparatus for separating the isotopes by the thermal diffusion method, using uranium hexafluoride. I understand from the Cambridge people that they have obtained fission of uranium using projectiles from the cyclotron. The effect seems to be fairly conclusive315.

The way had been opened for Oliphant’s crucial engagement in the exploration of nuclear fission, and its applications, once hostilities had begun.

Birmingham’s twin breakthroughs

Developments of major significance in both of these lines of work took place in the spring of 1940, within a few weeks of each other and both in the Birmingham laboratory. 21 February saw the first firing-up of a new form of generator of microwaves, soon dubbed the "”316. This was the work of John Randall and Harry Boot, first brought together as a team during the Ventnor exercise. They had worked from first principles to develop a device that far surpassed the capacity of existing forms of magnetron. Machined from a block of copper, continuously pumped and cooled by running water, the experimental rig generated hundreds of watts of power at the predicted, and much sought after, wavelength of 10 cm317. Within a few months,

315Oliphant to Lawrence, 24 August 1939. LPBL. 316 Burcham and Shearman (1990). See also Oliphant. Address at the 50 Year Magnetron Symposium, 21 February 1990. Box 22/7, SCUB. See also Cockburn and Oliphant (1981), p 85-86, 317 This target wavelength had been set with the needs of air-borne radar in mind, for example in night fighters. It took account of the size of the largest parabolic dish that could be fitted into the nose of a fighter and the need to make the resulting beam of “microwaves” narrow enough 170

engineers at the firm of General Electric at Wembley (where Randall had previously worked) had found ways to seal off the valve and cool it with air, so dispensing with the continuous pumping and water cooling that would have been impractical in operational use.

Oliphant later recalled the origins of the triumph in the following terms.

Randall and Boot joined in the centimetre wave work a little later than others and were assigned the task of building resonators into magnetrons318 in accordance with our general philosophy. Charles Wright was keen on the magnetron approach and agreed readily to the rather small expenses involved. The immediate success of the concept was due to the extraordinary extrapolation by Randall of the properties of the resonant circle of wire as used by Hertz, and the outstanding gifts of Duke as a master of vacuum and valve technologies and a designer of rare insight319.

The significance of this development for the outcome of the war has been much discussed. So influential were the varied applications of the magnetron in many theatres of the conflict that microwave radar has been widely described as "a war-winning weapon". Following the sharing of the new technology with the Americans as a consequence of the Tizard Mission320 in the autumn of 1940, it was dubbed by Churchill as “the most valuable cargo ever to cross the Atlantic”321.

to provide an acceptable direction-finding capacity. The calculations were made largely by Bowen. 318 Magnetrons had existed prior to this work, but they were low power devices producing only a few watts. 319 Oliphant to Rowe, 4 April 1983. EPA. 320 ”The ” is the common name for a British technical mission to the United States in August 1940, headed by Sir Henry Tizard, scientific advisor to the Air Ministry. It included, inter alia, John Cockcroft and EG Bowen. Its purpose was to exchange information on various technical developments of military significance. Those included the cavity magnetron, a sample of which was carried across the Atlantic by Bowen. 321 See Brown (1999), Chapter 4, Section 4.1. 171

Within a few weeks of the magnetron triumph came the spur to the other line of defence research which would so occupy Oliphant in the coming years, the quest to harness the energy of the atomic nucleus. The breakthrough again depended on teamwork. Rudolph Peierls and Otto Frisch322 were both German émigrés, Frisch recently arrived, Peierls on the staff of the University for several years and already a British citizen, but their backgrounds excluded both of them from the top-secret radar research for security reasons. Nuclear power research was not so secret, nor was it central to the work of the laboratory at that time. So they were free to pursue it.

The outcome of their work was to gain enduring fame as the “Frisch- Peierls Memorandum”323. In half a dozen pages they explained that an explosive chain reaction in uranium was possible if the fissile but rare U235 isotope could be separated from the much more abundant but non-fissile U238. Frisch had been working on one approach to such separation, which depended on the only usable difference between the two isotopes, their less than 2% difference in mass. The task was likely to be very challenging. Furthermore the calculations enabled them to estimate the amount of needed to produce a runaway chain reaction. This "critical mass" was likely to be of the order of only a few kilograms, not the tonnes which earlier work had suggested. If the reasoning and the calculations were sound, a deliverable "atomic bomb" was feasible.

Once he was aware of the memorandum and its significance, Oliphant was able to use his defence contacts, particularly those with Tizard, to

322 Otto Frisch (1904-1979). Austrian physicist. Jewish-born Frisch had been invited by Oliphant to come to Birmingham at the outbreak of the war. He was a nephew of the physicist Lise Meitner, who had contributed to the discovery of nuclear fission. For background see Peierls (1981), Frisch (1979). 323 Frisch (1979), p. 125-127, Peierls (1985), p. 153-156, Cockburn and Ellyard (1981), p 95. 172

generate action on the memorandum, something which Frisch and Peierls would have been unable to do unaided. He wrote to Tizard

I have considered these suggestions in some detail and have had considerable discussion with the authors, with the result that I am convinced the whole thing must be taken rather seriously, if only to make sure the other side is not occupied with the production of such a bomb at the present time…. I hope you will not think of this as purely a hare-brained scheme. It may well turn out to be impractical but in any case it is put forward with sincerity by the authors and with considerable belief by myself324.

Tizard's response was perhaps typically bureaucratic.

What I would like would be to have a small committee to sit soon to advise what ought to be done, who should do it and where it should be done, and I suggest that you, Thomson and, say, Blackett would form a sufficient nucleus for such a committee, and if you would like to bring someone else please make a suggestion325.

Meeting for the first time early in April, that small group later took shape as the , which over the following year guided a series of investigations, leading to the conclusion that the argument of Frisch and Peierls was indeed sound and that given an appropriate process for enriching uranium, a bomb based on nuclear fission could be built. In addition, the weapon would mostly likely be small enough to be delivered to a target by an aircraft. Oliphant was initially a member of the MAUD Committee and contributed strongly to its deliberations. A later

324 Clark (1965), p. 218. 325 Ibid. 173

re-organisation of the Committee saw him involved only as a member of a subsidiary technical committee. The government having decided that each university should have only one representative on the Maud Committee itself, Oliphant's position had been taken by his colleague Norman Haworth, the Professor of Chemistry who was already engaged in research relevant to the enrichment of uranium.

Following the delivery of the final report of the Maud Committee in the summer of 1941, the British nuclear energy effort was profoundly reshaped. Control passed to the DSIR, by then headed by Appleton, and a new division of “” established to oversee R and D in the field. Chosen to head the work was Wallace Akers, Director of Research at the giant chemical firm Imperial Chemical Industries. ICI was already active in relevant research, such as converting uranium into a gas for use in diffusion. Oliphant did not care for these developments and let out what Gowing called “a full-throated protest”. “...I see no reason whatever why the people put in charge of this work should be commercial representatives completely ignorant of the essential nuclear physics on which the whole thing is based…. This organisation is tantamount to that which exists in the United States where the whole thing is in the hands of non-nuclear physicists and is therefore being badly mismanaged” 326. This development was to colour Oliphant’s relationship with Akers for some years to come, a relationship that would be of importance in the hunt for funds for the Birmingham accelerator.

Oliphant in America 1941

Through 1940 and 1941 Oliphant was much involved in wide-ranging radar issues through his membership of the RDF Applications

326 Gowing (1964), p. 110. Postwar correspondence between Oliphant and Chadwick showed that the latter shared some of these misgivings though he did not say so at the time. Correspondence in EBA. 174

Committee327, but his personal preoccupation was largely with the klystron, refining the technology, seeking to increase its power output and using it to energise a makeshift radar set built on a sound locator carriage of Great War vintage. With a pair of dishes to transmit and receive the shortwave signals, the whole apparatus could be swung round to track a target. Its usefulness was demonstrated in trials both at the University and at the radar development centre at Swanage on the southwest coast, where according to some recollections it was destroyed in an air raid.

However, the magnetron was beginning to prove more promising than the klystron; elements of the structure of the latter appeared to limit the potential power output. A program of continuous development of the magnetron over the months following its first demonstration, involving both university researchers and industry, greatly increased its power output and reliability. Some challenges had to be overcome. One was the tendency of the magnetron to jump from one frequency of oscillation to another. A solution was devised at Birmingham by James Sayers, who found that "strapping", namely electrically connecting one segment of the device to another, eliminated the problem. It was to inform the Americans about this development that Oliphant visited the United States in August and September 1941, arriving with a full programme of visits to research groups and firms involved in magnetron development.

He reached the United States around 5 August 1941, having made an eventful flight across the Atlantic in a B-24 Liberator bomber, and planning, as he told Lawrence in a letter on that date328, to stay for about five weeks. He hoped for a chance to visit Berkeley. He could

327 For example, a surviving memorandum by Oliphant from July 1941 canvassed the feasibility of keeping a watch on the enemy-held coastline using beams of 3000 watt microwaves. G322 13-15, Cherwell Papers, Nuffield College. 328 Oliphant to Lawrence, 5 August 1941. LPBL. 175

argue that such a visit was necessary to allow him to talk to firms on the west coast regarding the production of microwaves but really it would be "to see your latest work in Berkeley". Replying on 12 August329, Lawrence was "pleased beyond words" at the news.

The surviving documentary record indicates that Oliphant visited Berkeley on 22 September 1941. While there he was able to view (and to be photographed with Lawrence in front of) the skeleton of the huge new 184 inch cyclotron, still standing in the open air since its containment building had not yet been constructed, and to discuss the experimental work that Lawrence planned to undertake330. Talk quickly turned to uranium fission and the potential for its military use. Reports from the Maud Committee had been forwarded to Lyman Briggs, director of the National Standards Laboratory and Chairman of the Uranium Committee supposedly investigating such possibilities. Oliphant had met Briggs and was shocked to discover that the reports were mostly locked in Briggs’ safe and had not been distributed to those who should have been in the know.

The impact of Oliphant's visit was illuminated by a memoir written by Lawrence in 1945. During his few days at Berkeley, Oliphant had provided Lawrence with some "very powerful ammunition", with which he [Lawrence] could pursue the case for America to take seriously the possibility of basing a military weapon on nuclear fission. Though not a member of the Maud Committee at that time, Oliphant was able to supply Lawrence with a summary of its latest findings. Lawrence quickly realised that the "Briggs committee was not working properly” and so began a personal crusade to inform and enthuse "certain people".

329 Lawrence to Oliphant, 12 August 1941. LPBL. 330 For a report on this visit see Oliphant (1941). 176

The influence of Oliphant on Lawrence, and through him on the gestation of the Manhattan Project, has been summarised by Gowing331. Lawrence was “deeply impressed” by the reports Oliphant gave of the work of the Maud Committee, and proceeded to inform and enthuse others, such as James Conant, who was deputy to at the newly-formed Office of Scientific Research and Development (OSRD), and Karl Compton from the , who was influential in the Academy of Science. Lawrence had been trying to stir action for some months and was “much encouraged in his proposals by Mark Oliphant”332. After noting that the British thought a visit later made by the US researchers Pegram and Urey to the UK has been a critical factor, “the decisive influences were those of the Maud Report itself on Dr Bush and Dr Conant and of Professor Oliphant on Lawrence”333.

The report Oliphant provided to Lawrence survives on the public record334. He stated that, in the opinion of the Maud Committee, the possibility that "a bomb of super-explosive violence" could be produced by the fission of U235 had been "demonstrated sufficiently clearly for it to be necessary to put an adequately high-powered effort into the problem”. Calculations suggested that 20 kg of the could be detonated with a force equivalent to one thousand tons of TNT, enough to "devastate an area of the order of magnitude of 1 mile in diameter". As for the technology to undertake the needed separation of the isotopes, only gaseous diffusion was thought by the British to be sufficiently safe and reliable. A plant to produce 1 kg of U235 a day would cost some $50 million to build and to operate for a year, an enterprise beyond the capacity of the UK, given the progress of the war

331 Gowing (1964), p 116. 332 Gowing (1964), p188n. Gowing here was quoting from “The New World”, p. 44. 333 Gowing (1964), p. 121. 334 Oliphant ML. Uranium Fission in Britain. Undated and unsigned memorandum. LPBL. 177

at the time. Hence the need for the challenge to be taken up strongly in the United States or in Canada.

In his covering letter to the report, dated 24 September335, Oliphant commented that "I feel quite sure that in your hands the uranium question would receive proper and complete consideration, and I do hope that you will be able to do something in the matter". That hope was quickly realised. Indeed, as we have noted, the initiation of the developments that led to the Manhattan Project a year later can plausibly be dated from that time336. The visit was good for Oliphant too. It strengthened his relationship with Lawrence and brought him to the notice (and in some cases into contact with) other leading figures in the scientific/military establishment337.

During the same visit to the United States, Oliphant met Richard Casey, the Australian Ambassador to the United States, discussing with him the work on RDF that had brought him to the US, and more tentatively, the recent developments in uranium research and their implications. Professing ignorance of the matter, Casey asked Oliphant for a brief memorandum on the subject. This was quickly provided. In it, Oliphant discussed the feasibility of both a "uranium bomb" and a "uranium energy machine”, with a strong emphasis, given the Australian context, on the latter. He envisioned such energy sources as being of "the greatest possible importance to Australia…. I am confident…. that

335 Oliphant to Lawrence, 24 September 1941. LPBL. 336 The development of the Manhattan Project has been well recounted by numerous authors. In addition to Gowing, who covered the matter from a British perspective, I have made use of Brown (1977), which reproduced the official US history of the endeavor (“The ”) and Rhodes (1987). Groves (1962) provided an overview of the project by the leader of the endeavour. 337 Gowing noted that his meetings during this visit including one with the Italian-born , who would later lead the team that built the first working . At the time of Oliphant’s visit, Fermi was reportedly “non-committal about the fast neutron bomb” and the Bohr-Wheeler theory of fission on which it was based. Gowing (1964), p. 118n. 178

Australian uranium will prove as valuable to the country as oil wells have been to America338,”

Oliphant in Australia 1942

On 7 December 1941, the attack by Japanese forces on American naval and air assets at Pearl Harbour in the Hawaiian Islands immediately precipitated the United States into conflict against Japan, and within a few days into the wider war in Europe. Over the following months, Japanese forces advanced rapidly through South-East Asia.

Oliphant’s response to the fall of Singapore to Japanese forces on 15 February 1942, even before the first bombing attacks on Darwin four days later, was typical of the man. It combined his sensitivity to the call of duty with an impatience which could have him acting precipitately. Communicating through the High Commission in London he at once offered his services to the Australian government, believing that his expertise and experience in the development of radar would be of value in the preparation of the nation’s defenses against enemy attack. Oliphant, also perhaps typically, may have wanted to be close to the centre of the action. A trip home would also reunite him with his wife and children who had been sent back to Australia for safety at the outbreak of war.

Oliphant’s cable offering his services survives in the public record. “Offer services defence Australia stop will seek release Admiralty and fly out immediately if real job offered stop Australia’s best hope conservation

338 Casey to Rivett, 17 September 1941. Australian Archives CRS A3300, item {217}. Casey had met Oliphant at a dinner party given by British scientist George Darwin in late August. Casey commented that "[Oliphant] seems to be a man of some note. Darwin speaks of him with great respect." 179

forces by new methods stop”339. The new methods were presumably those of microwave radar.

The gesture resulted in Oliphant being absent from Birmingham for nearly a year, departing on 20 March 1942 and not returning (accompanied by his family now that the threat of invasion had eased) until the end of February 1943. Journeys in both directions were by ship, whose passage was drastically slowed by the demands of blackout, convoying and zigzagging to avoid submarines. As a result, less than half of the elapsed eleven months were spent productively in Australia. Nor was the visit uniformly productive. It appeared that leading figures in radar in Australia340 were less than enthusiastic about his intervention, perhaps seeing it as a challenge to what they were already doing. That limited his opportunities to make a contribution on technical matters, let alone on policy. He was able to be more useful in , where he helped set up a new laboratory for valve development at the university, than in , which was nominally the headquarters for radar development.

When the time for his departure finally came, White341 and Madsen professed themselves satisfied with the outcome of the visit, even if Oliphant was not. Rivett342 called it “very productive” and proposed that, on his return to Britain, Oliphant should stay in touch by becoming an adviser on radar matters, working through the Scientific Liaison Office in London for an annual retainer of £250. Oliphant agreed. By now he was itching to be away. He longed to be back in the thick of things343.

339 Quoted in Masden to Rivett, 17 February 1942. CSIRO Archives, Series 3, File PH/OLI/8. 340 For example , Professor of Electrical Engineering at the , who had been very influential in establishing radar development in Australia. 341 Frederick White, in charge of the Australian radar R and D effort. This was located under the cloak of research into “radio-physics”, undertaken within the publicly-funded Council for Scientific and Industrial Research (CSIR). 342 , Executive Officer of the CSIR. 343 Cockburn and Ellyard (1981), p. 93. 180

Oliphant returns to nuclear physics344

On reaching Birmingham, Oliphant found that things had moved on. The cavity magnetron, invented in his laboratory three years before, was now in mass production in a variety of forms and for a multitude of uses. It was already powering AI ("air interception") sets in night fighters to ward off the threat of enemy bombers sneaking through the dark. It was being readied for use in ("Home Sweet Home”) sets to guide bombers venturing over Europe, and in ASV ("air to surface vessel") sets to be taken into the Atlantic in search of U-boats. Microwave radar, powered by the magnetron, was beginning to turn the tide of battle in many spheres; the Battle of the Atlantic would be won before winter by the use of this powerful technology. It was clear that the work of researchers in this field seemed mostly done. The task was now in the hands of the production engineers in firms like GEC and BTH to turn the tubes out by the thousand, ready for reliable use. It was time to look for a new challenge.

The new challenge was an old one, how to get a chain reaction going in uranium and so power an atomic bomb. As Frisch and Peierls had shown three years before, the secret was separating the isotopes of uranium, a challenging task. The most promising method appeared to be "gaseous diffusion", in which the uranium metal was turned into a gas (uranium hexafluoride) and then pumped through a series of very fine sieves which would discriminate between the two isotopes on the basis of their small difference in mass. With an equal energy of motion, the heavier isotope will travel more slowly and arrive at the sieve later.

344 The involvement of British researchers, including Oliphant, in the development of nuclear weapons is well covered in Gowing (1964). Additional and contextual information relevant to the rest of this chapter can be found in Cockburn and Ellyard (1981) Chapter Nine. 181

The process was slow and would need to be repeated many thousands of times to get any substantial enrichment.

Oliphant was ready to argue for a different process, one which would work both better and faster, perhaps requiring just a single stage for complete separation. The uranium atoms would be given an electric charge, formed into a beam and passed through a magnetic field which would bend the paths of the ions. The paths of the heavier U238 ions would be bent less and so become disentangled from the lighter U235. The principle was like a cream separator; the whirling cones fling off the heavier milk molecules and leave the lighter cream collected in the centre. It was the way Oliphant had separated the isotopes of potassium at the Cavendish a decade before to find out which one was radioactive and later the isotopes of lithium to distinguish their different responses to proton bombardment345. Here lay the possibility of linking peace-time academic research with the urgent needs of the war effort.

On 26 May 1943, less than three months after his return from Australia, Oliphant wrote to his old Cavendish colleague Appleton346, enclosing some “thoughts on ”. As head of DSIR, Appleton had oversight of the British atomic energy program. Oliphant wrote “I believe that such considerations might remove the whole project from the realms of gigantic chemical engineering and render it a much more practical proposition within this country itself”. The "gigantic chemical engineering" obviously referred to the many stages needed to separate the isotopes of uranium by gaseous diffusion. The motivation for the communication was unstated in the memo and in the covering letter, but was obvious enough. The goal was to separate U235 from U238 and so

345 Oliphant, Shire and Crowther (1934). 346 Oliphant to Appleton, 26 May 1943. AB 1/255. United Kingdom National Archives [hereafter cited as “UKNA”]. 182

produce a . Aware of security concerns, Oliphant referred to uranium as “X”.

Oliphant was proposing, as Lawrence had already done before him, that an electromagnetic (EM) process was a feasible way to bring about the needed separation. He noted that, according to reports, Lawrence had found a solution to the "space charge problem" which caused beams of positively-charged ions to spread out as they moved as a result of mutual repulsion. It was argued that a small amount of gas in the beam chamber at a pressure too low to interfere with the beam would be ionized by the beam, releasing electrons which would counteract the positive space charge. This would allow the production of much more intense ion beams than previously thought possible.

Some quick figuring, undertaken in his typically optimistic manner, suggested to Oliphant that the "efficient use" of the Nuffield cyclotron could handle a total ion current of 1 ampere. That would mean separating three or four grams of uranium an hour or a kilogram in a fortnight. "The product would be obtained in the single operation in a state of high purity." About 100 kilowatts of electrical power would be required to drive the process. Oliphant proposed the use of uranium metal directly rather than in a molecular form to reduce contamination.

Appleton passed Oliphant's correspondence on to Wallace Akers, in charge of the TA project within DSIR. Akers replied to Oliphant on 2 June347. After thanking Oliphant for his ideas on "methods of separating the isotopes in which we are interested”, Akers reported that the possibility of doing some work on electromagnetic separation methods (and "not to leave the whole field to Lawrence") had been raised by Chadwick at a meeting of the TA Technical Committee in

347 Akers to Oliphant, 2 June 1943. AB 1/255. UKNA. 183

March. Oliphant's timing was therefore very apposite. The Committee had left the matter for later consideration but had agreed that "the only condition under which this work could be effectively done in this country would be provided you [Oliphant] were available to direct the work. At that time it seemed unlikely we could hope to get you full-time on such work."

On the day of writing, Akers had discussed Oliphant’s memo with Chadwick, Peierls, Darwin and Simon348. They had noted that up-to-date information on Lawrence's progress was not available, due to the secrecy imposed by the Americans since the Army had taken over the project, but it was known that the Berkeley team had tried methods similar to the ones Oliphant was suggesting; the degree of success was unknown. "… it is clear to us that the whole problem is so novel that there would be a good chance of your evolving some technique, which has escaped them, which would have the effect of making a very difficult process into a really practicable one." While electromagnetic separation may not be able to handle the whole task, it would "provide an extremely elegant solution for the final separation after concentration of the active isotope had been brought up to say 50% by other methods".

The way forward, Akers suggested, was for Oliphant and his colleague Moon to work full-time on the challenge. If that was possible, Oliphant would, of course, become a member of the Technical Committee. Chadwick would be in touch soon to discuss the possibility, Akers himself being about to leave for Canada. In closing, Akers commented that he hoped good relations would again be established with the Americans,

348 (1893-1956). German-born physical chemist. Emigrated to Oxford in 1933. Very influential in the development of the gaseous diffusion process for the separation of uranium isotopes. For background see Kurti (1958). 184

... in which case it would be a very desirable and useful thing for us to have a section working in parallel with Lawrence. I think also this would have a good effect on the latter, as I know that he personally would welcome an opportunity of cooperation with you.

Related correspondence reveals some of the currents swirling around this development. Akers forwarded a copy of his Oliphant reply to Appleton, with some covering comments which are worth quoting at length349.

I may say that we were all quite definite that Oliphant whole-time would be a most valuable addition to our effort, but if he proposes to work part-time on this, then we should certainly not want him, as such an arrangement would be ineffective and dangerous. I say the latter, because Oliphant is, as we know, impetuous and none too discreet, so that we would not want to let him in on all the secrets of our TA work unless he were properly tied to us.

I believe that it is actually rather fortunate that I am going away so that the discussions with Oliphant, in the first place, have to be left to Chadwick, as Oliphant is still a little suspicious about the commercial element in TA, though I believe he does not regard Perrin and myself with the same distrust as he did originally350.

I am sure you will agree that the addition of Oliphant and Moon, together with the resources which they would bring, would

349 Akers to Appleton, 2 June 1943. AB1/255. UKNA. 350 This is a reference to Oliphant’s early opposition to the involvement of the firm Imperial Chemical Industries (ICI) at a high level in the British atomic energy program. Akers had been a senior executive with ICI before joining TA, as his TA deputy Perrin had been. 185

considerably strengthen our position, both absolutely and vis-a- vis the Americans.

Oliphant had copied his letter to Appleton to Simon at Oxford. Simon was a leading worker in and advocate of the gaseous diffusion process, which Oliphant was keen to challenge. Simon replied in a letter dated 31 May351, which placed it in advance of his meeting with Akers and the others on 2 June. His thoughts may therefore reflect those of the Technical Committee at its meeting in March, as well as his own.

As might be expected, Simon was not swept away by Oliphant's vision. He pointed out various caveats and limitations, one being that a large- scale electromagnetic plant would require a large number of trained physicists to maintain it, possibly generating a prohibitive cost. This view was reportedly shared by the Americans with whom Simon had discussed the matter in October 1942. Some of the sorting methods Oliphant was proposing had been tried by Lawrence and others without great success; likewise the proposal of using uranium metal rather than the chloride. According to Simon, Lawrence strongly advocated a simple magnetic focusing to separate the isotopes, but Simon commented, in a telling phrase, "that it may be that the greater success of [this method] is chiefly due to the strong personality and drive of Lawrence and the greater means that he is able to concentrate on this.” Was Simon fearing here that Oliphant's similar powers of persuasion might override due diligence?

The scale-up factor also concerned Simon. He claimed that no ion source achieved by the Americans had exceeded 25 milliamps. To reach Oliphant’s hoped-for full ampere of ion current, 40 ion sources

351 Simon to Oliphant, 31 May 1943. CHAD IV, 3/16. Chadwick Papers, Churchill College Archives [hereafter cited as “CPCC”]. 186

would have to be placed within the cyclotron magnet “which seems not very easy but certainly possible". Furthermore, if each "unit" separated 3 or 4 grams per hour, 1500 or 2000 units would be needed for a full- scale plant of the sort being envisaged at present. Might that not be as much of a challenge as building a diffusion plant with a similar number of compressors?

In summary, Simon argued that, at best, the electromagnetic process would be “very useful at the top end of the plant", say from 50% enrichment upwards. If Oliphant was arguing that an EM plant could carry the full load, then the matter needed to be discussed as soon as possible by relevant people. Given the outcome of the meeting on 2 June as reported by Akers, which essentially gave Oliphant the go- ahead without any further review, it does not appear that Simon's opinions carried much weight.

In a letter to Oliphant on 4 June352, Akers proposed that, since both he and Perrin would be on "much needed" leave, Chadwick could approve any expenditure that Oliphant sought to get the research underway. What such expenditure was is not recorded in any surviving document, but Akers anticipated that Chadwick would find it reasonable. Certainly, in the event, those costs were no impediment to moving the transfer forward. More protracted were likely to be the negotiations of the transfer to DSIR of the Birmingham Physics Department workshop, which had been equipped by the Admiralty, along with some staff still directly employed by the Admiralty. Akers wrote to Appleton353 on 4 June asking him to deal with these matters. He also reported Chadwick's view that Oliphant would indeed choose to join the TA effort and that "he has some good ideas and will probably do some things

352 Akers to Oliphant, 4 June 1943. AB 1/42. UKNA. 353 Akers to Appleton, 4 June 1943. AB 1/42. UKNA. 187

better than Lawrence, especially any methods depending on the use of magnetrons, as Oliphant has much more experience with those”.

Though no surviving documentation states so, Oliphant clearly accepted Akers’ invitation. Within 10 days, Chadwick was in Birmingham chairing a meeting (in his role as Acting Chairman of the Technical Committee in the absence of Akers) to discuss the arrangements for the transfer of a substantial number of the Birmingham physics department staff from radar to nuclear energy research. Attending the meeting were Oliphant, Nimmo, Moon and Sayers, together with two officials from DSIR, who not only took the Minutes354 but represented the agency with overall responsibility for TA research.

Oliphant told the meeting that the time had come to wind up the work being done on radar for the Admiralty and to hand ongoing responsibility for that to industry. CS Wright, the Director of Scientific Research from the Admiralty, had accepted this decision. It would take about three months to clear the remaining Admiralty commitments, but research on TA would start at once. The immediate priorities would be completing the cyclotron magnet, left in limbo since the start of the war, and building the necessary additional equipment needed for relevant research. Some research could commence even before the machine was ready.

Eight staff members, including Oliphant, Moon, Nimmo, Sayers, Titterton and Duke would transfer immediately to TA work, along with a number of the workshop staff, and others would follow within a few months. The intention was to move the whole of the laboratory over to the new program; existing equipment belonging to the Admiralty would be taken over by DSIR. The University would continue to be the employer, under

354 Proposed accession of Professor Oliphant and his team of workers to Tube Alloys research. Memorandum of Meeting at Nuffield Laboratory, University of Birmingham, Saturday 23 June 1943. CHAD IV (3/16), CPCC. Also in Box 22X/7B, SCUB. 188

a contract with DSIR. A TA contract had been in place with Birmingham since early 1942 covering the work done on gaseous diffusion under Haworth and Peierls. Under the agreement, the results of any research were to be the property of the Department. Very strict attention was to be paid to security, in view of the top-secret nature of the work. Of course Oliphant and his colleagues were already well aware of those requirements from previous experience with radar research. Arrangements for the transfer would be largely completed by August, and backdated to take effect from 1 July355.

Oliphant to America 1943

The key issue now was how to re-establish cooperation with the Americans, as Akers had hoped would soon occur. After considerable background diplomatic negotiation, the had been framed by British and American officials meeting under the aegis of the Canadian government. Headed Articles of Agreement Governing Collaboration between the Authorities of the USA and the UK in the Matter of Tube Alloys, the document was initialled by Churchill and Roosevelt on 19 August 1943. Seeking to "bring the Tube Alloys project to fruition at the earliest moments", the Agreement stated, inter alia, ”in the field of scientific research and development there shall be full and effective interchange of information and ideas between those in the two countries engaged in the same sections of the field”. This did not constitute the level of exchange previously enjoyed, since it implied that researchers would be entitled only to exchange information in areas of their own particular interest or expertise, but it would be enough, for example, for Oliphant and Lawrence to collaborate on electromagnetic separation.

355 Hogg to Perrin, 10 August 1943. AB 1/42, UKNA. 189

With the way now open for the resumption of collaboration, and with Anderson’s approval, Akers gathered an advance party together in Washington to set such collaboration in motion; it comprised Chadwick, Peierls, Simon and Oliphant. Gowing noted that the party arrived the day the Quebec Agreement was signed “in what appeared to the Americans as indecent haste”356. On 13 September, Chadwick and Oliphant met with US Army Corps of Engineers Brigadier-General , now at the head of the United States atomic bomb program code-named the Manhattan Project, and leading American physicist Robert Oppenheimer, who was in charge of the research base at Los Alamos in New Mexico, code-named site Y, where the bomb would be designed, built and tested. A few days previously, the Military Policy Committee, charged with high-level decision-making including the responsibility for making the Quebec Agreement work, had recorded in its Minutes

It is recommended that every effort be made to secure the services of Dr Chadwick and Dr Oliphant at the special American installation involved in development of the weapon…. it is believed that arrangements for placing Dr Chadwick and Dr Oliphant in our organisation should be made at once357.

A memo358 written by Oliphant on 7 October set out the plans now in place. Oliphant himself would take a small team to Berkeley in the first instance to work with Lawrence and to undertake research into ways to increase the yield of the existing American plant by improvements to the source of ions and the focusing system. The team would comprise Oliphant, Sayers, Duke, and Titterton, all members of the current Birmingham TA team, together with the theoretician (and old Cavendish

356 Gowing (1964), p. 171. 357 Minutes of the Military Policy Committee, 9 September 1943. LPBL. 358 Untitled memorandum by Oliphant, 7 October 1943. AB 1/42. UKNA. 190

colleague) . Both Sayers and Massey were still with an Admiralty team and their release needed to be negotiated. Titterton would go at once to Los Alamos with Chadwick; all or some of the others would follow after six months or so, to "work on problems associated with the bomb itself". In Oliphant's case, at least, that onward move never occurred.

Moon would be left in charge of work in Birmingham with a team of seven including Nimmo and two others, currently Admiralty men. Oliphant planned to make several visits each year, each for a few weeks “to keep in close touch with the work" and to "direct the work here into the most useful channels from the point of view of the joint effort in America and of our ultimate national scheme”.

Oliphant would be well remunerated for his efforts in America359. His full university salary would be topped up by the equivalent of the consulting fee still being paid by the Australian government for his advice on radar. With other allowances, this came to a total of £1950, which, as Appleton360 pointed out, was an "exceptional measure" and "on a purely personal basis". The figure was "not without serious embarrassment to the Department in relation to the general standard in the government service", and, perhaps not surprisingly, was to be kept confidential. Some months later even Chadwick was unaware of what Oliphant was being paid. In addition, he would be entitled to a Class 1 Mission Allowance for sustenance and to first-class travel. These extra-ordinary provisions appear to underline the value that Oliphant was expected to bring to the project.

359 Longair to Oliphant, 11 November 1943. EBA. 360 Appleton to Oliphant, 14 November 1943. EBA. 191

By 21 November, Oliphant was in Washington after "a rotten trip across the Atlantic by air"361. From the headquarters of the British Supply Council in North America he wrote to Chadwick362, who was yet to arrive in the US, reporting, inter alia, a two-hour meeting with Groves who was "very friendly and helpful and very frank indeed". Groves thought Chadwick should have similar powers to his own, and that the leaders of the British mission should have direct access to him (Groves), so as to be able to tell him bad news if necessary. Groves was, Oliphant reported, very dubious about the proposed imminent arrival of Akers, with his connections to ICI. Akers would not be welcome.

As for Oliphant himself, Groves had said that Lawrence wanted him to be a "special consultant" on the electromagnetic process in Berkeley, in effect (in Oliphant’s interpretation) a “second-in-command”. Groves had just returned from the west and "seemed pleased with the way things were going". Oliphant was to have full access to the production site in Tennessee and to other aspects of the large-scale effort. In his view, the arrangements were “very satisfactory and with goodwill should work well".

Oliphant at Berkeley

By the time Oliphant arrived at Berkeley in late November 1943, research by Lawrence and his team into the electromagnetic separation process had been in progress for nearly 2 years363. Lawrence’s vigorous advocacy had been initially empowered by Oliphant's visit in the autumn of 1941, followed soon after by the attack on Pearl Harbor and the entry

361 This flight was, apparently, one of a number that Oliphant made over the next year in the bomb bay of a B24 Liberator bomber. See Cockburn and Oliphant (1981), p. 114-115. 362 Oliphant to Chadwick, 21 November 1943. AB 1/42. UKNA. 363 An overview of developments up to September 1942 can be found in Lawrence. Brief Report on the Programme of Development of Electrical Methods of Isotope Separation with Special Reference to the Magnetron Method. EOL 72/117/29/4, LPBL. 192

of the United States into war against both Japan and Germany. Having accepted the findings of British research, namely that a powerful bomb could be made if U235 could be separated from U238 to a sufficient level of purity, researchers in various institutes were pursuing a range of options, all depending on the subtle differences in mass between the uranium isotopes. These included ultra-centrifugation and diffusion in various forms.

Of the range of possible separation techniques, Lawrence had settled on the use of electromagnetism as best suiting the skills and experience of his Berkeley team. The process was essentially that of mass spectroscopy, a technique dating back 50 years to JJ Thomson and Aston at the Cavendish. If a beam of similarly-charged ions was sent into a magnetic field, the path of the ions would be bent into a semi- circle and the isotopes of an element would be separated on the basis of their mass. Easily stated, this would in practice prove to be a very substantial challenge. In the case of uranium, the difference in mass of the two isotopes was very small, barely 2%. When the concept was first trialled inside the 37 inch cyclotron at Berkeley, the final separation between the beams of the two isotopes was only a few millimetres.

Scaled-up versions of this technology dubbed "”, a name incorporating a reference to the University of California, had been tested between the jaws of the 184 inch cyclotron. Plans were soon in place for an even grander iteration, where many individual calutrons (later known as "bins”) would be grouped into a large oval (or later still a rectangle), all feeding off the magnetic field generated by one very large magnet. The shape of the assembly gave them the name "racetracks"’. Each bin would be essentially self-contained with access gained via a door, through which the uranium-based "charge material” could be inserted and the separated isotopes removed, along with the large

193

amounts of "gunk" or "crud”, which represented the substantial majority of the charge material which remained unseperated.

The electromagnetic process honed at Berkeley would be put into productive use under the code name K-12 at a facility (known for security purposes as Site X) established at Oak Ridge in Tennessee, a rural area about 15 kilometres from Knoxville. Groves had chosen the site for its remoteness (and the resultant potential for secrecy and high security) and for access to electricity supplies from the nearby dams of the Tennessee Valley Authority, notably Norris Dam, since large amounts of electricity would be needed. Construction work at the site had begun in February 1943, and by November the first EM separation plants, known as “alpha” tracks, were to be in operation. With the wartime shortage of copper, the conducting coils for the calutrons were to be wound from (ultimately) 15,000 tonnes of coinage , supplied from the stocks at Fort Worth364.

By 27 November 1943, Oliphant had been in Berkeley long enough to form his first impressions, which he reported to Chadwick365: "This is a fine laboratory with a very good spirit and a ferocious energy". The electromagnetic plant under development there was "a winner", making redundant any effort to develop the diffusion process. The plant in operation was far simpler than a radar sett, though there were challenges, for examples in controlling "hash”, oscillations in the source and beam . Nor, in his opinion, was the current plant ideal. Already it was apparent from research, initially in Birmingham and now at Berkeley, that the yield of the plant could be greatly increased by using much stronger magnetic fields and a much smaller curvature of

364 Among Oliphant's colleagues, the story persisted that it was Oliphant himself who suggested this source of conductive material. This has not been corroborated, and given the timing of the events involved is unlikely. 365 Oliphant to Chadwick, 27 November 1943. CHAD IV 3/ 16 (3), CPCC. 194

the beam. Oliphant would return to this theme several times over the next year.

New ideas had to wait. The task now was to get the planned facility going as soon and as efficiently as possible, and then to build more units. Already the numbers were troubling. Oliphant assessed the likely output of the nine alpha tracks already authorised and due to come on line over the next year; four with single sources (“Js”) and expected to collect 10 grams of U235 per day, and five (the “Alpha2” tracks) with four Js per bin, which would collect 40 grams a day. The combined output would total only 20 kg of U235 by the end of 1944, barely enough for a single bomb, even at the most optimistic estimates of a critical mass for U235. The implications were clear. The current racetrack numbers should be quintupled, and work on the diffusion plant abandoned, if necessary, to free up resources. Oliphant hoped that Chadwick would argue this line with Groves.

Oliphant wrote at some length to Groves366 on the same day, on much the same lines: the electromagnetic apparatus “is without doubt capable of the performance claimed for it"; the rival gaseous diffusion scheme "was still on a theoretical basis, and the one essential, a satisfactory membrane, still does not exist"; the leaders of the diffusion and electromagnetic teams should meet "for a frank discussion of the present position of each method" (though apparently Groves was not in favour of such a meeting); the capacity of the electromagnetic plant should be increased by at least a factor of five to ensure enough material was ready for military purposes in a realistic time frame.

All this is typical Oliphant, showing a wholehearted commitment to the course of action on which he was embarked, even to the point of risking

366 Oliphant to Groves, 27 November 1943, LPBL. 195

over-selling the project and minimising its problems. There can be little doubt that in the long term, the electromagnetic process was oversold. It was able to assemble (barely) the amount of material needed to build a bomb only with inputs from the diffusion process once it came on-line, and it was quite quickly supplanted by the latter on the grounds of cost. By the end of 1946, the K-12 plant would be shut down entirely and an EM plant in the United Kingdom, for which Oliphant was to argue strongly, was never seriously contemplated.

1944: a year of progress

By the start of the New Year, the British TA team in the United States under Chadwick's overall leadership had grown to about ten. Frisch and Titterton had gone permanently to site “Y” in Santa Fe; Chadwick would be flying back and forth between Y, other locations and the technical mission HQ in Washington, where Bill Webster would hold the fort administratively; Peierls would be permanently located in New York in order to provide theoretical physics support to the team developing the technology of gaseous diffusion.

Oliphant, Massey, Tomlinson, Sayers and the most-recently-arrived Duke were all in Berkeley. Their focus would be on the “Js”, the code name for the ion sources at the head of the separation process. In these, gaseous molecules of uranium chloride would be fed into a carbon arc in the presence of a magnetic field and thereby converted into uranium ions, from there to be fed as a beam into the sorting magnetic field. The current versions of the sources worked tolerably well but there were clearly issues to be tackled and improvements to be made.

196

By April 1944, Oliphant was sufficiently across the state of ion source development to write "Some notes on the production of ions for the EM process"367. The front cover stated "there is nothing new in what follows. These notes are intended only to emphasise the outstanding problems and to serve as a basis for a programme of development." Every effort needed to be made to increase the "process efficiency”, since the more uranium molecules that could be converted into ions, the greater would be the throughput of each unit and the more uranium it would separate. One limiting factor in the generation of large currents was the appearance of "hash", oscillations in the flow of ions whose nature and origin was not yet fully understood.

At much the same time, Oliphant was seeking to boost his team and so bring more resources to focus on the task. A list dated January 1944368 included a dozen names, headed by Moon and Nimmo, together with Roberts from Birmingham and Skinner from Bristol (the latter would need to be released from radar research to undertake TA work). Also on the list were a couple of mathematical physicists from the Admiralty and the Australian , then running a valve development laboratory in Melbourne. Clearly Oliphant was spreading his net as wide as possible in search of the best people. Chadwick supported these efforts, though not without misgivings as he saw the implications of the “brain drain” on work in Liverpool and Birmingham. He was right to be concerned; by the end of 1944, nuclear physics research in the UK would have but ceased369.

Over the coming months the British team grew substantially, highlighting the success of Oliphant's efforts. According to a list of names held in the

367 EOL 72/117/29/28. LPBL. 368 "List of additional personnel who could contribute usefully to the work at Berkeley and site X”, Janaury 1944. LPBL. 369 Gowing (1964), p. 242. 197

Lawrence Papers, and bearing a hand written comment "Fall 1944", the team under Oliphant's direction had grown to more than twenty by that time, divided between Berkeley and Knoxville. Leading lights in Berkeley including Massey, undertaking theoretical research, Skinner, Curran, Bunneman, Wilkins and Duke, with Edwards among the technical staff. The slightly larger group of personnel at Site X was led by Allibone, Nimmo and Sayers. The team had a core of Birmingham expatriates, with the rest drawn from industry and other universities. Oliphant spent much time moving back and forth between the two locations, as well as visiting Washington to consult with Chadwick and others and crossing the Atlantic for meetings in London and to maintain oversight of the work in Birmingham.370

On his departure to the United States, Oliphant had left Moon in charge at Birmingham, with instructions, as Moon reported to Perrin in January 1944371, to finish the large magnet of the cyclotron. When complete, this would be used for experiments on generating and focusing ion beams of the type the electromagnetic separation process would require. The emphasis would be on research of immediate help to the Berkeley team rather than on longer range objectives. As Moon pointed out, the previous experience of men like Oliphant, Sayers and others fitted them especially to work on ion source problems372 and he guessed that

370 It is not easy to assemble a definitive list of the members of ‘Oliphant's team’. A number of lists occur at different points in the public record; some members stayed until the conclusion of the war, others left at the termination of their contract period. The most substantial list (which does not carry a date but was almost certainly written late in the war) contains approximately 30 names, of whom more than 20 are designated as theoretical and/or experimental physicists and/or engineers, two are design engineers, five are experimental assistants and two are secretaries. Not only was the team substantial, it was spread over two locations separated by thousands of kilometres. In addition a number of relevant staff remained in Birmingham. Management of this widely dispersed group would have presented Oliphant with a substantial challenge, but also added significantly to his managerial experience and expertise, so contributing to his "capital". 371 Moon to Perrin, 19 January 1944. AB 1/42, UKNA. 372 For example, much of the success of Oliphant's Cavendish "basement accelerator" flowed from his innovative improvements to the ion source which greatly increased the numbers of available. 198

sooner or later Oliphant would want more of the Birmingham team to join him in California. According to recent communications from Oliphant, two particular problems were of concern. One was increased efficiency of the ion source, so as to ease the difficulty of recovery of material from the slits and walls of the bin; the other was control of the “hash” that could spoil the focusing of the beams and the separation of the isotopes373.

Moon reported to Perrin that work on the magnet was proceeding well, with the magnet expected to be operational within the current month, though much of the "labouring" was being done by the scientific staff. "We have all done to a greater or lesser share, and some people have put great energy into tasks that would have been strange to them and which in ordinary circumstances they would certainly not have been asked to do." Here we have a foretaste of the duties many of the scientific staff, especially the graduate students, would be called upon to undertake in the construction of the synchrotron.

To assess the state of isotope separation in the middle of 1944, we can draw on a report374 compiled in July by Allibone, based at X. Four Alpha1 tracks were now in place there and operating satisfactorily and several of the more advanced Alpha2 design were under construction as the result of a decision to expand the capacity of the electromagnetic plant. Two Beta tracks375 had been built but so far only one was working. Constant small modifications were being made to upgrade performance, though, at least for the Alpha tracks, "most of the gremlins

373 As Gowing noted, the British authorities were anxious to keep the EM research going in Birmingham “since that now seemed to be the most likely method for producing U235.” See Gowing (1964), p 241. 374 Allibone TE. Brief Notes on the Equipment at X, July 1944. AB1/198, UKNA. 375 Despite Oliphant’s optimism in early 1943 that the separation of the isotopes by the EM process could be undertaken in a single step, it had long been decided that two stages would be needed. The Alpha tracks would enrich to some 15% of U235 (from the naturally occurring 0.7%). Their output would be fed into the Beta tracks to complete the enrichment (to at least 85%). 199

appear to have been eradicated”. In a typical Alpha track, 96 bins were arranged in an oval or rectangular shape, each separated from its neighbour by the coils which generated the magnetic field. Each was fitted with two or four J units to generate the ion beams. Each bin was accessible via a door to enable the separated uranium to be extracted, along with the very plentiful “gunk” and “crud”, as the material which had failed to separate was colourfully labeled. The massive nature of the equipment was in strong contrast with the minute amounts of enriched uranium being collected, milligram by milligram.

Months before, Oliphant had expressed his opinion very strongly that the capacity of the electromagnetic plant should be expanded to ensure sufficient supplies of fissile material. An opportunity to make that decision arrived late in May 1944 at a meeting attended by Groves, Oppenheimer and Lawrence (presumably a meeting of the Military Policy Committee). Chadwick relayed the outcomes of this meeting to Oliphant in a letter dated 26 May376. While Groves had gone to the meeting with an open mind on the matter, and Oppenheimer was prepared to push for an extension, it appears Lawrence was noncommittal. In view of the costs of the plant and the possibility of obtaining some output from the gaseous diffusion plant, the decision was postponed. Oliphant had clearly been disturbed to hear this outcome and that Lawrence had not himself pushed for an expansion.

These developments stimulated Oliphant’s well-developed capacity for impatience. By early June, he was writing to Chadwick apparently to suggest that it might be time to pack up and go home. He had copied the letter (a copy of which does not appear to have survived in the public record) to Lawrence; in a manner which reflected his alarm, Lawrence

376 Chadwick to Oliphant, 26 May 1944. AB 1/198. UKNA. 200

replied by telegram on 9 June377. “Your Chadwick letter received. You and your team are more than ever vital on this job, and I am seriously concerned that you should suggest otherwise378.” Whatever the outcome of the May meeting, Lawrence still seemed keen to expand the number of racetracks and to upgrade those in place, and claimed to have received expressions of interest from both Groves and Conant. The reasons were plain. The diffusion process was in trouble and would need a miracle to be ready in time, and the supply of could not be counted on. Only the EM process could be relied on in the required time-frame and so it needed to be expanded.

The surviving correspondence from this period is full of interest, not least for the light that it shines on Oliphant’s character, and is worthy of more detailed examination than is possible here. For example, we can see evidence of Oliphant’s tendency to state a position forcefully while not always in possession of the relevant facts, or even in defiance of some of them. A long letter to Akers written in early July379 proclaimed that the K-25 diffusion plant was “in serious trouble” and as a consequence plans had been agreed to modify the existing Alpha2 tracks (the “Alpha3 conversion”) to boost their output and bring forward the date at which a critical mass of enriched U235 could be available.

Oliphant sent the letter through Chadwick, as was appropriate (it was also to be copied to Cherwell, Anderson and Appleton) and invited him to attach a covering note if he disagreed with any statement (though he asked that the letter be forwarded unedited). Chadwick was as cautious and conservative as Oliphant was outspoken and sometimes

377 Lawrence to Oliphant, 9 June 1944. LPBL. 378 As Gowing later noted, there was good reason for Lawrence’s alarm. The Americans, though well supplied with engineers, were short of competent experimentalists, and needed the British team to solve problems with the equipment and to train the operators. See Gowing (1964), p. 257. 379 Oliphant to Akers, 3 July 1944, CPCC. 201

intemperate. Chadwick did add some comment and those tempered some of Oliphant’s more strident statements, though he was in agreement with many of the conclusions, such as that work must begin on an EM plant in the UK as soon as possible. He was able to add important information most likely unknown to Oliphant, for example that a new form of plant for low-level enrichment, known as a “thermal diffusion plant”, was under development, and that therefore a third path was opening to run beside the Alpha tracks and K-25 as feeds to the Beta tracks. This weakened Oliphant’s case for an upgrade of the Alpha2s, though it would not be running before January.

The closing of the final paragraph of Chadwick’s covering note was significant as a summation of Oliphant’s attitude.

He concentrates on his own part of the project and will not consider calmly the overall picture... . Notwithstanding, there is usually some point in his observations even when they appear wild and distorted, and our co-operation with Berkeley depends almost entirely on him. It is essential to keep him reasonably contented with his own activities and general policy380.

Thereafter matters seemed to settle, on the surface at least. Oliphant continued to press for experimental investigations that were likely to yield results applicable to the current EM system. According to a memo written in September 1944381, those included the merits of using uranium hexafluoride as the "charge material" instead of the chloride, and the implementation of a new form of wide “self-focusing" ion source which could replace the multiplicity of Js now used in each bin.

380 Around this time, Oliphant was threatening to resign from TA work altogether if the needed new men he had been calling for to boost the British presence at Berkeley were not provided “because of further funny business in London”. Oliphant to Chadwick, 4 July 1944. CPCC. 381 Experimental work urged by MLO as likely to yield results applicable to the present EM system. 28 September 1944. LPBL. 202

Doing it differently

In the midst of all this busy-ness, Oliphant found time to put on his visionary spectacles and propose significant changes in the way the process was working. He had been doing this almost since his arrival, as a memorandum written in February 1944382 revealed. Oliphant began by proposing a "new form of apparatus which could replace further "racetracks" if it were decided that additional equipment [for the separation of the isotopes of uranium] should be built in America or in England". That proposal was based on the observation that separation by electromagnetic methods "is more stable in operation and gives larger output as the magnetic field is increased". Oliphant was proposing to raise the magnetic field by an order of magnitude to 30 000 gauss, which lay beyond the limits of conventional iron-cored magnets. The memoranda referred, typically, to the “largest possible currents” and to "voltages as high as can be conveniently applied". He talked of cooling systems with liquid hydrogen or even liquid helium, the latter producing superconducting effects. It is not clear that any such systems were then in use. Oliphant also wanted to draw upon Birmingham experience in the design of the apparatus.

The equipment Oliphant was proposing, if it could be built, would be much smaller than any existing Alpha track, perhaps only one percent the size, but would deliver much the same output of enriched U235. The capital expenditure would be cut by a factor of 20, and running costs in proportion. "To secure an output of U235 of the order of 1 kg a day383, 100 such equipments would be needed.” The total power consumption

382 Oliphant, A Simplified Form of Electromagnetic Separation Equipment, 15 February 1944. CPCC. 383 The memo noted that at current estimates of the critical mass of U235 this would be equivalent to one weapon a month. 203

would be a massive 600 000 kW, though Oliphant did propose ways in which that might be reduced.

It is not our purpose here to assess the feasibility of the scheme that Oliphant was proposing, even if we had the capacity to do so. It is enough to say that within months of arriving at Berkeley, Oliphant was advocating a course which would overthrow much of what was being currently done. Of course it did not happen. Oliphant and his team had entered a situation where a stream of technology development was already underway, indeed had been for many months, and the best they could hope to do was to contribute ideas that would enhance the operation of that technology.

There was another threat to Oliphant’s vision that a greatly expanded electromagnetic plant could single-handedly generate sufficient material for a bomb that might win the war. That challenge came from "94", the code name for the newly discovered element plutonium. Evidence had accumulated that this element, generated in minute quantities inside a nuclear "pile"384, could be split by neutrons in a similar manner to U235. It was proving more difficult than expected to turn a mass of plutonium into a weapon385, but 94 still appeared to have a significant advantage. Experiments suggested that a critical mass of 94 would be only 20 kg whereas for "25" some 50 kg would be needed. Oliphant confessed to Chadwick that he was "shaken" by the statement386.

384 Plutonium was generated by a process which began with the absorption of neutrons by U238 without fission. That reaction produced element 93 ("neptunium”), which then underwent (“”) to form element 94. It was for this discovery that McMillan and Seaborg later received the Nobel Prize. 385 The critical issue was the presence in plutonium generated in reactors of small quantities of a plutonium isotope which could cause premature fission. The proposed "gun” method to assemble a critical mass of plutonium by bringing together two sub-critical masses, as was intended for a bomb based on uranium, was therefore not feasible, since it could not take place fast enough. Instead the "implosion" method was devised, in which a sub-critical mass of plutonium would be squeezed to criticality by a number of explosions set off around a spherical mass of the element. See Huddeson et al (2004). 386 Oliphant to Chadwick, 7 June 1944. AB 1/198. UKNA. Chadwick was at Y at the time. 204

Yet clearly some part of that vision remained in the back of his mind. A memorandum to Groves, written from Site X on 17 October 1944387, outlined "an alternative form of electromagnetic plant which shows economy of capital cost and running expense as compared with existing plant”. He intended that the proposal, which was "not final", would “promote interest in the future of the electromagnetic system and to show how the fundamental difficulties with the existing plant could be overcome without departing widely from existing experience." In other words, the proposed changes were significant but not revolutionary.

The key element of the system was, again, the use of uranium hexafluoride as the charge material, and the use of a solenoid388 to generate the magnetic field, since by this method the field could be made much more intense. Oliphant maintained that these and other improvements would simplify much of the chemical processing and reduce costs, by eliminating steel from the magnetic circuit and cutting the workforce substantially. A single racetrack based on these ideas would, he figured, process 300 g of U235 a day, enriched by a factor of 20. The closing words of his introduction were typical Oliphant. "I am convinced that development in these and other directions can make the EM system of separation the cheapest and simplest of all systems yet proposed, when due account is taken of the overall costs of production.”

Again it is not our purpose here to assess the merits of Oliphant’s proposals. It is sufficient to note that he put them forward at a time when, one would have thought, effort was urgently needed to enhance the operation of the existing plant. Yet the timing was not inappropriate.

387 Oliphant, A Revised Electromagnetic Separation Plant. 17 October. 1944. LPBL. The covering letter to Groves was dated 30 October 1944. 388 That is, an air-cored magnet without any steel. 205

As Oliphant remarked in a letter389 to Groves on 13 November, just before leaving for the United Kingdom, things were looking brighter. The output from the Beta tracks had shown an abrupt lift, a development which Oliphant called “very satisfying”. This was the result of a combination of factors that promised well for the future. The servicing of the Alpha2 tanks had also improved. These developments would have released time and mental effort for Oliphant and others to consider less immediate matters. As we shall see, those matters included his embryonic ideas for the proton synchrotron.

Oliphant was not the only one looking to the future. As early as May 1944, plans were being made about atomic energy issues in the UK once the war was over. The T. Technical Committee met in Washington in that month (so many of its members now being based in the USA) to discuss the possibility of a UK "research establishment" where a program independent of the Americans might be pursued. While a broad range of issues was discussed, Oliphant was on record as continuing to push the superiority of the electromagnetic process over any other if enrichment of uranium was to be a priority in the new facility.

The group met again on 11/12 November to move the discussion on. Among those present were Cockcroft, Peierls, Frisch, Webster and Massey, as well as Oliphant, who had a lot to say, according to the Minutes390. He read a statement about a suitable structure for the establishment, pushing the need for “good men from the universities” and circulated a memorandum continuing to expound the virtues of the EM process. Another gathering was held in the spring of 1945. Though it appears Oliphant was not present, “the work of the electromagnetic laboratory was set out in some detail".

389 Oliphant to Groves, 13 November 1944. LPBL. 390 Minutes of meeting, 11-12 November 1944. AB 1/213. UKNA. 206

Once he returned permanently to the UK, Oliphant was on an advisory committee for "Harwell", as it was later to be called, and was instrumental in finding a suitable site, on a large airfield near Oxford391. He continued to promote the electromagnetic process, but this was now a losing battle. As Gowing reported, opinion was hardening in favour of "94" as the basis of any future British nuclear weapon program, on the grounds that a plant to produce 1 kg a day of fissile material would cost only half as much if that material was plutonium than if it was enriched uranium. So a nuclear reactor rather than an electromagnetic plant would most likely be the source of the fissile material.

Oliphant would have been disappointed in this outcome, and with good reason. As Gowing commented, the British scientists in the USA working on the electromagnetic process (meaning Oliphant's team) had come to know much more about that process than their colleagues involved in the diffusion or plutonium routes to the bomb. "It was ironic that the one large-scale project for producing fissile material to which the British had free access was also the one which was not developed in the UK after the war. This was the electromagnetic separation392 .”

Time to go

From mid-November Oliphant was absent from the United States for nearly two months. He spent much if not most of this time in Birmingham with his family and with his colleagues at the University. As we shall see, it was during this period that he appears to have formulated his first ideas about the proton synchrotron, or at least first set them down on paper. He also met with Tizard who was offering support to the new

391 The name of the establishment was taken from the name of the airfield (RAF Harwell). 392 Gowing (1964), p. 256-261. 207

project. He had one meeting with Anderson on 9 January of which a transcript of sorts exists. He apparently discussed his thoughts on post- war atomic energy research in the UK, including the proposed establishment of the research facility. He does not appear to have discussed his plans for the new accelerator, which he would later claim to have talked about with Anderson.

Anderson had some grounds for ambivalence about Oliphant. On 3 October 1944 Oliphant had forwarded to London a document393 which purported to be a record of part of a conversation between Groves, Lawrence and Oliphant at Berkeley. The content of the two page document was far-ranging and need not be described in detail here, but it raised a number of sensitive topics. For example, it suggested that parts of the Manhattan project were still off-limits to the British and that Groves believed that Russia would already have full knowledge of the atomic bomb project through "communist sympathisers among the American or British scientists on the project". It is sufficient to say that the document was judged by Akers as "remarkable", and by Anderson as "dangerous stuff". He wanted to know how it had arrived and said that it should be "locked away". Oliphant would need to be warned about it.

Oliphant returned from the UK in mid-January 1945, spending some time in Washington before moving on to Site X and then to Berkeley. A pithy letter to Groves394 (whom he did not get the chance to see while in Washington) dated 22 January had significant content: congratulations for the very gratifying and significant improvement in performance of the Y-12 plant, an apology for his unexpectedly long stay in the UK (“the

393 Memorandum from Oliphant. Record of Part of a Conversation between Maj Gen LR Groves, EO Lawrence, and ML Oliphant, 3 October 1944. UKNA. 394 Oliphant to Groves, 22 January 1945. LPBL. 208

reasons for this cannot yet be discussed”395), and advice that he had informed the UK authorities “I must return permanently to my duties in Birmingham late in March. I mention this now so there will be no misunderstanding when the time comes.”

Oliphant had copied this letter to Chadwick, who apparently felt impelled to respond quickly396. He assured Groves that Oliphant was "not as uncompromising" as his letter had indicated. He would stay on in the US if he could be "really useful" and if it was necessary in the interests of the work. However, Oliphant had told him he had "urgent reasons not connected with TA which demanded his return to England." Oliphant felt he had exhausted his usefulness to the EM research and "was naturally anxious to be where he could be most useful". We have seen such an imperative before, in the abrupt decision to go to Australia and his later anxiety to get home again, and his quick (though not necessarily hasty) move from radar to TA research when the grass under the latter venture appeared greener. Chadwick also underlined the forcefulness of Oliphant's character when he commented "I am also sure that Sir John Anderson would wish Oliphant to remain here if the project demanded it but he will not be able to enforce his wishes if Oliphant makes up his mind to go”.

Indeed Oliphant had decided to go, but there were still tasks to complete. In early February 1945, Lawrence had asked him “to coordinate the growing volume of research into J phenomena", and he had convened the first of a series of meetings for this purpose397. Plans

395 This remark may have been a cover for his desire to get both his university ready for post- war life and preparations underway for his new form of accelerator, which had taken shape in his mind. As we shall see it was from Washington, and only five days before his letter to Groves, that Oliphant wrote a key letter to Akers seeking interest in the new venture. Perhaps even more important than these issues were discussions Oliphant had had on the shape of post- war nuclear research in the United Kingdom. 396 Chadwick to Groves, 23 January 1945. LPBL. 397 Oliphant; Memorandum on J Research. 8 February 1945. LPBL. 209

were in place for these to run at least until late March. Soon afterwards, he produced a memorandum398 summarising the issues to be discussed. His optimism about the future of the EM process was undimmed: "The spectacular success of the present plant encourages the belief that the EM process is inherently simpler and more practicable than others proposed so far." It was, however, more expensive in terms of the cost per unit of enriched uranium, due to higher building and operating costs. Those issues needed to be addressed if it was to continue to hold its advantage over other processes.

On 16 March 1945, Oliphant wrote a farewell letter399 to Lawrence, though he does not appear to have left Berkeley for another week or so. The letter is full of praise for Lawrence, his laboratory team and their collective achievements, together with thanks for the welcome that they had given the UK men, even if the latter were "interlopers at a late stage of the work". The plant now successfully operating at Site X was "a witness to the faith of those responsible for America's technological effort in the ability of the Laboratory to do the impossible. The success of the plant in operation, and the growing stockpile of material it produces, have justified that faith and have earned the admiration of all." Oliphant also expressed "his own personal gratitude for your frank treatment of me as your colleague".

Lawrence replied in a similar effusive vein on 20 March400, expressing appreciation of the contribution of the whole British team in bringing the EM process to fruition, and in particular to Oliphant for his "faith and vision and sound judgement". "You have been of vital influence in shaping things in the right direction all along." Lawrence appeared to

398 Oliphant; J Research Program. Undated memorandum (but circa mid February 1945). LPBL. 399 Oliphant to Lawrence, 16 March 1945. LPBL. 400 Lawrence to Oliphant, 20 March 1945. LPBL. 210

think that Oliphant would be back after a short visit to the UK, perhaps to go to Y, and might even change his mind about leaving at all if the situation required that he stay. Yet as Oliphant made clear in a letter to Massey401 on 21 March, he was leaving permanently to return to the UK, leaving Massey in charge of the UK team with Skinner as his number two.

Oliphant set out in some detail his reasons for going402; “I believe the work here is in an era of diminishing returns." The present plant was working well and there was understandable reluctance to interfere with it. In addition, few ideas being discussed showed promise of working well enough to be incorporated. The K-25 diffusion plant was about to come on stream, able to feed partially enriched material into the Beta tracks, and the Alpha tracks would presumably soon be shut down403. Since the war might well be over before the weapon was ready, the Army was reluctant to commit more expenditure.

Then there was the issue of whether “94” (plutonium) might not be better for military purposes, as it had only half the critical mass of uranium. A trial of the plutonium bomb, using the implosion method, might be possible in July or August. Since about 80 kg of U235, enriched to 75 to 80 percent, would be needed for a bomb, the total production of Y-12 would need to be accumulated to allow for "the production of a really big explosion towards the end of summer". This suggested there could be no testing before use.

401 Oliphant to Massey, 21 March 1945. LPBL. 402 Ibid. 403 Simon's 1943 prediction that the EM process might be more useful at the "high end" of separation turned out to be true. From March 1945, the gaseous diffusion plant code-named K- 25 provided a feed of slightly-enriched uranium to Y-12, so making the Alpha tracks obsolete, as Oliphant had predicted. From the summer of 1945, the S-50 thermal diffusion plant did the same. 211

Oliphant signed off by wishing Massey well, but urging that he "watch very carefully to see that our men are properly employed. If it becomes obvious that the project is folding up, all personnel should be sent home promptly". This included "my own boys on the technical side. They will be needed urgently in the UK”.

The circumstances surrounding Oliphant's departure from the USA were further illuminated by correspondence between Chadwick (who had of course known Oliphant personally for the best part of two decades and was to remain a lifelong friend) and Anderson. The surviving letter from Chadwick404, dated 24 March 1945, not only includes pertinent (and well-founded) observations on Oliphant's character and modes of operation. It also brings forward a wider issue, namely the nature of the relationship between Oliphant and Lawrence, not really evident in correspondence between them. This relationship was not, it seems, all that their warm letters of farewell might have indicated.

Chadwick began his detailed three-page letter by reminding Anderson that the "Oliphant problem is no new one". Oliphant’s desire to quit America and return to the UK (with many if not most of his team) dated back six months to September 1944 "arising from a change in the Berkeley program which went contrary to Oliphant's wishes and to what he considered the best interests of the project"405. Chadwick had hoped that time spent by Oliphant in the UK (from November) might have "relieved his mind to some of the unnecessary kinks” but clearly it had not.

404 Chadwick to Anderson, 24 March 1945. AB 1/19, UKNA. 405 Chadwick was most likely referring to a proposed shift in priorities at Berkeley away from military objectives and toward postwar civilian activities; Lawrence reportedly felt that enough enriched uranium could not now be collected in time for the weapon to be useful in the current war. Oliphant had strongly protested against this shift at the time. See Oliphant to Chadwick, 4 September 1944. CPCC. 212

To a large degree, Chadwick sympathised with Oliphant’s reasons for wishing to get home. He agreed there was little work of the right kind for him to do at Berkeley. Furthermore the relationship with Lawrence had proved difficult. Of Lawrence, Chadwick said "he is a delightful friend and a very able man but the only way in which he can run a laboratory is as a dictator". He had made Oliphant "a kind of assistant or deputy director and put a whole section under Oliphant's control", but could not refrain from interfering with the work or even countermanding Oliphant's directions. "Oliphant has borne his troubles with surprising equanimity and has succeeded in maintaining the most friendly relations with Lawrence, but he now feels he has reached the end of his patience." For that reason alone, it was better that Oliphant leave, rather than risk a rupture with Lawrence which might have serious and wide-ranging consequences.

Chadwick now came to his key point. Whether or not Oliphant has good reasons for leaving Berkeley, it is clear enough that Oliphant was determined to do so, and that further argument would only lead to trouble. Chadwick had consulted with Groves, who “understood Oliphant's little ways". They agreed that it was too late for Oliphant to go to Y, as had been the original plan, and (reluctantly) that there was very little to be done but accept the fact that Oliphant should go back to England. Chadwick had arranged for Oliphant to see Groves on 26 March, but regretted that he could not be there himself, as "Oliphant is apt to say a great deal more than the situation or the facts require”.

In closing, Chadwick mused on other reasons Oliphant had for wanting to get back to the UK, beyond the lack of relevant work at Berkeley and the strained relationship with Lawrence: the wishes of his family, the desire to get moving on a British EM plant, in order to prevent the UK falling too far behind, and on other potential industrial uses of nuclear

213

energy (which Chadwick thought he "over-emphasised"), the state of his Birmingham laboratory with the demands of the post-war world looming (a concern Chadwick shared, having a laboratory of his own waiting for him in Liverpool).

But there was possibly more. Chadwick suspected (with little evidence, as he admitted) that Oliphant had used his last trip home to make plans for the future of his own laboratory, perhaps to secure promises of money, to "get in on the ground floor". It seems Chadwick knew nothing in detail of Oliphant's plans, such as for a new accelerator; at least he did not mention them. Nor did he criticise Oliphant for so acting, though he would have preferred "to arrange such developments more reasonably and more openly and to consider them from the general interest of the country. If I were at home, I could help with such matters but it is impossible situated as I am.”

As we shall see, Chadwick's suspicions were justified. In the following chapter we shall see what use Oliphant made of his time away.

Summing up

A useful summary of some of the narrative contained in this chapter is found in an unsigned and undated memorandum, preserved in the public record and dealing with the role that Oliphant and the others had played in the work at Berkeley and at Oak Ridge. Despite its uncertain provenance, this document406 is a succinct overview of the progress of events, and is therefore worth quoting at some length.

406 British participation in research at the Berkeley Radiation Laboratory. Undated and unsigned memorandum. AB 1/213, UKNA. Since the authorship of this memorandum is not certain, we are not able to say if it was written from an American perspective or a British one. It may have been written by Oliphant himself. 214

In the early days of the war, before America was actively involved, there was complete exchange of information on policy and techniques for exploitation of nuclear energy for military purposes. In particular there were very full discussions with Dr Oliphant during his visit to the United States in August and September 1941. Dr Oliphant had talks with Dr Lawrence at Berkeley, and endeavoured to convince him that the TA project was one to which he should devote the greater part of his energies, as the problem of isotope separation was ideally suited to solution by the methods of teamwork developed in the Radiation Laboratory.

As a result, largely, of Dr Oliphant's visit to the USA, Dr Urey and Dr Pegram went to England and saw there all that had been done and talked over all that was planned. On their return to USA, they were able to stimulate a much, enlarged programme of work. Mr Akers, Dr Simon and Dr Peierls of the British TA organisation visited Berkeley at a later date and saw the early work on the electromagnetic process407.

The exchange of technical information on TA ceased abruptly when the US Army took over the project and it was re- established, and then not fully, only as a result of the Quebec agreement.

In September 1943, Dr Oliphant accompanied Dr Chadwick to Washington, and there helped to determine the form of collaboration. As a result of what he learned about the existing position of the isotope separation methods, he felt that the

407 It is presumably on the basis of this visit that Simon framed his negative comments to Oliphant on the prospects for the EM process in June 1943. 215

primary job was to establish the essential supplies of fissile elements. Accordingly, he was unwilling to take his team to “Y” until such time as those supplies had been assured. He elected to join Lawrence in Berkeley, as he believed that the electromagnetic project was more likely to prove successful in a reasonable time408. The development of the whole project has not been so fast as was hoped at that time, so the original arrangement that Oliphant move to “Y” after about six months was abandoned.

[The document then goes onto list the members of the team under Oliphant’s direction].

These men had taken part in the general work of the laboratory in Berkeley and at the site. Dr Massey has taken charge of the theoretical aspects of the project. Among the original contributions made by the British team are the "splutter" source of metallic ions, the "gas box” stabilising chamber for the arc source, the "photo-arc” for hash suppression, "negative" shims for correction of foci and numerous investigations into fundamental phenomena. Duke and Starling have contributed to the engineering and design of equipment, particularly Alpha3, M2X and Alpha1- 4J conversion.

As for Oliphant himself, official recognition of his contribution was represented by the proposal that he be awarded the Medal of Freedom with Gold Palm, the highest award the United States government could grant to foreigners. It had been proposed by Groves, whose letter of

408 That is, in comparison with the gaseous diffusion approach. It can be argued that had Oliphant himself written this document, he would have made reference to the better fit of the EM process with his own skill set and experience. 216

recommendation to the Secretary of State in August 1946 survives in the public record 409.

It is recommended that the Medal of Freedom with Gold Palm be awarded to Dr MLE Oliphant for exceptional meritorious conduct in the performance of outstanding services in the United States in connection with the development of the greatest military weapon of all time, the atomic bomb.

A physicist of eminence in the United Kingdom and the United States, Dr Oliphant served from November 1943 to March 1945 as special scientific consultant to the Radiation Laboratory, University of California, Manhattan Engineer District. In this position of responsibility and scientific distinction, he was in charge of extremely difficult and essential research and development in connection with the development of the electromagnetic method of separating uranium 235. His cooperation with his American colleagues has always been enthusiastic and his enterprise and interest in the problems at hand most stimulating.

Dr Oliphant’s sound scientific judgement, his initiative and resourcefulness, and his unselfish and unswerving devotion to duty have contributed significantly to the success of the atomic bomb project.

A shorter citation, as might have been used during a presentation, also survives. This says, inter alia,

409 This documentation quoted here (a copy of which is held in EBA) was obtained from the archives of the US Atomic Energy Commission by the co-author of Cockburn and Ellyard (1981). While the documentation has all the appearance of veracity (including acknowledgement of its declassification in 1961), more complete details of its source are not currently available. 217

As a special scientific consultant to the Manhattan Engineer District, US Army, he was responsible for extremely difficult and essential research and development work in connection with the development of the atomic bomb. A physicist of eminence in both the United Kingdom and the United States, Dr Oliphant exhibited extraordinary scientific judgement, initiative, resourcefulness and devotion to duty in work which was indispensible to the success of the Atomic Bomb project.

Oliphant was not the only one to be so honoured. The documentation quoted above indicated that Cockcroft, Peierls, Frisch and Penny were to be similarly recognised. Oliphant never received his award. The documentary record cited by Cockburn and Ellyard410 indicated that he was denied the opportunity on the grounds that it conflicted with the then policy of the Australian Government that Australian citizens not receive foreign awards.

As was stated in the introduction to this chapter, its purpose has been threefold: to outline the context within which Oliphant came to devise the concept of the synchrotron, to discover to what extent his wartime experiences added to the stock of "capital" available to be expended in the great enterprise of the proton synchrotron, and to examine aspects of Oliphant's character and modus operandi, as revealed by his wartime experiences which could impact upon the great enterprise, for better or for worse.

Regarding the first consideration: that the concept of the synchrotron arose during the development of the Y-12 plant has been referred to at

410 Cockburn and Ellyard (1981), p. 197-198. This states, inter alia, that only Oliphant was to get the Medal with Gold Palm. That appears to be an error. 218

appropriate times in order to give an indication of the context in which it had developed. The issue will be considered in detail in the following chapter. As to”capital”, some accumulation is evident, though in some cases the foundation was already well laid. The complexity of the equipment needed to realise the electromagnetic separation of the isotopes did not frighten Oliphant; as in previous circumstances, he revelled in it and would do so again in the challenge of the synchrotron. We may say that to a substantial extent the wartime experiences strengthened that attribute of technological “green fingers” already evident. He gained substantial experience in managing a large team, indeed one spread over several widely dispersed sites. He had not supervised so large a team before nor would he do so again but the experience would be invaluable in the post-war years in Birmingham.

Throughout this period, Oliphant had interacted with policymakers and bureaucrats at almost the highest level of authority, including people with the level of influence of Tizard, Groves, Lawrence, Chadwick and Anderson, and had had some varied success in securing acceptance of his point of view, which was always vigorously argued. Furthermore, he had interacted closely with leading figures who would be important in securing the Birmingham enterprise. Those included Akers and Anderson, though his relationships with them had been far from smooth. Certainly they now all knew who Oliphant was and his ways of working. It is clear in a number of instances that his colleagues were prepared to go along with his “ways of working” in the expectation that his participation would deliver substantial outcomes. His reputation as a leading figure in the Manhattan Project would cast a long shadow into the future.

As to Oliphant’s character as revealed by these events, we have some clear pointers. On the one hand his wartime experiences revealed him

219

as energetic and persistent, always ready with new ideas, or more broadly with a vision to empower those ideas. He was very hard to shift once he had made up his mind. On the other hand he was impatient, ready to change course quickly if a better route opened up, or appeared to. That impatience had already manifested itself and would do so again.

In addition, it seems reasonable to suggest that the experience of researching and implementing the technology for the Oak Ridge plant, where many problems were solved “on the run” as they appeared, and there was not time for methodical detailed planning to cover every eventuality, shaped his view that a willingness to innovate and “a fire in the belly" (as he would later express it) was the way to meet any challenge. The Birmingham enterprise about to begin would demonstrate that philosophy.

220

CHAPTER SIX The Birmingham Proton Synchrotron 1944 to 1946: Conception and Funding

In developing a chronology of the events and decisions that led ultimately to the building of the proton synchrotron in Birmingham, we have first to find a starting point. When and under what circumstances did the ambition for the project, and the first indications of how it might be achieved, first take shape in Oliphant’s mind? Once we have established the beginning, we can then track the steps by which Oliphant refined the idea, then moving to secure funding. From that point, the vision could become reality.

Oliphant’s own recollections, which were briefly quoted in the introductory chapter, were given at greater length in the Memorandum which he wrote in 1967, at the request of the University of Birmingham411. On the issue of the conception of the new technology, Oliphant wrote

In 1944, during the commissioning of the electromagnetic isotope separation plant at Oakridge (sic), Tennessee, it was necessary for the physicists to spend long hours supervising operations and eliminating the multitude of teething troubles inevitable in transforming a laboratory process to one which is successful on a large scale. EO Lawrence, in whose laboratory at Berkeley the process had been worked out, had made me his deputy, so it was necessary for me to take my turn in the so-called “owl watch”, from midnight to 8:00 am. There was little one could do, actively, in this dismal period, unless troubles developed412.

411 Oliphant (1967) Part 2. 412 Ibid. p. 1. 221

Here Oliphant located the genesis of the synchrotron very clearly in time and place; during the running-in of the uranium isotope separation plants (K-12) at Oak Ridge. That places the date in mid to late 1944.

I found myself speculating about what physics we could do in Birmingham when the war was over. The cyclotron would become a basic tool for nuclear research and training, but after the intensive effort and excitement of the war years, it seemed desirable to try obtaining bombarding particles with greater energy than this would provide.

The synchrotron principle was born in the following way. The maximum energy which could be obtained from a cyclotron was limited by the loss of phase of the circulating particles as the energy, and hence the effective mass, increased. In addition, a magnet of large pole diameter was very costly, since both the mass of steel and copper, and the power required to energise it, increased much more rapidly than the diameter. It seemed that it might be possible to decrease greatly the mass of the magnet by using an orbit of constant radius, and providing only a narrow range of magnetic field, the value of which increased as the particles gained energy. If the problem of matching the magnetic field, at every instant, with the momentum of the particles could be solved, there appeared to be no limit to the energy which they could be given, other than cost413.

The key motivations behind Oliphant’s vision were recalled here: the desire to advance nuclear physics by working with particles of much greater energy than previously available; the limitations on existing

413 Ibid. p. 3. 222

technology such as the cyclotron as vehicles for reaching such energies; and the way in which a new approach might overcome those barriers and reach those new energies at an acceptable cost.

Having reached these conclusions I went no further with the analysis but began to think of the hardware involved and the possible cost of an accelerator of about 1 GeV …. At the then ruling prices for steel, copper and electronic equipment, I estimated that an accelerator for 1 GeV could be built for less than £100 000, and in terms of the cost of the nuclear energy project this seemed trivial enough to be practicable. So I wrote to Sir Wallace Akers414, sending details of the proposal and asking him whether he thought that such an expenditure could be met by the Tube Alloys project, as a contribution to the development of basic nuclear physics, as soon as the war was over. I received a very noncommittal reply containing the implication that my job was still the development of nuclear energy, and that it would be better if I confined myself to that task415.

We now must ask to what extent these vividly-expressed recollections, set down more than 20 years after the events they describe, are reliable. How does other evidence, both documentary and oral, either support or contradict those recollections? Through such a process we will come, as close as can be, to a true account of the event.

414 Head of the Directorate of Tube Alloys (DTA) in the Department of Scientific and Industrial Research (DSIR). As has been noted, Oliphant’s attitude to Akers was ambivalent. He had opposed Akers, and ICI in general, being given so leading a role in the TA enterprise, and his wartime correspondence with colleagues such as James Chadwick contained references to the poor reputation Akers enjoyed with the Americans. Nonetheless, Akers was the man at the top at the time, with control of funding. It was therefore to Akers that Oliphant directed his radical proposal for a new accelerator. 415 Oliphant (1967) Part 2, p.3. As will be noted later, Oliphant’s recollection about receiving such a reply from Akers is not backed by any surviving documentary evidence. 223

We immediately are faced with a contradiction. For many years, Oliphant maintained that his first communication with Akers on the matter of the new accelerator was in 1943, not 1944. Referred to on a number of occasions, this was explicitly stated in March 1947416, in the first published paper describing the proposed machine.

In September 1943 one of us submitted to the Directorate of Atomic Energy in the Department of Scientific and Industrial Research a proposal for the acceleration of electrons and protons by a new method to energies above 109 MeV. Subsequently, and independently, similar proposals were made by McMillan (1945) in USA and Veksler (1945) in U.S.S.R.

The “one of us” is Oliphant. The “MeV” is clearly a typographical error, missed by the writers and the proof readers (the term should have been simply “eV”). However more discussion is prompted by the date quoted, September 1943. It would be tempting to think of that as a similar slip of the pen, had it not been quoted elsewhere (and earlier). As was noted in the attributional survey, this date was widely accepted and quoted by other workers in the field. No-one, it appears, thought to question the date, or indeed the existence of such a document.

However the existence of a copy of the proposal was not unequivocally confirmed until 1993, some 50 years after it was reputedly written. In his preface to the published proceedings of a one-day symposium to mark the 40th anniversary of the first operation of the Birmingham synchrotron, the editor, PM Rolph, noted

The School of Physics and Space Research possesses a carbon copy of (the then) Professor Oliphant’s September 1943

416 Oliphant, Gooden and Hide (1947). 224

memorandum to the Directorate of Atomic Energy, DSIR. Given both the historical importance and the intrinsic interest of this document, it has been reproduced and included in this volume as Appendix D. It seemed appropriate to reproduce the original typescript417.

The document was labeled in the Proceedings as “The Oliphant Memorandum”, by which title we will refer to it henceforth. It is headed The Acceleration of Particles to Very High Energies, and a notation, clearly in Oliphant’s handwriting, states that the document is a “Copy of memo submitted to the Directorate of Atomic Energy, DSIR, in Sept, 1943”. That is the only external evidence of its date. It is not signed or even initialled.

It must be said at once that should the Oliphant Memorandum have been written in 1943, as claimed by Oliphant and recorded by others, it would have been by far the earliest statement of an intention to build such a machine to pursue research in nuclear physics. It would have predated any other relevant document by several years. We can therefore appreciate the importance of seeking to authenticate it. There is a concurrent issue. It is easy to make the assumption, given Oliphant’s assertions about the surviving document, that the Oliphant Memorandum and Oliphant’s proposal to Akers are one and the same. The editor of the Proceedings made such a connection. As we shall see, this is not necessarily so.

Tracking down the Oliphant Memorandum

The most assiduous early seeker after a copy of the Oliphant Memorandum was the Nobel Prize-winning American physicist and

417 Rolph (1994), Preface (page not numbered). 225

accelerator pioneer Edwin McMillan418, whose efforts to find it extended over several decades. The following account is drawn from correspondence between the author and McMillan in 1980419.

Writing to Oliphant on 30 December 1946, McMillan asked “Also I would appreciate it if you could send me a copy of your original paper of 1943, which was listed in one of the documents you sent but was not included”420. Oliphant had apparently sent McMillan some papers dealing with the synchrotron project he was by then pursuing, one of which referred to something written in 1943. The request did not, apparently, achieve a result, as McMillan made a further inquiry three years later.

My next contact came on August 31 1949 when I stopped at Birmingham on my way to an international physics conference at Basel and Como. Oliphant, Gooden, Fremlin and some others were there and I had a good look at the still unfinished machine. I asked Oliphant about his 1943 proposal and he searched through some filing cabinets but couldn’t find it. He said that things were in disorder because he was making arrangements to return to Australia421.

There matters rested, as far as McMillan was concerned, until 1965.

Oliphant had visited us in Berkeley. We discussed the matter again and on his return to Australia he sent me a letter, dated 8th

418 Edwin McMillan (1907-1991). American physicist and chemist and Nobel Prize winner as co-discoverer of the first transuranic elements. Later Director of the Lawrence Berkeley Laboratory. For his National Academy of Sciences biographical memoir see http://www.nap.edu/html/biomems/emcmillan.pdf 419 McMillan to Ellyard, 5 November 1980. EBA. The correspondence contained a number of attachments which are referred to below. The correspondence was undertaken in the context of research for Cockburn and Ellyard (1981) . 420 McMillan to Oliphant, 30 December 1946. EBA. 421 McMillan to Ellyard, Op cit 226

October 1965, and with it a copy of a letter to the archivist at the U.K.A.E.A in London. But [these items] do not really solve the mystery of the 1943 memo, as the memo to Akers is dated January 17, 1945422.

Though McMillan claimed not to have found what he was seeking. Oliphant was quite sure it was the right document, describing it in his covering letter as “my original letter to Akers”. Whether McMillan continued to hunt for it is not recorded. In 1980 he commented that “The 1943 memo seems to have vanished”. In 1982, McMillan noted that a search by at Oliphant’s request failed to find anything of significance in the public record before July 1945423.

We need to assert here, inter alia, that a September 1943 date for the Memorandum is most unlikely to be correct. This is despite the appearance of “September 1943” in Oliphant’s handwriting on the surviving copies, and Oliphant’s referral to that date in citing a proposal (though not necessarily this document) in his letter to McMillan in 1946 and in his paper with Gooden and Hide in 1947. We have noted earlier that in September 1943, Oliphant was preparing to move to the USA to work with Lawrence in Berkeley as part of the Manhattan Project. The new co-operation arrangements with the USA were barely in place. He did not make that move till November. Under the circumstances, it would be surprising if he had the time, the opportunity or the motivation to produce such a document.

As noted above, Oliphant himself gives support for a changed date. In his recollections of the genesis of the concept of the synchrotron written in 1967 and quoted above, he gives the date of his first memorandum as

422 Ibid. 423 McMillan to Ellyard, 27 September 1982. EBA. 227

1944 and underlines that by describing the circumstance of time and motivation for the generation of the idea, namely the long and tedious “owl watch” at the uranium separation facility as Oak Ridge (Site X). The Oak Ridge plants were not running until mid to late 1944; before then there was no “owl watch”. Based on these considerations, Lawson has commented that the document “post-dated September 1943” should really be dated 1944424.

Some commentators take the issue of the date of this memorandum rather lightly. Walker, speaking at the 40th anniversary reunion event, commented

The date of 1944 [as quoted by Oliphant in his 1967 memoir] is slightly at variance with that of “September 1943” handwritten in at the top of one of the memos on display [in the exhibition staged in conjuction with the symposium] and quoted in the paper written by Oliphant, Gooden and Hide425 fairly soon after the end of the War. However, it is not really surprising with the pressures under which people worked during the War that some differences in memory arose, and there is absolutely no doubt that the period 1943-44 saw the idea of the synchrotron, as it was later called, arise in Oliphant’s mind426.

We will return to the issue of the timing of the Oliphant Memorandum, and in the context of considering some related documents, propose an even later date.

The later history of the Oliphant Memorandum, following its creation, is not clear. Symonds has claimed that the existence of the Memorandum

424 Lawson in Rolph (1995), p. 24. 425 Oliphant, Gooden and Hide (1947). 426 Rolph (1995), p. 29. 228

was well-known around the laboratory in the early years of the construction of the machine427, and that senior staff made sure that everyone working on the project had seen it. This suggests that the Memorandum enjoyed an iconic status as the foundation of the enterprise being undertaken. This is despite the fact that, as we shall see, the machine described therein differed very substantially in technical detail from the machine then being built.

Thereafter it seems to have gone missing for at least two decades, maybe more. The copy reproduced in the 1993 Proceedings, and which must have once been in Oliphant’s possession, is most likely that reported in 1996 to be in the archives at the University of Birmingham (with another copy at CERN), placed there by Lawson428.

It can be argued that the document may have turned up and been annotated by Oliphant between 1965 and 1967, the first date being the year Oliphant provided McMillan with his January 1945 letter to Akers incorrectly claiming it was the missing memorandum (and so implying that at that time at least Oliphant did not have a copy of the Oliphant Memorandum itself), the second being the year by which Oliphant had stopped claiming the document had been written in 1943. However this is merely surmise at this stage.

The 1951/2 Mann Correspondence

Another retelling of Oliphant’s version of early events comes from a letter in late 1951, generated by requests from Wilfrid Mann, then working with the US National Bureau of Standards, for information which

427 John Symonds was an early member of the Birmingham synchrotron team. From transcript of interview with the author in 2010. EBA. 428 The two copies of the Oliphant Memorandum now available, one from the Proceedings, the other from the CERN archives, are identical in content, but differ in layout, the copy from the Proceedings having been repaginated to fit onto A4 pages when being photocopied. 229

would help him complete an epilogue to his monograph on the development of the cyclotron. Mann intended the new chapter to cover the conversion of the 184 inch cyclotron at Berkeley to a synchrocyclotron and it would therefore deal with the “phase stability” issue, which will be addressed here later.

Writing to Ernest Lawrence on 19 December 1951429, Mann reported reading the Oliphant, Gooden and Hide 1947 paper with its reference to the reputed 1943 proposal to Akers.

Not knowing how this might relate chronologically to Veksler’s430 paper or in other ways to Ed’s431 work, since I felt they must have discussed the problem during the war, I wrote to Oliphant asking him for a memorandum or other reference as I had been unable to locate any at the Division of Atomic Energy in London. I will enclose an extract of the letter which I have just received from him in reply.

Mann appeared to have enclosed the whole of the reply from Oliphant, written from Canberra on 11 December 1951432 . After a warm if quick greeting, he said

I don’t think I have ever made any claim for any authorship of the synchrocyclotron. The idea of frequency modulation in the ordinary cyclotron has been discussed for many years and the folk at Berkeley certainly deserve all the credit for having set to work to show that it actually produced the results expected.

429 Mann to Lawrence, 19 December 1951. Attachment K to letter from McMillan to Ellyard, op cit. 430 Vladimir Veksler (1907-1966), Russian physicist and accelerator pioneer. A co-discoverer of the “synchrotron principle”. For more information and an appreciation of his achievements see Rabinovich (1967). 431 i.e. Edwin McMillan 432 Oliphant to Mann, 11 December 1951. Attachment J to McMillan to Ellyard op. cit. . 230

Oliphant then recounted the early stages of his involvement with the conceptualisation of the proton synchrotron. The presentation of the chronology is strongly consistent with the 1967 exposition which it predates by a decade and a half. The recognisable elements of the story are here; that a letter went to Akers “late in 1943” proposing a post-war project; that the machine would involve “both rising magnetic field in an annular magnet and frequency modulation to match”; that only an elementary treatment was given of phase stability; that the magnet would be energized by a bank of condensers; that Akers had said no plans could be made at that time about what might happen after the war, though he thought it should be pursued.

Oliphant continued

Unfortunately, I have no copy of the original memorandum, since it was written when I was in America where I put the general idea to Lawrence from whom I received some encouragement. A copy was sent to the University of Birmingham and I believe it is now in the possession of Dr John Symonds. …. After the publication of MacMillan’s (sic) paper, I asked Akers to turn up the correspondence and memorandum, and though he remembered them perfectly he was unable to locate them. They may have turned up since.

Mann asked Lawrence to give his and Oliphant’s letters “confidential, limited circulation.”

I would be grateful for your comments which I would also treat as confidential and any suggestions as to how I should treat the matter in the “epilogue”. I will also write to Philip Moon to try to

231

get a copy of the letter from Dr Symonds. If it were not for the statement (re 1943) in the Oliphant, Gooden and Hide paper of 1947 all would be plain sailing but I do not wish to leave out any relevant references. Oliphant also mentions the fact that he discussed the matter with you, so your comments will be most helpful.

Lawrence replied to Mann on 11 January 1952, saying, inter alia

I have just seen the letter Ed McMillan wrote to you about Oliphant’s suggestion and his account gibes (sic) well with mine. I rather vaguely remember Oliphant telling me one time around 1943 that he had an idea of pulsing a current though some coils and at the same time pulsing a frequency modulated RF accelerator, and if I remember correctly, I commented at the time that the idea certainly looked feasible in principle. His idea seemed to be more closely akin to an air-cored synchrotron, though I don’t recall the utilisation of phase stability in achieving the acceleration. At least I did not appreciate the point at the time433.

Lawrence’s recollections, nearly a decade after the events referred to are interesting on several grounds. They indicate an early date (that is, a date during the war) for Oliphant to have worked out his ideas in sufficient detail to be able to discuss them with Lawrence. That the date recalled is 1943 can perhaps be attributed to confusion generated by the pressures of wartime activity with so many enterprises underway. It is possible of course that Oliphant already had some of the concepts in his mind at the time he moved to Berkeley late in 1943, though as we have

433 Lawrence to Mann, 11 January 1952. Attachment M to McMillan to Ellyard. op cit.

232

already argued, it is unlikely that he would have written them down at that time. Most importantly, the brief description he gives of Oliphant’s concept fits well, as we shall see, with that laid out in the Oliphant Memorandum.

Following the trail

We have devoted considerable space to the Oliphant Memorandum because of its reputation as containing Oliphant’s first thoughts on the new technology. But it is only one of a series of documents which appear to chronicle the development of the concept in Oliphant’s mind. One consists of a single page; another is a letter to Akers with technical details contained within it. In attempting to link these documents and to determine the order in which they were written, we are drawn to conclude that the Memorandum is the third of these, rather than the first, and that it was created as late as the spring of 1945. We will come later to consider the violence that such a judgement appears to do to Oliphant’s version of events.

At the end of November 1944, Oliphant, back for a time in Birmingham, wrote to Moon, on assignment at Los Alamos434. He reported that he was “now home, family flourishing but not so at university”. A big influx of students, especially older ones, was anticipated; as a consequence, a return to their university posts should be a priority. Oliphant himself was thinking of getting back to Birmingham by the end of March. “I realise

434 Oliphant to Moon 27 November 1944. File F20, Moon Papers, UBSC. This is a carbon copy only, therefore lacking attachments. This was clearly written from Birmingham as the reference is MLO/KH, KH being Oliphant’s long- time secretary. 233

that the whole TA job is approaching a stage where we can do very little to help the Americans”435.

He proceeded to give some clues about his plans once he was home for good.

I am hoping very much that the development of TA will not stop with the practical realisation of the present possibilities of nuclear chain reactions. I am convinced that a whole new world of nuclear physics is awaiting attention in the region of bombarding energies of the order of a thousand million electron volts. I believe also that such energies can now be realised in the laboratory. Of course large currents would not be available, and for the majority of purposes we should need very few particles so we can get them just as required. I am hoping that the authorities will see fit to encourage work in this region on the grounds that it represents one of the next regions of nuclear physics to be explored and perhaps one day to be exploited.

I am adding a note on one method which I think might produce particles with these cosmic ray energies.

The letter clearly indicated that as of late 1944, Oliphant was envisioning an accelerator able to be built in a laboratory, capable to generating particles of an unprecedented 1000 million electron volts and promising “a whole new world of nuclear physics”. A month later, Oliphant wrote again to Moon from Birmingham with more exciting news436. He had meet with Henry Tizard, who was not only a friend and colleague from

435 As we have seen, Oliphant’s feeling that time was running out for any real contribution by him to the bomb project was reiterated a few months later in a letter to Massey. See Oliphant to Massey, 21 March 1945. Box 7B, UBSC. 436 Oliphant to Moon, 28 December 1944. File F20, Moon Papers, UBSC. 234

the days of radar research, but also well-connected with the Nuffield Foundation437.

I think you will like to know that Tizard believes he will be able to find for us from the Nuffield Foundation about £5000 a year for 10 years to provide three Fellowships and the necessary workshop and technical assistance. This is contingent on us carrying out fundamental research and is designed to prevent us becoming completely submerged in the practical aspects of the TA problem. He is especially anxious that we develop new methods of research in new fields and was delighted to hear that we have been giving some thought to methods of accelerating particles to very high energies and to the type of research that might be possible with artificial cosmic rays. Tizard is confident that he can find the money and later, if we need it, a capital sum to build equipment. The Foundation will not meet again till March and he wants to have by then very clear ideas about our program. I am sure that you will be as greatly encouraged as I am that he should think of our laboratory as carrying forward the standard of nuclear physics in our country.

The letter contains some clues as to the methods Oliphant had in mind.

[I am] confident that pulse methods of operation offer possibilities of avoiding the huge and expensive equipment which is popular for producing particles of high energy. To be successful we will have the need for one or two real experts in [radio-frequency methods].

437 Clark (1972). 235

To wrap up that element of the vision, Oliphant proposed that he and Moon get together to discuss details and to agree on “who we shall invite” (presumably to fill the anticipated new Nuffield Fellowships).

This correspondence, whose dating cannot be doubted, establishes that by November 1944, Oliphant had conceived the accelerator project, set its objectives and written an account of a possible solution to at least some of the technical challenges. Already in these letters we can see the shape of Oliphant’s vision, the elements of which he would repeat in various formulations in letters and submissions over the next year. We can summarise those elements follows.

 Fundamental research into nuclear physics must continue, even as efforts were being made to further exploit the understanding of nuclear processes revealed by earlier research.  The bombarding of targets with energetic particles would be a central technique, as it had been for over 20 years, but new methods were needed that could accelerate particles to energies that were an order of magnitude higher than anything currently being reached, energies high enough for the particles to be dubbed “artificial cosmic rays”.  It was now feasible to reach such energies with laboratory- scale apparatus.  New techniques, such as those using pulses of radio frequency energy, could reduce the size and cost of the necessary equipment well below what might be envisaged simply by scaling up existing techniques.  The appropriate source of funding for such enterprises was the public purse (“the authorities”), though private sources

236

such as the Nuffield Foundation could provide significant assistance.

Is it possible to identify the “note” that Oliphant had drafted describing “one possible method” and which he planned to send to Moon (later evidence indicates that he did not in fact include it)? There is certainly a candidate. A file held in the University of Birmingham Special Collections438 contains, inter alia, a one-page typed document entitled An Impulsive Method for the Production of Particles with very high Energies. The document carried Oliphant’s name at the bottom (though there is no signature) together with a footnote that it was written from the Nuffield Laboratory. This document will be referred to hereafter as the “one-pager” in the absence of any other suitable title.

The document was not dated, so we must rely on other evidence to assign a date to it. The final paragraph stated “it is my hope that the proposed system can be developed in the laboratory once peace returns (author’s emphasis)”, clearly placing it in the period of the war. The document having been written from Birmingham limits the date either to the period mid-November 1944 to mid-January 1945, during which Oliphant was in residence in Birmingham, or to the period from late March 1945 once he had returned permanently from the US. The brief and relatively rudimentary treatment given in the document suggests the first period. Well before March, we shall see, Oliphant had developed more elaborate and somewhat different ideas, contained in a letter to Akers in January 1945.

In addition we have, as noted above, a reference to “a note on one method” in his letter to Moon, written at the end of November 1944.

438 Oliphant: Notes on production of high energy particles, c 1944/45. UBSC. This is the notation on the whole file, rather on the particular document in question. It is not known who made the notation, but it is not in Oliphant’s hand. 237

That description fits better with this short account than with the much longer Oliphant Memorandum. Further the term “impulsive method” did not occur other than in this document. It appears reasonable therefore to suggest that this was the document to which his note to Moon referred, and that it was written in the second half of November 1944. We can also surmise that being written so soon after his return, with many matters needing his attention, the document was based on ideas formulated a little earlier, that is, while still in the US, perhaps at Site X. That fits with Oliphant’s own “owl watch” recollections.

A hand-written draft of this document also survives in the same file. It is very similar to the final version, with only a few variations, especially at the beginning, though it lacks the final paragraph referring to "when peace returns”. Both the draft and the final version cover the same ground, so their common contents can be summarised as follows, in terms that will become increasingly familiar.

 Progress in nuclear physics depends on giving bombarding particles energies similar to those carried by cosmic rays.  Reaching such energies with existing techniques such as the cyclotron and the betatron will require massive and expensive equipment.  A new method, combining some features of the cyclotron and the betatron, may provide an economic method for production of short pulse of very energetic particles.  The new method constrains the particles to an orbit of approximately constant radius by increasing the strength of the guiding magnetic field as the particles gain energy and therefore speed.  At the same time, the frequency of the accelerating electromagnetic field, applied across equidistant gaps between hollow electrodes,

238

akin to a multi-segment cyclotron, must be increased to keep the applied potential in step with the particles as they pick up speed.  The magnetic field would be supplied by coils without iron, largely removing limits on size.  Methods of varying the frequency of the accelerating field and the strength of the magnetic field are well known.  The use of pulse techniques (that is, rather than continuous operation) allows for a reasonable average level of energy dissipation, and the use of higher accelerating voltages.  There appears to be no reason why energies of 1000 MV should not be produced by this method.  The method will be applicable to protons and heavier ions more readily than to electrons, since the light particles very rapidly approach the speed of light.

The draft contained some points that were not included in the final version.

 Some detail was provided of the origins of the cyclotron (begun by Lawrence) and the betatron (by Kerst and the General Electric research laboratory).  The current methods are limited for practical reasons to around 100 MV.  Extending the current methods to the desired higher energies would result in equipment so massive and expensive that few laboratories would be able to afford it. The use of the new methods would result in equipment that would be considerably smaller.  Particles should be fired into the system at a relatively high initial velocity (though the reason for this is not discussed).

239

A diagram in the draft is not reproduced in the final version; this shows the variation with time of the frequency of the accelerating voltage and the strength of the magnetic field. The latter rose almost linearly, while the former rose quickly initially and then flattened out. It is likely that this diagram referred to the acceleration of electrons rather than protons since, as was noted immediately above the diagram, electrons rapidly approach the speed of light. A constant velocity requires a constant frequency of the accelerating voltage (as in the cyclotron).

It appears highly likely that these documents represent Oliphant’s earliest thoughts on his new technology, committed to paper a matter of weeks after they came to him. If this is so, his proposal to accelerate protons by this method, though briefly sketched, was two years in advance of any other, the next not coming till November 1946. He was also more than half a year ahead of the more general and theoretical considerations of McMillan. He was roughly contemporary with Veksler, though the Russian’s paper was not available in English until more than six months later. We will consider the contributions of McMillan and Veksler below.

Oliphant to Akers 17 January 1945

The second document in the series (“second” in that we are considering it second, but we shall also be arguing that it comes second in time), was a letter from Oliphant to Akers dated 17 January 1945. This letter poses some problems. The first is the provenance of the documentary evidence and therefore the confidence which we may place in the letter. No copy is available through the official record. What we have to work with is not an original, but a retyped document, presumably from an original now lost. It is labeled (COPY) at the top. It carries a date but no address.

240

Copies of this document have come from two sources439. One copy was found among Oliphant’s papers when I was researching his biography in 1980; the other came from Edwin McMillan later in 1980 in response to my enquiries. According to McMillan, and as noted above, this came to light in 1965, apparently in response to repeated requests to Oliphant by McMillan. McMillan described the circumstances under which he received a copy in his 1980 letter to the author440, a shorter extract of which has been quoted above.

In 1965 came a copy of the famous letter to Akers. Oliphant had visited Berkeley, we discussed the matter again and on his return to Australia Oliphant sent me a letter, dated 8th October 1965 and with it a copy of the Akers letter and of a letter to the Archivist at the U.K.A.E.A. These letters…. do not really solve the mystery of the 1943 memo, for the memo to Akers is dated January 17, 1945. Also what I was sent was not a photocopy but a typed copy, on the same watermarked paper as the covering letter, so that it lacked any identifying marks such as one might expect on an historical document.

What then are we to make of this letter? If it is genuine, it appears to be based on an original copy that perhaps still existed in 1965, presumably among Oliphant’s papers in Canberra, and presumably in such poor condition that it had to be retyped. The whereabouts of the original are unknown, as are its location and movement for the previous 20 years. In the covering letter to McMillan441, Oliphant wrote

439 Oliphant to Akers, 17 January 1945. Copy in EBA, also as Attachment G to McMillan to Ellyard, op.cit. 440 McMillan to Ellyard op. cit. 441 Oliphant to McMillan, 8 October 1965. Attachment N to McMillan to Ellyard, op. cit. 241

Herewith a copy of my original letter to Akers, who was at the time the Chairman of the “Tube Alloys” project in U.K. In private life he was a member of the Board of Imperial Chemical Industries, responsible for research and development, so you will appreciate that the letter was written to an intelligent chemist and not to a physicist.

Akers’ reply was non-committal, merely suggesting that I raise the matter again after the war.

Oliphant’s assertions quoted above give rise to some confusion. As McMillan has already pointed out, Oliphant appears to be maintaining that the letter now being provided is the first communication on the matter to Akers, despite it being dated January 1945. This appears to be at variance with his other claims that the first communication to Akers was created in 1944 or even 1943.

Oliphant’s letter to the U.K.A.E.A archivist442, sent the same day as the letter to McMillan, was more extensive.

On 17th January 1945, I wrote a letter to Akers concerning plans of mine for a new type of particle accelerator which I proposed should be built after the war. A copy of the letter is attached. This letter was written from the Liaison Office in Washington, where a copy was filed. Some weeks later a reply came from Akers, and this too was filed there.

Thus in the records of the British project, there should be two copies of each of these letters. In 1946, I requested confirmation of the existence of these letters, but they could not be found. I

442Oliphant to UKAEA Archivist, 8 October 1965. Attachment N to McMillan Ellyard, op. cit. 242

imagine that the papers are now in far better order then they were at that time, and I wonder if they can be traced? They are of some interest as records of an early proposal for a proton synchrotron. Dr E McMillan, Director of the Lawrence Radiation Laboratory, is anxious to have as much historical material about this subject as possible and if they are found he would appreciate copies.

I wrote again after my return to U.K, and though subsequent correspondence is less important, it would be helpful if I could have copies of anything referring to this project which may be traced.

Sir James Chadwick tells me that many documents of this time are missing. Is it possible that they remain within D.S.I.R. or in Washington?

With respect to the comment which Oliphant assigned to Chadwick, the following quotation from a letter from Chadwick to Oliphant in December 1965443 (two months after the letter to the Archivist) is relevant.

In a letter you wrote to me from Berkeley you mentioned a letter of yours to Akers in January 1945 in which you discussed a scheme for a new accelerator. I have looked for this in my papers. I cannot find it but I am going to go through all of them again. Only some of my papers were sent to me and many interesting ones are missing. I suspect that they were destroyed, for Penny tells me there are none in Washington that survive from

443 Chadwick to Oliphant, 13 December 1965. EBA. 243

my office. What happened to the documents in Tube Alloys office in London I have never been able to discover. 444

The correspondence that Oliphant generated “after my return to UK” was presumably the letter and submission of 16 June 1945 which we are yet to consider. It is clear that at the time of writing to the UKAEA archivist Oliphant did not have a copy of that, though he apparently had a copy of the 17 January 1945 letter. He was provided with a copy of the June 1945 document a few weeks later when the archivist replied445, though we do not currently have a copy of her reply. In a letter to McMillan dated 24 November 1965 and reporting her response, Oliphant does not refer to any other (i.e. earlier) correspondence, implying that it had not been found.

So in seeking to assess the import of the 17 January 1945 letter, we are left only with copies of uncertain provenance446. We must look therefore to internal evidence as a guide to its veracity. This too raises issues, including significant differences in the details, particularly as regards the proposed technology, between this letter and the Oliphant Memorandum. We can however suggest that the January 1945 letter is in fact the communication which Oliphant variously claimed to have written in 1943 or 1944. In that case, Oliphant would have been right in

444 This implied deliberate destruction of part of the Tube Alloys documentary record has serious implications for historians involved in reconstructing the broader history of TA, as well as for the task being undertaken here. It may possible to fill in some of the gaps by closer study of such of the Akers/Chadwick correspondence as survives, such as in the Chadwick papers at Churchill College, though Chadwick’s remarks do not encourage optimism in that regard. Such search has not been undertaken. 445 The archivist at the time was Margaret Gowing, later to write the authoritative history of the British atomic energy program, and to become Professor of the History of Science at Oxford University. See Gowing ( 1964). 446 No obvious avenues are open in attempting to provide a more substantial provenance for this key document. As noted, no copy is held in the most obvious place, the public record as represented by the UK National Archives AB series, nor in Birmingham (not that that would be expected given that the letter was apparently written in the USA). The two surviving copies are both traceable to Oliphant himself, since it was he who provided Macmillan with a copy in 1965. It may be speculated that in 1965 even Oliphant himself did not have a copy, and instead had the current document typed up from suriving hand-written notes. 244

his assertion to McMillan that it constituted his first communication to Akers on the matter. The argument depends both on timing and on the content of the documents.

If we are right in dating the first writing down of the concept ton late November 1944, and if that resulted in Oliphant sending a note to Akers, it seems highly unlikely that he would have forgotten that a mere two months later, as the January 1945 letter implied. Likewise the notable differences in the technology proposed counts against the possibility of two communications to Akers on the matter within two months.

The 17 January 1945 letter began as follows.

We are extremely busy here, as you can imagine. However there are periods of boredom when one is watching or waiting for results, or when travelling, and I have filled some of these with speculation about the future of nuclear physics in England. It seems to me that further advances in the field of nuclear energy are likely to stem from more detailed study of the nucleus and the properties of the particles of which it appears to be composed. This new information can be obtained in two ways. Firstly by more intensive, and possibly more intelligent, investigations of nuclear reactions at the energies now available from cyclotrons and high voltage equipment. Secondly it may come from observations at much higher energies, such as that of the hard component of cosmic radiation, where entirely new phenomena are likely to be found. It is difficult to know which approach is likely to be most profitable. However there are plenty of laboratories working at the lower energies, and I would put my money on attempts to produce bombarding beams with energies of a different order of magnitude from those employed so far.

245

It seems fair to say that this and what follows reads like a letter written at the time and from the US, and that it reflects Oliphant’s later recollections of the circumstances of how the concept of the synchrotron arose. We have many references to “here”, meaning locations in America, and the talk of “boredom while one is watching and waiting for results and while travelling” could refer to the “owl watch” and to the trips Oliphant frequently made between Berkeley, Oak Ridge and Britain. In establishing the veracity of this document none of this is conclusive but it is suggestive. The most curious aspect of the opening is that it makes no reference to any earlier correspondence. It reads as if it was the first time Oliphant had raised the matter with Akers. This supports our contention that such was indeed the case.

In general, the rationale laid out in the opening paragraph mirrored that we shall find in the Oliphant Memorandum, namely the need for the controlled production of very high energy bombarding particles to advance knowledge of nuclear physics and thereby enhance the production of nuclear energy. We are presented with a contrast between what knowledge might be gained by continuing to work at the currently accessible energies, and what new insights might arise from working with energies comparable to those found in cosmic radiation.

Oliphant went on to state that, in his opinion, “cyclic methods of acceleration, akin to the cyclotron” would be able to reach the extreme energies proposed, but that he doubted that the cyclotron as presently constituted would be able to do the job, due to the “relativistic change of mass with velocity, and hence the increased time of revolution in a uniform magnetic field”.

246

Thomas, and I understand, also Bethe, have proposed that this limitation might be removed by introducing extra focusing, which I do not understand completely, due to undulations in the magnetic field. This form of focusing is claimed to be sufficient to enable the magnetic field to be increased with radius to compensate for the increase in mass.

But a problem remained.

Now, I do not think that it will be very rewarding to aim for an energy less than 1000 million electron-volts (the mass energy of the proton or neutron), if interesting new phenomena are to be observed. Even if the relativistic increase in mass can be compensated, the pole diameter for a cyclotron at this energy would be about 30 ft, and the mass some 80 000 tons.

Oliphant backed this statement with some calculations, based on target energy of 940 MeV and a maximum magnetic field strength of 12 500 gauss. This yielded an orbit radius of about 5 metres. He concluded “so I assume that the building of such a cyclotron would be beyond us in England”. That statement would seem beyond debate. With as much steel needed for the machine as for a battleship, it is likely such a project would be beyond anyone. Having cleared the ground, Oliphant proceeded to expound his vision for “… a different kind of accelerator which I believe will work and which will give us protons of 109 electron- volts or, with unlimited funds, even more”.

The letter, relatively short as it is, contained a large amount of technical information. Oliphant referred to the new device as being “rather like a ring cut from the outer edge of a cyclotron”, a useful image not used in other expositions. A guiding magnetic field would rise in strength as the

247

particles gain energy, so constraining them to an orbit of constant radius within the annular magnet. He spoke of an oval-shaped vacuum chamber (within which the particles would orbit) of glass or ceramic, with an internal conducting surface, part of which would form the accelerating electrodes, as in a cyclotron.

A key issue addressed for the first time here was the manner in which Oliphant planned to inject particles into the machine, and the injection energy that could be achieved. He proposed to send particles into the accelerator at 500 KeV from a Cockcroft-Walton apparatus. As the particles were accelerated to a final 1000 MeV, their velocity would increase 30-fold. To keep pace, the alternating voltage to accelerate the particles would have to rise in frequency in a similar 30 to 1 ratio, from 300 kilocycles to 10 megacycles. This variation might be obtained by varying an inductance or capacitance associated with an oscillator. He sketched a possible technical solution to this challenge, admitting that it “sounds a bit tricky and I have no doubt there would be problems of development, but I am sure it is correct in principle”. As we shall see this matter was not to be lightly dismissed.

The injection energy affected how long the particles must be accelerated. He proposed a period of one second; the particles would make some 2 million revolutions during this time, gaining 50 KeV in each orbit. That requirement had affected the choice of magnetic field source. With such a rise time for the field, an iron-cored magnet was possible, laminated to reduce eddy currents. Oliphant was setting aside his earlier thought about using an iron-free magnet. We have previously noted his interest in iron-free magnets in his alternative designs for A-M plants; perhaps now he was concerned about the difficulties in designing such a system.

248

Oliphant was not relying solely on his own judgement here. Earlier in the letter he wrote “I might say that I have discussed this with Lawrence on two occasions. He does not commit himself, but he says he can see no reason why it should not work. Of course he is very occupied with other things -- as no doubt I should be also -- and it is clear he has given the scheme little attention. But I do not think he would let it pass if there were a fundamental flaw in the argument. His instinct in these matters is uncanny.”

The underlying purpose of the letter is clear from the two concluding paragraphs. Oliphant is after money to make his dream real. The sum needed was substantial.

I estimate the cost of building such an accelerator, at current prices of steel and equipment, to be not more than £200 000, exclusive of buildings. This may sound a lot, in terms of accepted expenditures of physical equipment in England, but it is very small compared with the costs of the present project.

I would be grateful if you would let me know whether TA would support this proposal as something that we, in England, might try to do when the war is over? By building such an accelerator, we could continue to contribute to basic physics, as in the past. I doubt whether we can ever emulate America in building and operating the extremely costly plant required for the development of nuclear energy. We can hope to make a real impact upon the basic knowledge required if we are bold and take steps such as I am advocating. This might appeal to Anderson447, as there would

447 Sir John Anderson, Chancellor of the Exchequer in the UK Government, with Ministerial oversight of the atomic energy (TA) program, 249

be no problem of American co-operation or of use of American know-how which we have acquired here.

The reference to the likely interest of (Sir John) Anderson in the proposal is puzzling on the following ground: Oliphant had had an hour- long meeting with Anderson in London on 9 January 1945 barely a week before. The surviving record of that meeting does not contain any reference to this matter. Given that, in his 1967 account, Oliphant claimed to have discussed the matter with Anderson, why did he not do so on that occasion?

Dating the Oliphant Memorandum

Before we examine the content of this important document, we need first to consider its timing and its relationship to the 17 January letter. As difficult as it is to do so, it appears we must set aside the view, reinforced in various ways by Oliphant himself, that the Memorandum was written in 1943 or 1944, and that it constituted his first approach to Akers on the matter, thereby preceding the January 1945 letter. We can instead suggest that neither of these statements is true. An alternative chronology, not itself without difficulty, is as follows: following the one page document, which we have established beyond reasonable doubt was written in late November 1944, and was in all probability the first written statement of the concept, Oliphant then wrote the 17 January 1945 letter. At some time later he wrote the Memorandum, which would place its creation in the spring of 1945.

Given that the Memorandum is undated (beyond Oliphant’s later annotation), what evidence can we bring to support this alternative view? As we shall see the Memorandum was all but identical to the proposal Oliphant put before Akers in June 1945 (and which is on the

250

public record). On the other hand, it is noticeably different in key respects from the proposal put forward in January. Given the contrast between the detail provided in the Memorandum and the simplicity of the one-pager, it is obvious the latter came first. So the Memorandum cannot have been written before late November 1944.

We now have two possibilities. The first is that Oliphant wrote the Memorandum, and then within a matter of weeks, wrote the January letter, with its alternative approach, returning six months later to the Memorandum as the basis of his final proposal in June. The second is that he wrote the January letter as a first elaboration of the one-page document. Over coming months, he refined the concept into the Oliphant Memorandum, leaving behind many of the ideas expressed in the January letter. With a few small amendments the Memorandum became the basis of the proposal to Akers in June.

Prima facie, the second chronology seems more plausible, and there is other support we can adduce for it. As we have noted, the January 1945 letter reads as if it were the first time Oliphant had raised the proposal with Akers. There is no reference to any earlier communication, and certainly not to one written little more than a month before. On our new chronology, the omission is not surprising since there was no earlier communication. The January 1945 letter was, as Oliphant had claimed to McMillan in 1965, his first such proposal to Akers.

Furthermore we can see in other materials evidence of Oliphant developing the key difference between the January proposal and that in the Memorandum. This was the notion of injecting the particles into his new accelerator at much higher energies using a cyclotron. The materials in question consist of pages of hand-written notes and

251

calculations contained in the file which held, inter alia, the one-pager448. These papers are not numbered or dated; we do not therefore know how they related in time to the one-pager. It is however likely that they concern the period around the late winter or early spring of 1945.

Oliphant was clearly concerned with the issue of varying the frequency of the accelerating voltage over a substantial range in the very short period of acceleration. He may have said that methods for doing this were well known, but he still wanted to contain the problem. One option was therefore to inject the particles into the system at as high an energy as possible; the higher the initial energy, the higher the initial accelerating frequency, and the smaller the variation in frequency that would be needed to accommodate the final energy.

His solution was to use a cyclotron as the injector. On one page in the notes449, we read “suppose we use an ordinary (sic) cyclotron to get up to 10 MV then the frequency variation will not be excessive. We would then surround the cyclotron with the new system if we can get the particles transferred without great difficulty…. perhaps by two electric deflections.” Oliphant then illustrated the idea with a simple but remarkable pen-drawn diagram. This showed the outline of the two dees

448 It appears the file once contained a variety of papers, now mostly missing. According to a notation scrawled in Oliphant’s hand on the front of the envelope which once held the papers, the contents were “Misc. letters etc”, including correspondence with Chadwick (“written to me at X and Berkeley”) and with Tizard, some material on a conversation with Groves which had been “handed to Dale”, and material on Y-12 sent to Dawton (?). There are certainly a number of pages in Oliphant’s hand dealing with the design and operation of an EM plant, but none of the other materials on the list are in the file. The materials we have been considering are not on Oliphant’s list and must have been added later. The cover sheet has a notation, not in Oliphant’s hand, as follows; “Oliphant. Typescript, MS notes re production of high energy particles, prob c 1944/5. To do with synchrotron?” 449 This page of the notes presents something of a puzzle. The style of paper, ink and penmanship are similar to those of draft of the one-pager (and unlike any other pages of the notes). A first thought was that it represented possible content for the one-pager not included in the final typescript. However the level of detail does seem consistent with the rest of the draft, and it is my view that it therefore represents a later stage in the development of Oliphant’s ideas. 252

of a cyclotron, with a beam of particles leaving the cyclotron and being deflected into an orbit around it.

Other calculations made the matter clearer. Injecting at 5 MeV would mean only a ten-fold variation; at 20 MeV it would be less than five to one. We can put a date on some of the musing by reference to a notebook in Oliphant’s hand held in the Barr-Smith Collection450. An entry dated 15 May 1945451 reads “let us consider the possibility of injecting protons from the cyclotron at 19 MeV... or thereabouts”. The frequency variation required was only 4.5 to 1.

The Oliphant Memorandum.

It is time to consider the content of this document in some detail, not because it was the first statement of Oliphant’s accelerator vision (as we have noted it was almost certainly not) but because it was essentially the basis on which Oliphant sought funding from the public purse. It was also, even with this now-proposed later date of creation, more than a year ahead of any other detailed proposal to use the new technology to accelerate protons and ahead by a number of months of any proposal for a “synchrotron” of any kind.

In its original layout the document ran to six pages. It began by setting a context, in large measure an elaboration of ideas already expressed.

The properties of elementary particles with energies of 10 or 20 million electron volts (10-20 MeV) have been investigated extensively during recent years. In the war period, work has been intensified in some directions though it has lapsed in others. A

450 Series 10, Oliphant Papers, Barr Smith Library, University of Adelaide. 451 “The” cyclotron was the still incomplete Nuffield Cyclotron. 253

great deal of work remains to be done, but it is probable that the main outlines of nuclear physics in this region are now clear. In any event there is ample equipment in existence in the USA and elsewhere to fill in most of the gaps in our knowledge, while much will be accomplished in the government laboratories that will be set up in various countries to exploit the possibilities of nuclear fission.

The greatest hope for an increase in fundamental understanding lies in experiments at energies above 1000 MeV. Cosmic radiation offers a source of particles with energies in this region, or higher, but, due to the low intensity and the uncertainties about the nature of any individual particle, there are difficulties in the proper interpretation of experimental results. Investigators in this field of physics have shown remarkable ingenuity and patience and very striking results have been obtained. However the rate of progress would be greatly accelerated and obscure points much more easily settled, if there were available a method for accelerating particles of a known kind to known energies in this region. It is certain that new and important phenomena452 would be discovered because of the higher intensity and the freedom from obscurity as to the time and energy of the bombarding particles, while knowledge of the fundamental properties of these primary particles would reflect on the whole of nuclear physics.

Here, Oliphant stated his fundamental motivation. Major advances in nuclear physics knowledge depended on greatly increasing the energy

452 It is not clear what new sort of “new and important phenomena” Oliphant had in mind in this reference. He appeared to generalizing from the experience of the previous decade or more, when each new realm of beam energy had opened major new lines or enquiry, and adding the expectation that controlling energies equivalent to those of cosmic rays would progress the field even more (it had of course been through cosmic rays studies that particles like positrons and mesons had first been discovered). 254

of the bombarding particles available for experiment. The only currently available source of particles of such energies, cosmic radiation, suffered from sufficient uncertainty as to the nature and energy of the radiation as to greatly complicate the interpretation of experimental results. Hence, some new method was required to accelerate particles of known characteristics to clearly defined energies in the cosmic ray realm. Were any of the current types of equipment capable of this task? Oliphant thought not.

The induction accelerator, or betatron, undoubtedly affords the simplest system yet devised for the acceleration of electrons to energy as high as 100 MeV. However, both mass and cost of equipment to reach 1000 MeV are prohibitive. An examination of the possibility of producing pulses of particles by “coreless” betatrons, using very large currents from charged condensers or from special short-circuit machines that are used by Kapitza, shows that this approach involves very large scale equipment and presents formidable problems of engineering. Similarly, the extension of the cyclotron principle to extreme energies appears to be prohibitive in cost and to be difficult of solution when the particles have acquired velocities close to that of light.

In what follows we describe a method employing some of the principles of each of these systems, which is reasonable in cost and which can be applied in a modern laboratory.

Under the heading The new system, Oliphant proceeded set out the basic principle of his proposed accelerator, indicating how it would draw on and then surpass what is offered by the betatron and cyclotron.

255

The essential feature of the proposal is that the particles should be constrained to move in a circle of constant radius, thus enabling the use of an annular magnetic field of the correct form but over a total volume which is small enough to require only minor power for its excitation. The magnetic field would be varied in such a way that the radius of curvature remains constant as the particles gain energy through successive accelerations by an alternating electric field applied between coaxial hollow electrodes as in the cyclotron. The varying magnetic field performs the function of the guiding field of the betatron, but the acceleration is provided by an applied potential rather than by a changing flux. In this way, it is possible to apply much higher accelerations per revolution. The changing magnetic field can be produced by an application of modern pulse technique, while the accelerating potential can be provided by the same general method. Essentially very large powers are available during the accelerations of a single burst of particles, a relatively long quiescent period reducing the average power consumed to a reasonable value.

With respect to the path being taken by the particles under acceleration, Oliphant’s intention was to confine them to an orbit of constant size, therefore requiring only a ring-shaped magnetic field for guidance. This is in great contrast to the spiral path covered by particles being accelerated in a cyclotron which required the guiding magnet to generate a field throughout a cylindrical volume. Though Oliphant does not say so here, the mass and cost of such a solid magnet would increase at a rate between the square and the cube of the diameter of magnet, while the cost of a ring-shaped magnet would increase only linearly with its diameter, or perhaps a little more. These considerations

256

underlie his assertion that extending the principle of the cyclotron to extreme energies would involve prohibitive cost.

With regard to the acceleration of the particles, Oliphant’s proposed machine differed from the betatron, which supplied additional energy to the particles by increasing the strength of the guiding magnetic field. Oliphant on the other hand proposed to use an alternating electric field and cylindrical electrodes through which the particles passed, as was done in the cyclotron.

In summary, the essential features of the proposal were

 Confining the particles to a circle of constant radius so that an annular ring of magnetic field could be used in place of the full disc field required for the cyclotron. This would substantially cut both the weight of the magnet and the energy needed to excite it.  Varying the strength of the magnetic field as the particles were accelerated to maintain constant radius of curvature, so confining the particles to a narrow annulus.  Accelerating the particles by repeated passage through an alternating electric field, as in the cyclotron.  Using pulse techniques of the kind developed for radar to generate both the rising magnetic field to control the orbit size and the alternating electric field to accelerate the particles.  Restricting the total power requirement through pulsed operation, with bursts of particles with a high power demand during acceleration being alternated with relatively long quiescent periods.

257

 Shaping the profile of the magnetic field (that is, its variation with radius) in order to reduce oscillations in both the radial and axial direction.

In the Memorandum, Oliphant considered both electrons and protons as the particles being accelerated, dealing first with the simpler case of electrons. Even at the relatively low energies, around 1 million eV, at which the electrons might be injected into the accelerator, the particles would already be close to the speed of light. During acceleration, therefore, their velocity would increase only a little, making their period of revolution almost constant and allowing the frequency of the accelerating field to remain at one value. The slight change in velocity could be compensated for by a small change in the magnetic field.

An additional challenge arose in attempts to accelerate protons by this method. They would experience a much greater change in velocity, and therefore it would be necessary to substantially vary the frequency of the alternating electric field. Here his argument was quite different from that made in the 17 January letter. Rather than injecting at 500 KeV from a Cockcroft-Walton apparatus, he was now proposing to use a cyclotron (presumably the still-incomplete 60 inch Nuffield cyclotron) as the injector. It was designed to supply protons at 45 MeV, which would be moving at 30% of the speed of light (0.3c) as they entered the accelerator. At the hoped-for final energy of 1200 MeV, their speed would be 0.9c, which meant the accelerating field had to increase in frequency by a ratio of 3 to 1 (from around 3.6 MHz to 11 MHz) to keep up with them. This was a much easier task than the 30 to 1 shift briefly spoken about in the January letter, but Oliphant doubted that such a change was possible (in the 100th of a second he was envisioning as the accelerating period) by purely electronic means. He was already thinking of mechanical tuning methods.

258

Oliphant examined the use of both iron-cored and air-cored magnets, deciding, for the moment at least, to use the air-cored type. This is a return to one of the few technical details in the one-pager, and a repudiation of the choice made in January. The major consideration was cost, since calculations showed that some 360 tonnes of steel, laminated to reduce eddy currents resulting from the required rapid change of magnetic field, would be required, “a formidable amount of high-grade material”. An iron-cored magnet would take less energy to generate the same field strength (due to less leakage) but not enough of a reduction to make it preferable.

Oliphant admitted that the design of an air-cored magnet (essentially a system of coils carrying current) had not yet been done. He did have an order of magnitude estimate of how much electrical power would be needed to excite it; the momentary power would be of the order of a million kilowatts (produced by, say, a current of 45 000 amperes at 250 000 volts. The numbers did not seem to faze Oliphant. “This is large” he wrote, “but by no means impossible.”

To supply the energy, Oliphant proposed either discharging a bank of condensers or short-circuiting an alternator. The January letter had considered only an alternator with a flywheel. For reasons of flexibility, such as to be able to vary how long the field would take to reach its peak, he preferred the first option. Either way, it would be expensive. “The condensers or machine required to produce the magnetic field are by far the biggest items of equipment with which we are faced.”

The particles would be accelerated as they crossed gaps between electrodes, across which an alternating voltage would be applied at the proper frequency. Oliphant supplied diagrams showing electrode

259

arrangements with 2, 4 or 8 gaps, more gaps meaning a lower voltage across each -- 250 000 volts for 2 gaps, 125 000 volts for four gaps – but also a higher frequency of oscillation. Producing the needed voltages and frequencies would be “especially easy with pulsed operation”.

Some issues were not considered. There was no discussion of the necessary vacuum system, perhaps on the grounds that it was well- known technology. In proposing a technique to accelerate electrons to 1000 MeV, Oliphant made no reference to the limitations imposed by synchrotron radiation. Perhaps he was unaware of them. The memorandum also made no reference to costs or to seeking public support for the venture. Presumably such matters would have been in a covering letter, if one was written. They were raised in the 17 January letter. However, as far as can be ascertained, the public record contains no copy of such a letter, nor of the reply Akers was supposed to have sent, as indeed it contains no copy of the Oliphant Memorandum itself. This raises the possibility that there was no such letter, and that Oliphant in his later recollections confused it with his letter to Akers of January 1945. The challenge to that hypothesis comes from the fact that, as we have seen, Oliphant was making references to a 1943 communication with Akers as early as 1946.

Developments in early 1945

Through the early months of 1945, Oliphant was busy with many matters, including his imminent departure from Berkeley and return to Birmingham. In his 1967 memorandum, Oliphant wrote

For some months [after the letter to Akers] I was preoccupied with Tube Alloys business, spending much time in discussion of

260

the future programs for nuclear energy in Britain and in the search for a suitable site for an Atomic Energy Research Establishment. But I continued to pester Akers and anyone else in authority whose attention I could obtain, including Sir John Anderson453.

Among his correspondents was Edward Appleton, a former Cavendish colleague, now in charge of the Department of Science and Industrial Research (DSIR) within which was housed the TA effort under Akers. Writing to Appleton from Berkeley on 14 February 1945454, with the letter copied to Chadwick and Akers, Oliphant announced that he would be returning to the UK at the end of March. He needed to be available at least half time to his University. As a consequence the nature and extent of his ongoing involvement with TA required discussion.

He also declared, if in veiled terms, his longer-range plans.

As soon as war conditions permit, I intend to start work again on nuclear physics of a type which is not immediately applicable to the TA problem. I believe this is essential if we are to retain in U.K. a proper appreciation of the advancing aspects of the subject, and if we are to train, for TA, the right type of student.

This means that the whole of the facilities of the laboratory will no longer be available to immediate TA work, though I am prepared, of course, to do my part in providing accommodation and facilities for some portion of the experimental effort, until such time as our central laboratory is ready455.

453 Oliphant (1967) Part 2. 454 Oliphant to Appleton 14 February 1945. AB 1/42, UKNA. 455 This is a reference to the proposed national “experimental establishment”; that is, what became Harwell. 261

We see here a point to which we will return when discussing of the funding of projects like Oliphant’s. He had already acknowledged that such an enterprise could not proceed without substantial government investment. To the original rationale, that of advancing fundamental knowledge of nuclear physics in the search for more efficient ways of unlocking nuclear energy, he was here adding a second, the need to train the new generation of nuclear physicists by exposing them to the advanced forms of equipment. As we shall see, the second justification outweighed the first in many eyes.

He went on to announce himself impressed by the resources and facilities that the Americans would be making available for industrial research after the war, citing as an example the General Electric engineering plant at Schenectady. There the work force of 600 (“in its swollen wartime state”) would become 3000. Westinghouse and other firms were to undergo similar expansion over their wartime capacity. This he contrasted starkly with the situation in the U.K. as he saw it. “It is necessary for us to realise that we can not retain any ability to hold our own in world markets unless the inadequate provision and plans I heard discussed by [Metropolitan Vickers] and other firms, and by those who are planning the government effort, are expanded many times.”

A personal note also intruded. “I have had three offers of employment at a much greater salary than any available in U.K. Several of my team are very anxious to take such offers and to remain here after the war.”

Appleton replied on 5 March456, undertaking to make the necessary arrangements with the University to permit Oliphant to maintain a part-

456 Appleton to Oliphant, 5 March 1945. AB 1/42, UKNA.

262

time connection with TA (“other than committee work”). Regarding Oliphant’s plans at Birmingham, he wrote “I am glad to see that you propose to start out on a new line in nuclear physics, and can well understand that we cannot expect the whole of the Birmingham laboratory to be available in TA You know my views concerning the function of University laboratories!” In the present context, this cryptic (and loudly stated) remark is presumably supportive. It seemed that, in his view, universities should serve primarily as sites of fundamental research and sources of trained personnel.

As might be expected, Appleton sought to counter Oliphant’s negative views on the state of affairs in the U.K. Plans for “our own post-war TA organisation” were progressing, and awaiting agreement from the government as to the size and shape of the enterprise. He would be able to start offering jobs once that was done, and hopefully at attractive salaries. DSIR was also working on broader post-war plans. “But of course you must remember that we are very much at war still, and it is impossible to take people who are engaged in war work into our post- war schemes. Building research naturally looms large with us at the moment.”

At much the same time Oliphant was in correspondence with Moon (then still at Y), continuing the conversation started in November on the prospects of the new machine. On 2 February Moon wrote

First of all I am sure you have chosen a fine field of work. I haven’t thought about the techniques as deeply as I would like to do, but as far as I have gone I am highly enthusiastic and feel very lucky to have a chance of helping you with this.

263

I believe there is a real chance of establishing in Birmingham a laboratory that will carry on a large part of the traditions of the Cavendish.

There is no doubt there will be competition, especially for first entry into these unexplored continents of energy. I know that Kerst has plans for a super-betatron and Alvarez has told me he has hopes of securing a good deal of radar equipment for multiple gap acceleration of particles you were thinking about 18 months ago457. I believe your best chance of building up a real lead in this field is just as you have said – to get there quickly with relatively simple equipment aiming at producing relatively small numbers of particles458.

Oliphant responded on 27 February.

I expect it is because the EM plant is going well that I found my thoughts going to post-war work but I like to believe that some national importance attaches to the job of getting teaching and research going again in our universities.

457 This comment indicates that Oliphant and Moon discussed the possibility of some form of multiple gap accelerator using pulse techniques in middle to late 1943. We may note the proximity of this date to that at one time ascribed by Oliphant to his first proposal regarding the synchrotron. However, it is not clear that Moon and Oliphant had discussed a circular form of accelerator. They may have been instead considering a linear accelerator, a form with which Oliphant was familiar from the Cavendish.. The reference to Alvarez supports that view , since he did build a linear accelerator using pulse techniques at Berkeley in the immediate post-war years. Therefore we need not see Moon’s comment as undercutting the argument made elsewhere in this thesis that a 1943 date for Oliphant’s first synchrotron proposal is not plausible. 458 Moon to Oliphant, 2 February 1945. EBA. Moon’s enthusiasm for the project, so evident here, was in strong contrast with his criticisms expressed (at least privately) later. This will be taken up in Chapter Seven. 264

I do not think we should worry too much about competition in this field. On the technical side we have valuable expertise. On the purely nuclear side we have a unique background459.

At much the same time, the wheels were beginning to turn at the Nuffield Foundation. Events following Tizard’s offer to seek assistance for Oliphant can be traced through the records. The ninth meeting of the Foundation’s governing body in March 1945460 received a report from Tizard titled Post War Needs for Research in Physics at Birmingham University. This noted the need for financial assistance to cover the cost of three research assistants and of workshop and laboratory assistance and materials. The Minutes recorded “The trustees viewed the proposal sympathetically and recommended the setting aside provisionally of a sum of £25 000 for grants over 5 years, to meet the costs thereof, the procedure in regard to the appointment of research assistants to be the same as that proposed for research students in Professor Dee’s department at the University of Glasgow461.”

Some context is needed here. Clark has recorded462 that in September 1944 Tizard had impressed on the government, via John Anderson, the need to revitalise university teaching and research, so badly affected by the exodus of university staff to defence-related work, including to the Manhattan Project. Action was needed to encourage those vital human resources back to the universities once the war was over, otherwise they might be lost permanently to industry or to overseas enterprises.

459 Oliphant to Moon, 27 February 1945. EBA. 460 Minutes of the 9th meeting of the Board of Trustees. March 1945, Item 224. Nuffield Foundation Archives. 461 At the same meeting the trustees had received a request from Dee for a sum of £50 000 - £60 000 over 6 years to pay for research studentships and fellowships, technical assistance and research apparatus, Dee’s intention being to build a betatron. The trustees were in favour of granting the application, subject to a formal request from the University and agreed to provisionally set aside the sum. The meeting agreed that appointments were to be made by the Foundation on nominations received from the University. 462 Clark (1972), p. 58-59. 265

That also called for expenditure on modern facilities. As he pointed out in a memo circulated to his fellow directors in December, the matter was particularly acute in the case of nuclear physics. "[Research in nuclear physics] is extremely expensive work, relatively speaking, and cannot now be adequately done without engineering equipment of a size, cost and complexity which is quite new to university laboratories.”’ We can contemplate the extent to which such thoughts had been stimulated by his discussions with Oliphant.

Tizard had also been talking to Phillip Dee, another Cavendish alumnus and former Oliphant colleague, still working on radar at TRE but soon to return to his professorial position in Glasgow; Tizard's December paper had included a proposal that the Nuffield Foundation make £60 000 available to Dee to rebuild his department.

Clark has underlined how important these grants were in the context of the times. The Government was not yet ready to move on the matter, with the war still underway. Indeed as we shall see, it was not until October 1945 than any process was put in place to channel support to University-based nuclear physics projects. Meanwhile the Foundation was active. Between the end of 1944 and the spring of 1946, it allocated nearly £160 000 to the natural sciences; most of it going to nuclear physics

…. an operation which allowed the university departments concerned to build up their staffs with men who might otherwise have been attracted to the United States, or to industry, rather than to fundamental research, which enabled them to plan ahead for the expensive equipment which would be needed; and which

266

enabled the transformation in this field to be carried out with a smoothness which would otherwise have been impossible463.

A year later the trustees would be told that Lord Nuffield had approved the recommendation464, The University of Birmingham would now be informed of the terms and conditions, including that the trustees reserved the right to review the grant should Oliphant resign his Chair during the grant period. It appears that the trustees saw Oliphant as essential to the future of the endeavour. The stipends of the three new Nuffield Fellows were set at £750 a year.

How the idea grew

We have already noted the existence in the Barr Smith Collection of a note book maintained by Oliphant through the middle and late months of 1945. He was now back permanently in Birmingham, endeavouring to move his department into a peace-time mode of work, but also seeking to advance his vision. The first entry was dated 15 May, the last 25 July (other than an entry in late December dealing with a particular issue to be considered later). From the start he was concerned with the challenge of changing the frequency of the accelerating voltage as the particles picked up speed. “If we consider the possibility of injecting protons from the cyclotron at 19 MeV”, their velocity would be about 20% that of light. At a final energy of 1000 MeV, the particles would be at 90% of light speed. So the time taken for a particle to complete an orbit would be cut by a factor of 4.5, and the accelerating frequency must rise by the same ratio (from 2.2 Mc/s to 10 Mc/s). “The question is can we change f by this factor?”

463 Ibid. 464 Minutes of the 10th meeting of the Board of Trustees. March 1946. Item 263. Nuffield Foundation Archives. 267

Within the next couple of days (the page is not dated but it was filled in before the 18th) Oliphant wrote that “we are now a position to write a tentative specification”. The machine Oliphant was at this moment envisaging for electrons, rather than for protons as a day or two earlier. Injected by an “impulse generator or linear resonance accelerator” at 1.25 MV, the particles would begin to orbit at a radius of 380 cm, provided the field strength at the time was 12 gauss. Over the next 400 microseconds, the field would be ramped up to around 10,000 gauss and the electrons reach a peak energy (unspecified) in a 400 cm radius orbit. The numbers, of course, are not of interest in themselves, since they were prone to change. Rather the nature of the calculations indicated the variety of issues going through Oliphant’s mind, as he sought to transform his concept into a buildable machine. Already the numbers were looking challenging, especially the requirement that the variations in field strength and accelerating field frequency would need to be in synchrony to one part in 8000.

At the bottom of the page, Oliphant set down some thoughts about materials. “The electrode system would need to be made of materials of high resistivity or must be laminated [?] to prevent eddy currents from distorting the field. This will be one of the most difficult parts of the job.”

On 18 May, Oliphant “digressed” (his term) to muse further on the possibility of firing electrons into the machine using a linear accelerator. His figures suggested a series of 100 “resonators”, each five cm long and each adding 100 kV to the energy of the electron beam by the application of “modern pulse techniques”, would generate 10 MeV at the end. “This looks good and practicable”, he observed, but he did not pursue the idea.

268

On 12 June, Oliphant drew up a list of the tasks still to be attacked, both in calculation and in design. These were substantial. The coils to generate the guiding magnetic field awaited design, involving calculations of, inter alia, the optimum size of the conductors and the best form of insulation. In the queue for design were oscillators for the accelerating electric field and ways to link it to the rising magnetic field with the needed precision, circuits and controllers to excite the magnet, injection and extraction systems for protons and electrons, an impulse accelerator for electrons and “fitting the proton accelerator” to the cyclotron. And of course, control systems would be needed the whole apparatus. At the end of the list were half a dozen tasks needed to get the Nuffield cyclotron completed and operational, both for its own sake and for the role it was to play in the new accelerators.

Oliphant to Akers 16 June 1945

Five months after his previous note to Akers (and only a few days after his list of work still to be done), Oliphant wrote again. The document, consisting of a memorandum entitled The Acceleration of Particles to Very High Energies and a covering letter, dated 16 June 1945 and entitled Nuclear Physics and TA, can be found in the public record465. For the first time we have a proposal document whose provenance is secure. The memorandum was undated (as has been noted by a reader in a written comment on the document). It carried no signature but only Oliphant’s typed initials. There can be no doubt however that the letter and the memorandum belonged together.

Also in the public record is a response from Akers, dated 23 June 1945, in which he wrote “I would like to acknowledge in writing your letter of June 16th enclosing a memorandum on ‘The Acceleration of Particles to

465 Oliphant to Akers, 16 June 1945. AB 1/111. UKNA. 269

Very High Energies’ and suggesting that an experimental program of this kind, which would cost, in equipment alone, about £50 000, should be supported by the T.A Directorate of D.S.I.R.”

I agree entirely with you and will arrange an early discussion with Appleton to whom you have sent a copy of the note.

The whole idea seems to be highly ingenious and, I gather, very likely to be successful to the extent of producing beams of particles of energies of the order of 1000 M.e.V.466.

What then is the proposal to which Akers gives so enthusiastic a response? Oliphant’s covering letter is worth quoting in full.

Dear Akers

I have been giving serious consideration to the long-range aspects of the utilisation of the nuclear energy, and I have certain suggestions to make.

The present method of extracting nuclear energy from certain of the heavy elements involves the use of chain reactions produced through the ‘fission’ process, during which a small fraction of all the total mass energy is liberated. The total energy which could be obtained by a transformation of the whole mass (mc2) is 1000 times as great as the energy set free in the fission process.

It is obvious that there is every justification for the search for more cataclysmic disintegration of nuclei than is achieved by T.A, as we know it at present. The first steps only have been taken in

466 Akers to Oliphant, 23 June 1945. AB 1/111. UKNA. 270

a new field. The possibilities offered by chain reactions based on fission must be followed up vigorously in the proposed TA establishment and elsewhere. At the same time, the possibility that there are other methods waiting to be discovered, one or more of which may be an improvement on fission, must not be neglected.

One possible method of investigation which might lead to the discovery of new processes is to carry out experiments on 2 nuclear interactions at bombarding energies which exceed moc for the proton and neutron. A phenomenological survey of nuclear reactions at energies above 1000 M.e.V., combined with a careful analysis of scattering processes using both electrons and protons, could not fail to yield results of the highest importance.

Experimental observation in this realm of energies is confined, at present, to cosmic rays467. The occurrence of phenomena is rare, and when observed they are often difficult of interpretation due to uncertainties about the nature and energy of the initial particle and of the products. Knowledge would advance at a rate greater by some orders of magnitude if beams of particles, accelerated to energies above 1000 M.e.V., were available in the laboratory.

I have developed some proposals for new methods for accelerating protons and electrons to these very high energies. I am convinced that we can carry out this work successfully in my laboratory if we have the necessary backing. The accompanying

467 Cosmic rays are streams of very high energy subatomic particles generated and accelerated by processes in deep space far beyond the Earth. Cosmic radiation had been discovered only some 30 years earlier. At the time Oliphant was writing, the nature of the incident particles was not well understood, and their arrival remains largely unpredictable to this day. 271

brief report describes the method with greatest promise at this time.

It is seen that we have to provide large pulses of electrical energy in order to produce the guiding magnetic field of the correct form. The cost of equipment for this purpose is estimated to be £50 000 spread over 18 months, a sum beyond the resources of this laboratory, though we can ourselves provide a large proportion of the salaries of research workers who will carry out the project.

In view of the intrinsic importance of such investigations to the future of TA, I believe this is a proposal which could be properly fostered by the Directorate of Tube Alloys. I would like to suggest therefore that equipment and priorities be provided for this work by your department, subject, of course, to such interest in the ultimate disposal of this equipment, and of the results obtained with it, as may be thought desirable.

Oliphant had copied the letter and enclosure to Appleton and to Chadwick. The covering letter summarises Oliphant’s now increasingly familiar motivation for seeking funding, the key points being

o There is the potential for much greater release of energy from nuclear energy than is currently available from chain reactions based on fission. o Such release will require major advances in fundamental knowledge about sub-atomic particles and processes. o Such knowledge is all but certain to be revealed if bombarding particles of 1000 MeV or more can be produced in a controlled fashion in the laboratory.

272

o Oliphant is proposing a new type of accelerator to produce such very high energy particles, but its capital cost (£50 000) would exceed the resources of his laboratory. o The Directorate of Tube Alloys would be an appropriate source of the needed funds, given the “intrinsic importance of such investigations to the future of TA”.

The letter reads as if this was the first time Oliphant had written to Akers on such matters, though he later claimed to have raised the issue before. We have already noted that the letter of January 1945 makes no reference to any earlier approach, raising the possibility that there had indeed been nothing earlier, and that Oliphant was confused in his later recollections. There can be no such explanation here, if we accept the January 1945 letter as genuine. It may be that Oliphant had forgotten he had advanced the matter before, perhaps a consequence of the pressure of war-time and post-war work, though that seems unlikely. Perhaps he thought it was better to start again from scratch, not relying on any earlier communication that Akers himself may have forgotten about, especially since this was a more formal approach. Certainly Akers in his response does not appear to recognise the proposal, which is broadly similar to that he supposedly saw in January, though differing in many details. Perhaps having sighted the original idea and dismissed it as inopportune, it may well have been driven out of his mind by other more pressing issues. A third possibility, and perhaps the most likely, is that Oliphant was now writing for a broader audience (having copied the correspondence to Appleton and Chadwick) who would not have been party to the earlier correspondence.

The attachment was almost a replica of the Oliphant Memorandum. In only a few places were there any significant differences from that earlier presentation. The rationale and the technology proposed are the same,

273

even though, as we have noted, Oliphant was continuing over a number of months to refine and develop various aspects of the proposal. This suggests that the two documents were closely linked in time, with perhaps only a few weeks between them. If so, the writing of the Memorandum could have been as late as May or even early June 1945. Oliphant’s notebook yields a clue, already referred to. The 18 May entry was labelled a “tentative specification”. It seems unlikely that Oliphant would have written that if a more detailed specification already existed in the form of the Memorandum. It would follow that the Memorandum was first drafted after 18 May, perhaps stimulated by the “specification”. A few amendments then generated the attachment to the 15 June letter to Akers.

It is therefore not necessary to go though this document in detail, but only to point out the places in which it differs from the Oliphant Memorandum. There are only two notable omissions: some of the mathematics dealing with the variation of the magnetic field over time in the acceleration of electrons, and the diagrams that show alternative layouts for the accelerating electrodes. There are two significant additions, one little, one large, to the original Memorandum, which show that time had passed and some progress had been made. After discussing the issue of the likely high cost of equipment to energise the magnet, he adds “Precise designs and estimates are being obtained”. Then, in a paragraph added at the end, we read

Program

Sufficient paper work has now been done to show that the proposed method of acceleration is practicable. It remains to calculate in detail the form of coil needed to produce the magnetic field and to make measurements on small scale models

274

to establish the design. At the same time, the details of the electric circuits and of the injection and extraction systems must be worked out. Much of the circuit work must be carried out by experts whom I hope to attract to the project.

I estimate that the whole of the detailed design can be worked out and the construction completed in two years. It is probable that, with reasonable priority, the condensers or the machine required to supply the energy needed to produce the magnetic field can be designed and manufactured in this time.

For the first time Oliphant put a time scale on the project; two years. It is not stated with real conviction; completion would ultimately take seven years. In the covering letter he had put in another key figure, an estimated £50 000 for the cost of the equipment (spread over 18 months). This represents perhaps the most striking difference between the January and July visions. Six month earlier the figure quoted to Akers was £200 000, albeit for a machine different in important detail. As we shall see, the sum Oliphant ultimately sought from the authorities lay between these extremes; the final cost exceeded the upper bound.

This, then, was the final form of the proposal, the version in which the enterprise entered the bureaucratic process which led in less than a year to the securing of funding. In passing we might note that, as we shall see, the machine which was finally built resembled much more closely that proposed in January 1945, than that described in the Oliphant Memorandum and the June 1945 submission.

The establishment of “priority”

275

We have set out in considerable detail the development of the concept of the synchrotron in Oliphant’s mind, as conveyed in this series of letters and memoranda, partly to examine how circumstances of time, place, personality and concurrent events had affected the course of that development, and partly to lay a foundation for examining what priority he was entitled to claim in the "invention” of the synchrotron or the “discovery” of its operating principles. Debate on priority in science is both a valid and a valuable exercise. As Miller commented in his examination of claims of priority in the discovery of the compound nature of water

Being first in science is important, as it is in geographical discovery. Entering virgin territory geographically or conceptually is a privilege that few of us will experience. “Proudly to have thought where none had thought before” might be the scientist’s Star Trek experience. However merely thinking and discovering are worlds apart, as we shall see. Discovery is a social as well an intellectual process. It is, we shall argue, a property ascribed to certain intellectual and practical processes rather than inherent in them468.

We can delineate two meanings of the word “priority” in this context. If we agree on the nature of the “discovery” or “invention” under consideration, priority is a matter of timing. If on the other hand we find that there are distinguishable elements which contributed to the totality of the “discovery” or “invention”, then priority can be ascribed to various people with less concern for timing.

As we noted in our attributional survey, five individuals are commonly ascribed significant roles in the development of the first generation of

468 Miller (2004), p. 1. 276

synchrotrons: Oliphant, the Americans Edwin McMillan, William Brobeck and Stanley Livingstone and the Russian Vladimir Veksler. Of these, Brobeck and Livingstone were essentially machine builders, responsible respectively for the Berkeley Bevatron and the Brookhaven Cosmotron. The other three all had a hand in conceiving such a machine, and we shall shortly look at the contributions of McMillan and Veksler,

The essential nature of the synchrotron lies in the way in which it differs from the pre-existing technologies for the cyclic acceleration of charged particles. Unlike the cyclotron, in which the particles follow a spiral path, the synchrotron confines them to orbits of fixed size by use of a rising magnetic field. The benefit is that only an annular magnetic field is required, with greatly reduced size and cost of the required magnet, and bringing the possibility of reaching very higher energies at an achievable cost. From earliest days Oliphant had sought to work out (albeit in a preliminary fashion and with a minimum of mathematics), how such a machine could operate, how to make the machine work. All this has the character of an “invention” since nothing like it had been built before. Oliphant wanted to build it because he wanted to reach the highest possible energies.

As we shall see, McMillan and Veksler came from a different starting point and via a different route, which resembled more a “discovery”. They were concerned (at least at first) to extend the energy range of the existing cyclotron technology by overcoming the limitations imposed by relativistic mass increase. As a consequence, rather than as a motivation, they came across the existence of stable orbits and the concept of phase stability, which they set out with mathematical rigour. That led both men to propose modifying the existing cyclotron using the new understanding and then the building of a machine to accelerate electrons to new energies.

277

McMillan and Veksler

In Oliphant’s 1967 account, we read

My belief that I had found a unique new method for the acceleration of particles to very high energies was shattered by the publication of the comprehensive and beautiful papers by McMillan and Veksler. However these did provide powerful support for the proposal that we build such a machine in the UK.

The two papers cited here both appeared, as far as English readers were concerned, in mid to late 1945, though Veksler’s paper had been published in a Russian-language journal the previous year. In English, its title was A New Method of Accelerating Relativistic Particles469. McMillan’s paper The Synchrotron – a Proposed High-Energy Particle Accelerator470 was published in September 1945.

The papers were strikingly similar in many ways. Each identified a major barrier to attempts to raise the energy imparted to particles by a cyclotron, namely, the increase in mass of the particles as they neared the speed of light. Even though the particles might gain in energy, the increasing mass might see their velocity decrease, increasing the time taken for the particles to complete an orbit and so causing them to fall out of phase with the fixed-frequency oscillating electric field that was accelerating them. If they arrive half a cycle out of phase, the field would have caused a deceleration. This appeared to place a limit on the peak energy obtainable from such a machine. McMillan, for example, estimated that while the almost complete cyclotron at Berkeley, with its

469 Veksler (1944). 470 McMillan (1945). 278

huge 184-inch magnets was intended to accelerate deuterons to 100 MeV, it would have been very difficult, or even impossible, to reach even 60 MeV.

The solution proposed by both men was the same; provide the capability to vary the frequency of the electric field to match the orbital frequency of the particles. This would create a frequency-modulated cyclotron, synchro-cyclotron or phasatron (the term later preferred in the Soviet Union). The relativistic particles remain in synchrony with the changing field, so overcoming the energy limit. When the principle of phase stability was applied to the 184 inch cyclotron, it immediately reached nearly twice the planned deuteron energy.

McMillan’s own account of the growth of the phase stability concept was set out in his letter to the author in November 1980471, McMillan commented “This [the notion of phase stability] was recognised in my first memo on the subject, written at Los Alamos and dated July 4, 1945, not long after I had the idea”.

That memorandum472 was headed The Synchrotron, McMillan’s newly- coined name for the proposed machine, here used for the first time and now universally applied to accelerators of the type Oliphant was proposing. However, careful study of the document shows it fell well short of Oliphant’s vision. It began “This is a device for the acceleration of particles to high energies. It is essentially a cyclotron in which either the magnetic field or the frequency is varied during the acceleration, and in which the phase of the particles with respect to the high frequency electric field automatically adjusts itself to the proper value for acceleration.”

471 McMillan to Ellyard, 8 November 1980. EBA. 472 The Synchrotron, Attachment B to McMillan to Ellyard op. cit. 279

McMillan went on to expound the notion of phase stability, and did so at greater length in his first published paper on the matter a couple of months later473. McMillan noted the success of the cyclotron in accelerating charged particles to very high energies by the repeated application of an oscillating electric field. He suggested that if a very large number of repeated accelerations were required, there may be difficulty in keeping the particles in step with the electric field, particularly as the particles gained mass as they approached the speed of light. Paradoxically, this would cause them to slow down and fall out of phase with the constant field.

The device proposed here makes use of a “phase stability” possessed by certain orbits in a cyclotron. Consider, for example, a particle whose energy is such that its angular velocity is just right to match the frequency of the electric field. This will be called the equilibrium energy. Suppose further that the particle crosses the accelerating gaps just as the electric field passes through zero, changing in such a manner that an earlier arrival will result in an acceleration. This orbit is obviously stationary. To show that it is stable, suppose a displacement in phase is made so that the particle arrives at the gaps too early. It is then accelerated; the increase in energy causes a decrease in angular velocity, which makes the time of arrival tend to come later. A similar argument shows that a change in energy from the equilibrium value tends to correct itself. These displaced orbits will continue to oscillate in both phase and energy about their equilibrium values.

473 McMillan (1945). 280

To accelerate the particles it would now be necessary to change the value of the equilibrium energy, by varying either the magnetic field or the frequency. While the equilibrium energy is changing, the phase of the motion will shift ahead just enough to provide the necessary accelerating force. Having devoted the bulk of the paper to the acceleration of electrons, McMillan concluded

The application to heavy particles is not discussed in detail, but it seems probable that the best method will be the variation of frequency. Since this variation does not have to be extremely rapid, it could be accomplished by means of motor-driven mechanical tuning devices.

The synchrotron offers the possibility of reaching energies in the billions volt range with either electrons or heavy particles; in the former case, it will accomplish this at a smaller cost in materials and power than the betatron; in the latter, it lacks the relativistic limit of the cyclotron.

The construction of a 300 MeV electron accelerator using the above principle at the Radiation Laboratory at the University of California in Berkeley is now being planned.

Regarding this paper, McMillan commented in 1980:

The first draft of my letter to the editor of the Physical Review was also written at Los Alamos and was dated August 26, 1945. This letter was submitted by way of Berkeley although I did not actually go there until late in December. In this letter the application to both electrons and heavy particles was discussed and my proposal for an electron accelerator outlined, but no

281

specific proposal for a heavy ion accelerator was given. Such a proposal, leading eventually to the completion of the 184 inch as a synchro-cyclotron474, was originated at Berkeley soon after my arrival, I think by JR Richardson.

He went on

These proposals were for machines in which the magnetic field or the frequency was varied. A more complicated design in which both the parameters are varied together in such as way as to keep the orbital radius constant (now called the “proton synchrotron”) was proposed by William M Brobeck sometime in 1946. I recall many discussions with him on the topic, but have no notes or documentary evidence of dates, except for a drawing dated November 12, 1946 which is a conceptual design of the Bevatron in considerable detail. Shortly after this drawing was made, II Rabi of Columbia visited Berkeley and took a copy back with him. That is how the proton synchrotron design got to Brookhaven.

We will come to the timing for the development of the proposal by Brobeck for the Bevatron later. At this time it is reasonable to assert that Oliphant’s proposal to Akers for a proton synchrotron was many months ahead of any proposal originating at Berkeley, and even more in the case of Brookhaven. As we shall see, by the time plans for the Cosmotron at Brookhaven and the Bevatron at Berkeley were

474 In a “synchro-cyclotron”, also referred to a “frequency-modulated cyclotron”, the frequency of the oscillating electric field was reduced as the particles gained mass (and therefore lost velocity) to keep them in phase with the field. Without this intervention, the 184 inch cyclotron would not have been able to reach its design energy of 100 MeV for deuterons. In the event, it was able to double that energy.

282

significantly advanced, Oliphant had secured his funding, completed the design and was getting ready to commence construction.

We need not go to such length in considering Veksler’s paper475, though it was earlier in publication. The challenge identified was the same, namely the impact of the relativistic mass increase of the particles on the energy limit of the cyclotron, as was the solution proposed, a variation in the frequency of the accelerating voltage. Like McMillan, Veksler saw an interesting implication, that under the right circumstances, the particles would keep pace with the accelerating field through a phenomenon he called “auto-phasing”, phase stability by another name. Like McMillan, his interest in the principle, beyond applying it to cyclotrons476, was in accelerating electrons; by strengthening the magnetic field as the particles gained energy they could be confined to a stable orbit. The more complex possibility of accelerating heavy particles in a similar apparatus was mentioned but not pursued.

Discussing the impact of the McMillan and Veksler papers, Livingstone commented

Meanwhile, the principle of phase stability in synchronous accelerators was announced by McMillan and Veksler in 1945. This principle led to the development of the synchrocyclotron and of the electron synchrotron, at the University of California and elsewhere. Both papers describe two techniques for synchronous acceleration, involving the used of frequency modulation for heavy particles or an increasing magnetic field for electrons. Both papers also include the possibility of acceleration of protons at

475 Veksler (1944). For a discussion of the genesis of Veksler’s ideas see Ratner (1964), Chapter 19. 476 He called such a machine (a “synchro-cyclotron” in Berkeley terminology) a “phasatron”. 283

constant radius by varying both the frequency and the magnetic field, but neither paper was primarily concerned with this more complicated application477.

In Oliphant’s reply to Forman478 discussed earlier, he commented that, in the formulation of his proposal “there was no consideration of the problem of the cyclotron, and so I missed the synchro-cyclotron idea”. This supports the contention that the genesis of the concept of the synchrotron in Oliphant’s mind differed from the way it came to McMillan and Veksler. They were both explicitly concerned with the consequence of relativistic mass increase for cyclotron operation, and devised a strategy to combat it. The concept of the synchrotron flowed from that. Oliphant’s focus was on a broader issue. While he was aware of the relativistic energy limit on the cyclotron, he was driven more by the need to cut costs and so open the way to higher energies through an affordable technology.

Three-fold simultaneous discovery?

Oliphant may have been dismayed to find what he thought was his idea alone in print above the names of other researchers, but it made little difference otherwise. There is no way he could have been influenced by the published proposals of McMillan and Veksler prior to finalising his own proposal to build a proton synchrotron. He would have been able to read both papers only after submitting his proposal on 16 June. By McMillan’s own admission479, his first memo on the matter was not written till July 1945. He had formulated the concept only a little time

477 Livingstone ( 1969), page 99. 478 Oliphant to Forman, 8 March 1974. Attachment Q to McMillan to Ellyard, 8 November 1980. 479 McMillan to Ellyard op cit. 284

earlier, and had not applied it seriously to the acceleration of protons even in the paper written in August.

Is it possible that some informal contact between the three men had either stimulated or disseminated the key ideas? This was obviously not feasible in the case of Veksler, but feasible at first sight for Oliphant and McMillan since both had associations with Berkeley and in the last years of the war both were working within the Manhattan Project. Oliphant, however, was mostly at Oak Ridge when not at Berkeley or in Britain, while McMillan was employed mostly at Los Alamos (from where he wrote his July 1945 memo). There seems to have been very little opportunity for interaction, though this could be made more certain by constructing a travel diary for each man based on other records. In any case, both men have denied any prior knowledge of each other’s enterprises in this field. Oliphant claims to have been surprised by the appearance of the McMillan paper, and McMillan has written that he knew essentially nothing of Oliphant’s plans until well into 1946480.

Confirmation of this conclusion can be found in a letter from Moon to Oliphant in October 1945481. Having been promised by Oliphant on several occasions over the previous year that he would be sending a note about the new method, Moon had finally received a copy of the “accelerator memorandum” when Peierls brought it482 to Los Alamos. Moon commented that it had been a pity Oliphant’s earlier attempts to send the document had not succeeded. He could, he said, have shown it to McMillan, who had just left to return to Berkeley. “It would have been nice to discuss your plans in some detail [with him]. Even though he is committed to a somewhat different scheme of which you know

480 McMillan to Ellyard op cit. 481 Moon to Oliphant, 16 October 1945. F 22, Moon Papers, BUSC. 482 This document was presumably the attachment to Oliphant’s letter to Akers on 16 June 1945. 285

already, he might have had some comments which would have been useful to you.”

The conclusion seems unavoidable. As has been stated by others, though with less evidence, Oliphant, McMillan and Veksler had indeed independently conceived the synchrotron, though Oliphant reached his conclusions by a different route and with different considerations in mind.

Did Oliphant understand phase stability?

Central to the concept of the synchrotron, and its strategy of confining the accelerating particles to a narrow orbit, is the issue of stability, of orbit and of phase. The particles must be kept from straying too far from the mean orbit, either radially or axially, lest they strike the sides of the vacuum chamber, and they must be kept in the right relation to the alternating electric field that is accelerating them, lest they fall out of phase and cease to be accelerated. The latter constitutes McMillan’s phase stability, dubbed by Veksler auto-phasing.

In his 1967 memoir, Oliphant has the following to say about the general issue of stability of particles in their orbits.

It was necessary to convince oneself that the orbits could be made stable, both in position and phase. There was little difficulty in showing that, provided the field decreased radially at a rate greater than 1/r but less than 1/r2, the orbits would be geometrically stable. By considering what would happen to particles which either led or lagged in phase with respect to the voltage across an accelerating gap, it was possible to show that they would adjust themselves to the phase where they received

286

just the energy required to match the magnetic field. By sensing the magnetic field and using a derived signal to control the frequency of the oscillator, it should be possible to keep field and frequency in step throughout the period of that acceleration. The last problem lay outside my experience and capabilities, but I was confident that those versed in electronics would be able to find a solution.

While the issue of stability had hardly a mention in the Oliphant Memorandum, and therefore in the June 1945 submission, Oliphant devoted a long paragraph to it in his January 1945 letter to Akers. The words chosen show that he understood the importance of the concept.

In order that the protons be accelerated, it is essential that they do not drift appreciably in phase with the electric field. On the accompanying sheets I have plotted typical orbits for particles of different energies moving in a field that falls off as 1/r 1.5, an arbitrarily chosen gradient that seemed to give reasonably good stability. It will be seen that particles following paths corresponding to larger energy than those on the mean radius will arrive at the accelerating gap later, while those with smaller energy will arrive earlier. Hence, if a bunch of protons sits on the falling part of the R.F. amplitude, somewhere near the centre, those which have too much energy will receive less acceleration, and those with too little energy will receive greater acceleration. The accompanying analytical treatment shows that the phases will be stabilised. Experience with the problems of electron paths in and magnetrons, gained in our work with centimetre- wave radiolocation, makes me confident that bunching and phasing of the beam will present no problems.

287

We are left to wonder what Oliphant was referring to as the “accompanying sheets” or the “attached analytical treatment”, since neither has survived in the documentation now available. They were most likely elementary when compared with the more complex treatment presented by McMillan and Veksler, or even the later work done in Birmingham. Oliphant admits as much. Yet he does seem to have been well aware of the issue. Indeed his explanation of how the particles would be kept in phase with the accelerating field differs little in essence from McMillan’s treatment. There is therefore little support for those, including Oliphant himself, who claim he did not understand “phase stability”.

How do we sum up this discussion? What priority can we assign to Oliphant in these momentous events? The evidence supports the widely-held view that all three players (Oliphant, Veksler and McMillan) came independently to the concept of the “synchrotron”, that is, to the strategy of confining the particles under acceleration to an orbit of essentially constant radius. This we have called “the synchrotron principle”.

By itself, that statement underplays the complexity of these developments. There were significant differences in the route taken to the “invention” of the synchrotron. Oliphant was driven by the desire to secure the highest possible energies at an acceptable cost. The others began with the desire to overcome a problem with existing technology. That led them to “discover” phase stability and to explore the concept mathematically. Without such detailed treatment of the phenomenon, Oliphant’s understanding was relatively elementary, but it was sufficient to give him confidence that his “invention” would work. Furthermore, Oliphant moved swiftly to apply the idea to protons; indeed it quickly

288

became his preoccupation. For the other two, the first application they sought was the simpler acceleration of electrons.

As for priority in time, it seems that Veksler was first to write down a form of the “principle”, with Oliphant a few months later, though not in detail, and McMillan six months later again. Oliphant however was first, by many months, to seriously propose using the principle to accelerate protons, and to explore solutions to the many technical problems such an endeavour must inevitably raise. In this matter, he clearly had priority. Furthermore, as we now explore, he was first to seek and secure funding for such a machine, and to begin construction, with a lead-time of more than two years. Finally, as we shall explore in Chapter Eight, he had influence, perhaps a significant influence, on the initiation of the two rival American machines in Berkeley and Brookhaven. Taken collectively, these findings support our primary hypothesis, that Oliphant was, on many grounds, not merely “first among equals” of the pioneers of the proton synchrotron, but much more influential and worthy of acknowledgement than that phrase suggests. It may be too much to describe Oliphant as the “inventor” of the proton synchrotron, but the evidence we are examining here shows he had a greater share of that invention than any other individual.

Finding the money

With a proposal reasonably well honed, but still subject to change, it was time to secure funding. From the first, Oliphant, with some experience of seeking funds from private sources, had looked to the public purse for support. In his 1967 memoir, Oliphant wrote

289

A Nuclear Physics Advisory Committee had been established through the influence of Professor PMS Blackett. I think that at first this had no official standing whatsoever, but was called together by him to discuss what projects might be pushed for the good of the country. The Committee viewed my request for allocation of £250 000 with favour. The project was reviewed by the Council of the DSIR which would make the grant if any. We were successful, and at once began to gather together a small team of enthusiasts to carry the project through.

As might be expected in so brief an account, Oliphant greatly simplified both the bureaucratic process and the winding course to be followed before funding was achieved. It was certainly true that, in financial terms, the fate of Oliphant’s vision rested with the Department of Scientific and Industrial Research (DSIR), within which was housed the Directorate of Tube Alloys (DTA) and which would continue to manage TA matters until a separate organisation was created specifically for that task. Only DSIR, and at the policy level the Advisory Committee on Atomic Energy (ACAE), chaired by a powerful figure in the form of Sir John Anderson, could recommend that the Government spend the large sum required.

It followed that the key people to be convinced were Appleton as Secretary of DSIR, and Akers as Director of TA. Hence Oliphant’s keenness to keep them well informed. He had already written to Akers seeking support, copying the letter to Appleton; in late July he wrote again to Appleton to keep the pot boiling. He had continued to hone the proposal, reporting to Appleton that following advice from Wilkinson from the firm BTH, he had become convinced that "the right way to energise the magnetic field is to employ a short circuit generator, of the general type used by Kapitza and in switchgear testing. The possibility

290

of using a condenser discharge for the coils is less attractive, as prices obtained from BI Cables and from Philips would involve expenditure above £100 000”.

Estimates of the cost of equipment including the motor generator and switchgear, magnet field coils and the oscillator and ancillary equipment came to around £50 000. This was less than he had previously indicated, but raising the sum was still a major challenge. He had, he said, enough funds at his disposal for research fellowships and general expenses but, like most University departments, no capital fund on which to draw for large-scale equipment. So he wanted to know if DSIR, perhaps through TA, would be prepared to foster this work. “It would result, not in a detailed examination of energy regions which have been explored in the broad sense, but in the establishment of an entirely new technique as a route into new regions. I'm confident we can carry the proposals to a successful conclusion.”

In terms of process, the key agency was to be the ACAE’s Subcommittee on Nuclear Physics. It had been established by the ACAE at its meeting on 4 October 1945 “to make recommendations regarding the program of nuclear physics to be pursued in the country as a whole”483. Chadwick was to chair it, with the leading lights in the field as its membership (Blackett, Cockcroft, Darwin, Dee, Feather, Oliphant, Peierls, Pryce and Thomson). The Subcommittee was later enlarged in membership to include representation from most of the universities undertaking nuclear physics research.

As the Subcommittee would note in its Interim Report in December, its attention “was first directed to two schemes for accelerating particles to high energies which had been put forward by the Universities of

483A.C.A.E. (N.P.) (45) 2. 23 October 1945, AB 1/111, UKNA. 291

Birmingham and Glasgow respectively”. We should note that these two projects, championed by Oliphant and Dee respectively, had also been those granted initial assistance by the Nuffield Foundation. It is tempting to suggest that the establishment of this committee may well have been stimulated by the existence of these two proposals, though as we shall see, the interests of the Subcommittee broadened as time went by.

A note from Akers on 26 October 1945484 added impetus to the work of the Subcommittee. Prepared as a hand-written note for the “Nuclear Physics Subcommittee, Advisory Committee on Atomic Energy”, the document cited correspondence from Cherwell and Dale. On that basis, Akers commented that interest in “high-tension” equipment was growing, and the Subcommittee needed to survey “the whole position so that the proper advice can be given to establishments which wish to install high- tension equipment, and also to ensure that this type of equipment is available, in adequate amount and modern form, this country”.

Summarising the present state of the field in Britain, Akers noted that the only such equipment linked to the TA program was a pressurized Van de Graaf generator being bought for the “Experimental Establishment” at Harwell, with a proposal that a similar machine be built at the Cavendish. A scientific mission was studying a German betatron near Hamburg, and drawings of a cyclotron of “the latest American design” 485 were likely to guide the building of such a machine at Harwell-to-be. The summary ended “Finally, Professor Oliphant has produced a design for a novel type of apparatus for producing beams of very high energy”. Here is further evidence of the probable influence of

484 Memo by Akers High Tension Equipment, 26 October 1945. AB 1/111, UKNA. The original title has been amended by hand with the addition of the words “accelerating systems”. 485 This may have been of the Berkeley 184 inch cyclotron, though before its conversion to a synchro-cyclotron. 292

Oliphant’s proposal in getting the Subcommittee going. Dee’s enterprise is not similarly indicated.

Among the first decisions of the Subcommittee when it met for the first time on 18 October486 was to ask Oliphant for a paper, outlining his proposal and the way it would fit into other programs at Birmingham. (In due course, other universities would be asked for a similar document). By the time the Subcommittee held its second meeting on 30 October, Oliphant had his memorandum ready, but consideration of it was deferred until the next meeting, along with a report on nuclear physics research at the University of Liverpool, prepared by Rotblat487 and tabled by Feather.

The Subcommittee meet for a third time on 5 November, with Chadwick, newly returned from the USA, in the chair. Oliphant’s memorandum had already been circulated. In reality this was a copy of his letter to Akers of 15 June and its attachment, together with the summary of what other related activities were underway in Birmingham. There was some reporting of activities elsewhere; the reactivated Cambridge cyclotron should deliver a beam by Christmas, HT1 was running, HT2 was “in the process of reconstruction”, while plans were being made in Liverpool to enlarge the cyclotron and move it to a new site, with a target date of November 1946. But the focus of discussion was Oliphant’s proposed project.

486 Minutes of meetings of the Nuclear Physics Subcommittee are generally not found in the relevant files held in UKNA. The source of the information used here is the set of Oliphant’s own copies, held in Box 22X 8/31, SCUB. These have the added interest of preserving Oliphant’s own comments and markings. 487 , a physicist later well known for his engagement with the anti-war Pugwash Movement, in which Oliphant too was involved. 293

Headed Nuclear Physics in Birmingham,488 Oliphant’s note placed the accelerator enterprise in a context.

This programme relates to experimental work in nuclear physics and does not include the investigations in theoretical nuclear physics which Professor R. Peierls will pursue, though these programs will be integrated very fully. The existence in Birmingham of a full department of theoretical physics and the forthcoming creation of a research professorship in electron physics, together with a department of electrical engineering which will cooperate in our work, make it a good centre for the developments which are proposed below.

The rest of the document outlined, firstly, plans for the completion of the cyclotron and for experimental work using it to accelerate deuterons and alpha particles to 35 MeV and, with an upgrade, protons to 60 MeV. The research programme was to include a “very thorough” investigation of scattering phenomena at those energies, and a survey of nuclear reactions produced by protons. Oliphant expected the cyclotron to be in operation in three months, given that “the magnet and the bulk of the apparatus are now complete for operation”.

Oliphant went on to introduce his new endeavour, the “Acceleration of Protons and Electrons to Energies above 109 eV”. There is a minimum of technical detail, since Oliphant attached his much more comprehensive earlier correspondence to Akers. We need not therefore go through it, other than to note that Oliphant had gone back to using the discharge of condensers as the source of the energising current for the magnet coil.

488 Nuclear Physics in Birmingham. AB 1/42, UKNA. The document is undated, though a handwritten note of unknown origin has added “Oct 45?” 294

…. [short-circuiting an alternator] is cheaper than condensers and for a time we favoured it. However further discussion with the engineers shows that they are very dubious about repeated short-circuits at the power level required (500 M.V.A)489. The cost of condensers for this purpose is about £80 000. We believe we could carry out the whole scheme, using condensers to store the energy, for about £120 000.

It would appear that the anticipated cost of the project rose and fell depending what method was proposed to energise the magnet490. To underline the importance of his request, Oliphant gave his rationale, which we have heard before. “Experiments in the region of energies above 1000 M.e.V. can give information about fundamental particles which cannot fail to be helpful in the further development of the practical uses of nuclear energy.”

The Minutes of the meeting491 recorded that “in discussion, it was debated whether these proposals went sufficiently far”. In his copy of the Minutes, Oliphant underlined those words and added a “?”. The tone of the discussion must have surprised (and mostly likely pleased) him. The Minutes continued: “The Committee felt that the scheme for reaching energies of 109 volts would enable this country to take part in

489 The available copy (a carbon) in the public record is indistinct at this point and has been overtyped. This figure appears very high. 490 The Special Collections at the University of Birmingham holds a undated 2 page document which appears to be closely allied with the submission to the Subcommittee, to which it apparently makes reference. Entitled “Nuclear Physics Program at the University of Birmingham”, it concentrates on the synchrotron. We can assume that Oliphant wrote this document. It quotes a similar price for the machine (£120 000). One paragraph is worth quoting. “A special feature of these proposals is the facilities which will be created for the training of research students by the virile pursuit of a far-reaching attack on fundamentals.” The work would complement work with cosmic rays in Manchester and research with the Liverpool and Cambridge cyclotrons. Box 22X/8 (31), SCUB. 491 Minutes of 3rd meeting of Nuclear Physics Subcommittee of ACAE, 15 November 1945. Box 22X 8/(31), SCUB. 295

extending the frontiers of nuclear physics and that this is a possibility we cannot afford to ignore. The proposal was undoubtedly a bold extrapolation but it represented at once a scheme that was practical and not too visionary.”

Oliphant highlighted the last sentence in his copy of the Minutes and added another “?” It was clear that, at this early stage at least, the Subcommittee was won over. The meeting agreed to recommend to the Advisory Committee “that funds should be made available for apparatus described in Professor Oliphant’s memorandum to be erected in Birmingham”.

The Panel on “Apparatus for Accelerating Particles”

In the meantime, Oliphant had a job to do. At the 30 October meeting, the Subcommittee had agreed to “set up a panel to consider the question of manufacture in this country or procurement from abroad of apparatus for the acceleration of charged particles”. Oliphant was to chair this panel, which gives some indication of his standing in discussions at this time, and perhaps of the more advanced state of his proposals. There was to be representation from potential players in the development of the technology for particle acceleration; universities such as Oxford, Cambridge and Glasgow, government agencies such as TRE492, and industrial firms such as Metropolitan-Vickers and British Thompson-Houston. DSIR would provide the Secretariat.

Invitations to this gathering, to be held at the headquarters of BTH in Rugby on 16 November (the day after the third meeting of the Subcommittee), were issued over Oliphant’s name on 7 November. The

492 The Telecommunications Research Establishment, set up during the war to advance the technology of radar. 296

venue may have been chosen as being more central for most participants than London, and perhaps as highlighting the potential for involvement by British industry. Correspondence indicates that Oliphant would be provided with accommodation in the guest house at BTH.

The Panel brought together researchers from the Universities of Birmingham, Cambridge, Glasgow, Liverpool, London and Oxford, most of whom has recently returned from service on the Manhattan Project, together with industry representatives (BTH, English Electric, General Electric and Metropolitan Vickers) and government agencies (Telecommunication Research Establishment, National Physical Laboratory and Royal Arsenal). The Minutes recorded that Oliphant had outlined his plans for the Birmingham proton synchrotron. Phillip Dee from Glasgow did the same for a proposed 200 MeV betatron; others discussed possibilities for linear accelerators. No interest was expressed in cyclotrons.

Minutes from the meeting of the Panel493 were tabled at the fourth meeting of the Subcommittee on 5 December.494 Taking aboard advice from the Panel, the Subcommittee then formulated an interim report to the ACAE495, prefacing its proposals with some context-setting remarks; “We cannot conceive any satisfactory program in this field [nuclear physics] without considerable additional government expenditure……..It is essential that work be done as soon as possible on the acceleration of particles to very high energies”.

493 Minutes of the First meeting of Panel on Apparatus for Accelerating Particles, 16 November 1945. Box 22X/8 (31), SCUB. 494 Minutes of 4th meeting of Nuclear Physics Subcommittee of ACAE, 5 December 1945. Box 22X/8 (31), SCUB. 495 Interim Report of Nuclear Physics Subcommittee to ACAE, 8 December 1945. Box 22X/8 (31), SCUB. 297

The report recommended that support be offered for Oliphant’s accelerator to an amount of £120 000, that one or more leading firms such as BTH, GEC or M-V be contracted to build a 200 MeV betatron for Dee in Glasgow at a cost of £50 000, that TRE be encouraged to work on linear accelerators and that contracts be placed with GEC or BTH for the development of a technology called a Machlett tube for use in high- tension accelerators. Regarding the Oliphant and Dee projects, the report stated “these schemes appeared to us promising and to offer a relatively early prospect of important results.”

Defending any appearance of lavish spending on its proposals, the Subcommittee pointed to the very backward state of nuclear physics facilities in the UK, especially in comparison with “America”. Britain possessed only three cyclotrons, one of them (at Birmingham) still incomplete and the operating hours of the other two (Liverpool and Cambridge) severely restricted by technical problems. The United States, the report noted, “has now a program for six betatrons and some 20 cyclotrons. At least six American universities are planning to spend at least a million dollars each on nuclear physics research and that the Berkeley programme alone will cost three to four million dollars.”

The Subcommittee judged that, with perhaps one or two exceptions, the universities would be unable to provide the needed facilities unaided. Even staffing and servicing them would be beyond most of them. It gave the example of the high-tension apparatus needed for a cyclotron or betatron, comparable to that required by a main radio transmitting station, and resulting in total costs of £50 000 or more.

The days when the Cavendish Laboratory during Rutherford’s hegemony was run on an annual budget of a few thousand pounds are long past. Nuclear physics has become the most

298

expensive of today’s sciences. The scale of apparatus in other fields, e.g., physiology in which great advances may be expected, is fortunately much smaller and the cost therefore less496.

So Big (nuclear) Physics had arrived. There was no going back. Only the government could afford to pay for it, but the money would be well spent.

There appears therefore to be no alternative to extensive Government financial assistance on a scale exceeding that which the Government grants in other fields of scientific research. Such assistance is justified by the special responsibilities of the Government in this field. It would be useless for the Government to spend the sums which are proposed for the establishment at Harwell and for the production piles unless there was an adequate basis of fundamental research in the country. Unless there is sufficient apparatus to train an adequate number of nuclear physicists, the knowledge will not exist in this country to prevent the production program being based on obsolete methods, and there will be no leaders in the next generation able to direct either Harwell or the piles.497

We can see an interesting contrast between this rationale for the funding of high-cost equipment (including the Birmingham machine) and that which Oliphant had himself advanced. The Subcommittee saw a short- term and practical outcome for the investment: sufficient trained personnel to permit Britain realistically to plan to be a nuclear weapon state. Oliphant had seen a bigger picture: the potential for the discovery of new subatomic phenomena that might open the way to vast new

496 Ibid. 497 Ibid. 299

energy sources of interest not only to the military. Indeed, as he had told Appleton in March 1945, he planned to pursue investigations which might not immediately be of any interest to the current atomic energy program.

This development brought a risk: the potential to marginalize other areas of physics or even other areas of science. According to the Minutes of the ACAE meting, such a concern had been expressed by Chadwick (still in the USA) in a cable.

He agreed however that the program could not be safely reduced or delayed. It was pointed out that the (interim) Report did not propose that any teachers in other subjects should be diverted into nuclear physics. Some students would be attracted who might have gone into other faculties but exceptional treatment for nuclear physics was justified because of the Government’s special interest and because this was at the moment the most promising area in physics.

We can see considerable self-interest in this argument, which was being made by researchers who might not be judged objective. Rather they had made their careers and their reputations in the field and would be keen to see it prosper.

A slightly broader point of view, and one a little closer to Oliphant’s, is found in a later paragraph in the Interim Report.

It might of course be argued that a large claim for a single branch of physics may operate to the disadvantage of other branches of physics or of other branches of science. We think, however, that it is inevitable that in the coming decades investigations of the

300

fundamental nature of matter and energy will be the main focus of interest in physics, and that other branches of physics, however important they are, will necessarily take second place. This country must play its proper part in the forthcoming fundamental discoveries in nuclear physics, and must be able to share the development of new methods of harnessing atomic energy.

We can see from the perspective of today that such reasoning has proved groundless. While major discoveries have been made in nuclear physics using large-scale apparatus, they have made little if any impact on the methods of extracting nuclear power from uranium or plutonium. Such improvements as have been made have been of an engineering nature, based on advances in materials science or control methods and not on any new insights into the “fundamental nature of matter and energy”.

Speed bumps for Oliphant

The ACAE had received the Subcommittee’s spending recommendations at its 10th meeting on 8 December. The recommendations included, along with other proposals arising from the meeting of the Panel on 16 November, support for Oliphant’s enterprise. As to the latter, the Advisory Committee was unwilling to give wholehearted endorsement. While agreeing that it was “urgently necessary” to support all the lines of work outlined, “In the case of the work to be undertaken by Professor Oliphant, further planning and preliminary research were needed before it could be approved in detail.”

301

An extract from those Minutes was circulated to Subcommittee members on 13 December. Some members at least of the Subcommittee took that caution on board. At the meeting on 29 January, Blackett asked from the Chair “… whether the Birmingham project, which was a completely new departure, has yet reached the stage where actual machinery could be ordered. Ought there not to be some further examination of the stability of the orbits?498”

Oliphant was required to defend his vision. He claimed there were no grounds for anxiety about the stability of the orbits, especially following new work done at Birmingham based on Veksler’s recent paper, and unlikely to be any serious difficulty about the phasing. The main worry was “how to get the particles out”, but nonetheless he was ready to go ahead with the scheme, and thought it desirable that the condensers, which would take a long time to manufacture, should be ordered at once.

The Subcommittee was not convinced. “… [the Subcommittee] felt that before the government were asked to make a large grant in support of a new departure of this kind, it was reasonable that calculations be checked by some independent person”. The person to be given that task was Oliphant’s former Cavendish colleague Harrie Massey.

At the same meeting Dee’s proposed betatron was given an easier ride. The Minutes recorded his claim that, unlike the Birmingham venture, his proposal was not “entirely new”, and that “there was no doubt it would work”. But doubt lingered. Was the betatron the best way to reach such high energies? Maybe, maybe not, but the Subcommittee had already decided it could not afford to wait until it was certain of the best

498 Minutes of 1st meeting (for 1946) of NP Subcommittee, 29 January 1946. Box 22X/8 (31). UBSC. 302

methods, “before all the best research is done elsewhere”. We are reminded of Oliphant’s plea to Rutherford, on much the same grounds, not to delay investment in the High-Tension Laboratory.

So both Dee’s and Oliphant’s plans were to some extent gambles, though Oliphant’s was clearly more so. Yet the Subcommittee was keen to push ahead with both. The Chairman was “anxious lest the making of vital contracts be delayed”. The Secretariat was to write to DSIR seeking urgency, it having been determined by this time that requests for grants for new equipment would pass through DSIR. Both Dee and Oliphant would attend the next meeting of the relevant DSIR committee to argue their respective cases.

The Scientific Grants Committee of DSIR’s Advisory Council met on 27 February 1946, with Lawrence Bragg (Rutherford’s successor at the Cavendish) in the Chair. Heading the Secretariat was Appleton, who had first heard whispers of Oliphant’s plans nearly a year before. Dee and Oliphant had the opportunity to address the meeting. According to the Minutes499, Oliphant told the meeting that he was seeking £141 000 to build and staff his “equipment” and to operate it for 3 years. Dee’s request was more modest, £50 000 for a 200 MeV betatron. The lower costs reflected both the lower target energy and the more advanced state of the technology, betatrons already being in existence. Indeed, unlike the untried synchrotron, construction of the Glasgow betatron could be outsourced in its entirety to an industrial firm, and the Metropolitan Vickers Electrical Company Ltd was the proposed contractor. “Metro-Vick” was already gearing up.

499 Agenda and Minutes for meeting of Scientific Grants Committee, DSIR, 27 February 1946. DSIR 2/469, UKNA. 303

The firm had, on Professor Cockcroft’s advice, done work on this subject during the war, and now had got part of a 20 MV betatron assembled. They hoped to have it running in three months, to experiment with different forms of pole piece, and then to proceed to design the apparatus for 100-200 MV500.

In other words, the Dee proposal was a relatively sure bet.

The meeting papers summarised the two proposals. The description of the Birmingham enterprise drew heavily on previous submissions. At the head of the information paper, prepared presumably by Appleton or his staff, we read

The Nuclear Physics Subcommittee of the Advisory Council on Atomic Energy reported that there is presently a grave shortage in all the laboratories of this country of the costly equipment without which it will be impossible to carry out research of equal importance with that which is currently forging ahead abroad. They consider that it is essential that work be done as soon as possible on the acceleration of particles to very high energies501.

In this statement lay the justification for the support of the two current proposals, Oliphant’s and Dee’s, for which DSIR would be “an appropriate channel for grants”.

Firming up the program

By this time, there was more to talk about than just the proposals from Dee and Oliphant. On 5 December the Subcommittee had agreed on

500 Ibid. 501 Ibid. 304

the importance of identifying needs and opportunities in the nuclear physics area over a broader range of outcomes than had been considered by Oliphant’s panel. As a consequence, letters had gone on 11 December over Chadwick’s signature to solicit information on the state of nuclear physics in all universities, the Subcommittee feeling “under-informed” on this matter. Universities were also asked to provide estimates of capital and recurrent costs associated with possible expansion of that work.

Responses to the invitations were tabled at the meeting on 29 January 1946. The 35 invitations had elicited 21 replies, the fourteen responses still outstanding including the six colleges of the University of London. Sifting the responses, it was clear that seven universities [Bristol, Birmingham, Cambridge, Edinburgh, Glasgow, Liverpool and Oxford] were planning programs of research into nuclear physics over the next five years. Six of these [Oxford had been remiss on this matter] had given estimates of their needs for government funding, with requests totaling £430 000. On the basis of this additional information, the Subcommittee prepared to finalise its recommendations.

As for the proposed review of Oliphant’s calculations by Massey, there is no documentary evidence that this was actually sought. Certainly we can find no record of his opinion among the Subcommittee papers that survive. We also find no further evidence of caution about the proposal. When the Subcommittee met on 7 March to settle the list of projects for funding, the Birmingham machine was listed first [alphabetically] on the “immediate program”, with £120 000 recommended to fund the accelerator. Also on the list was £55 000 for Glasgow, mostly for the betatron, £85 000 for Liverpool including £60 000 for a new 60 inch cyclotron, and £70 000 for Oxford, with £50 000 for some “high-energy apparatus”.

305

Further down the list we find details of “second stage” funding. This brought in Edinburgh with £10 000 allocated for an artificial neutron source, Cambridge (a total of £85 000 including £50 000 for further development of a “high energy apparatus”, and £25 000 for a new cyclotron) and Imperial College, London (£15 000 for a Van de Graaf generator). Oxford was also to get £15 000 for a Van de Graaf and Liverpool £12 000 for a betatron. Recommendations were also made for grants for buildings, although those funds would come through the University Grants Commission. Birmingham was recommended for £15 000 to house the new accelerator, though as we shall see, Oliphant chose not to have a new building, but rather to cram his machine into an existing space.

Bubbling in the background of these developments were some more general concerns about the magnitude of the proposed expenditure, the manner in which it had been determined and the appropriate vehicle for conveying the funds. It is clear from the Minutes that advice on these matters was sought from the leading Treasury official Alan Barlow, who was present at several meetings of the Subcommittee. For example at the meeting on 12 February502, Barlow stimulated discussion as to whether the program of proposed purchases was “too diffuse”, and whether it might be better to concentrate on a few established centres. Might it also not have been better to ask what apparatus the nation needed (and then decide where to put it) than to ask what each university what it wanted?

In its defence, the Subcommittee asserted that the proposed program was “very slender”, and “hardly excessive”. The current program

502 Minutes of 2nd meeting (for 1946) of Nuclear Physics Subcommittee of ACAE, 12 February 1946. Box 22X/8 (31), SCUB. 306

contained only one large betatron (at Glasgow), two cyclotrons (Liverpool, Cambridge]) three Van de Graafs and “one special type accelerator”. Furthermore, training was the most important concern and justified a dispersal of facilities. “Where there was a large potential honours class in nuclear physics, there should be a major piece of equipment”. It appears that Barlow was playing devil’s advocate to some extent. The Minutes quoted him saying “The Treasury would not be frightened at the total cost of the program, if it was convinced it was in the national interest”503.

The Advisory Committee on Atomic Energy met on 28 March 1946, for its sixth meeting for 1946504. It was a high-powered gathering. Sir John Anderson, who had been Chancellor of the Exchequer until the mid- 1945 change of government, took the Chair. Those present included the chiefs of the Army and Air Force (Field Marshal Alanbrook, Marshal of the Air Force Tedder), Edward Appleton, representing DSIR, Henry Dale, until recently President of the Royal Society, representatives of the Treasury and the Foreign Office and the prominent physicists George Thomson and Patrick Blackett. Three representatives of the and Aircraft Production, led by Viscount Portal, were also present.

As the first and apparently only item on the agenda, the ACEA addressed the final report from the Nuclear Physics Sub-committee. It had recommended grants totaling £330 000, to be made by DSIR for equipment in the following universities; Bristol, Birmingham, Cambridge, Edinburgh, Glasgow, Liverpool, London (Imperial College) and Oxford.

503 Ibid. 504 Minutes of the 6th Meeting of the Advisory Committee on Atomic Energy (1946), 28 March 1946. AB 1/111, UKNA. 307

Of this sum, Birmingham was to get more than a third (₤120 000505) for its new accelerator. This, along with £53 000 to Glasgow for its betatron, had already been agreed to by the Committee for recommendation to the government at the meeting on 7 December 1945. In addition, the tabled papers indicated annual costs of £6000 for staff, equipment and power, to be met by the university.

Oliphant’s share was the largest but not by much; Liverpool was to get £85 000 for a cyclotron and associated equipment including a Van De Graaf generator. Comparing the two allocations of funds is instructive. The funding to Liverpool was only 30% smaller, even though the technologies of the cyclotron and the Van de Graaf were well established and the output energy would be considerably less than that of the Birmingham machine. Oliphant was moving into new territory. At the time of bidding in mid-1945, the Birmingham team was still to have no clear idea of the technical challenges the synchrotron would throw up, let alone the scale and cost of the solutions.

Addressing the proposal before the Committee, Thomson recognised that “the proposed grants amounted to a considerable sum of money. It would be useless, however, for the government to spend large sums of money at Harwell and on piles [nuclear reactors] unless there was an adequate basis of fundamental research in the universities”.

The Minutes noted that Chadwick had sent a cable expressing concern that the implementation of the funding proposals in the Report might operate to the detriment of other branches of physics. The view of the meeting was that “exceptional treatment of nuclear physics was justified

505 In his 1967 memorandum Oliphant recalled that the funding offered was £250 000, but he may have been including funds from other sources, such as the Nuffield Foundation, or later increases in the amount allocated by the Government. The often-quoted total of £141 000 appears to include £15 000 for a building and £6000 for other expenses. 308

because of the government’s special interest and because this was at the moment the most promising area of physics”.

The following day, Anderson forwarded the Report to the Prime Minister with a covering note506. The grants recommended totaled some £400 000, once allowance was made for necessary buildings. He contrasted that sum with the amounts planned for Harwell (£2.8 million for 1946/7 alone, £25 million to build one nuclear reactor). Yet, he commented,

…. these large expenditures will not be justified unless there is an adequate basis of fundamental research in this country and adequate provision for the training of the nuclear physicists of the future. It is, therefore, necessary to rebuild nuclear physics, which has largely come to a stop in this country during the war. This will be an expensive task, since a single piece of apparatus may now cost £100 000. The Committee feels, however, that their proposals cannot be cut down if they are to serve their purpose”507.

Surviving records confirm that the ACAE had considered the funding recommendations on 28 March and made some revisions. Blackett was able to report on 12 April508 that the recommendations had now received the general approval of the relevant ministers, and that the issue of recurrent (rather than capital) expenditure was being worked through. When the Subcommittee again on 15 May 1946509, the agenda carried as an attachment the proposed contract with the university, together with a letter to the Vice-Chancellor. The contract stated that the

506 Anderson to Prime Minister, 29 March 1946. PREM 8/3703, UKNA. 507 Ibid. 508“Nuclear Physics Research”. Note by Chairman. ACAE (NP) (46) 23. 12 April 1946. Box 22X/8 (31), SCUB. 509 Agenda of Meeting of Nuclear Physics Subcommittee of ACAE, 15 May 1946. Box 22X/8 (31). SCUB. 309

equipment to be built would revert to the DSIR once the research was complete. Results arising from the research were to be published but the Department reserved the right to determine if publication was in the national interest, "this being seen a consequence of the unusual circumstances attaching to special grants in nuclear physics". The DSIR was to be supplied with copies of all results prior to publication. There is no evidence as to whether or not permission to publish was ever withheld.

The covering letter to the Vice Chancellor stated that in making the grant, the Council had given consideration to "the exceptional timeliness and promise of research into the acceleration of particles to very high energies being conducted under Prof ML Oliphant at University of Birmingham and particularly the need for certain expensive equipment and additional staff”.

Conclusion

A succinct summary of the conventional wisdom regarding Oliphant and the origins of the proton synchrotron came from Seidel. Discussing the beginnings of the Bevatron, he commented “In Birmingham, Mark Oliphant had already embarked on a project inspired by his wartime stint in the Radiation Laboratory and McMillan’s principle of phase stability to build a 1.3 BeV proton synchrotron”510.

We can now see that this is an inadequate view. To begin with, it suggests that Oliphant was a follower rather than a leader. The evidence and argument laid out in this chapter leaves little doubt on that matter, even if we have along the way corrected Oliphant's own recollections. He independently conceived such a machine, without

510 Seidel (1983), p. 393. 310

input from any other source, and put in place efforts to secure funding for a proton synchrotron at least a year in advance of any similar effort elsewhere.

We have noted that Oliphant cannot have been aware of McMillan’s work at the time he began to frame his ideas for the new machine. McMillan did not publish his paper (nor indeed was Veksler’s paper available in English) until after Oliphant had placed his proposal before the funding authorities. Oliphant had been musing on the matter for most of a year before any similar thoughts began to stir in McMillan’s mind. Furthermore, when McMillan did publish, he said little about accelerating protons; the machine he later built accelerated electrons only. Oliphant considered accelerating electrons, but it was soon apparent his interest was primarily in protons, as it had been since the early 1930s.

Was Oliphant “inspired” in this venture by his year at Berkeley and Site X? Certainly the latter months of that period provided the settings of time and place in which the ideas began to bubble up, even if he first wrote them down in Birmingham. He claimed to have had positive feedback from Lawrence. What is clear is that the “capital” which he continued to accumulate during his time on the Manhattan Project, and which we summarised at the end of the previous chapter, was of great value as he worked his way through conception, design and funding issues in the years spanning the end of the war.

As we have worked to construct a chronology of this period, a number of issues have been clarified. It is apparent that, while Oliphant, McMillan and Veksler each generated versions of what we have called the “synchrotron principle” independently of each other, their motivations differed. The American and the Russian were seeking to overcome

311

limitations on the energy achievable with a cyclotron, proposing a technical fix to the problem of relativistic mass increase.

The Australian was driven by a different motive. He sought to generate beams of the maximum possible energy, wanting to create in the laboratory reactions so far seen only in cosmic radiation. He set a target of 1000 MeV, later linking that to the rest mass of the nucleon. This drive for energy was the latest phase in a quest he had been pursuing for a decade and a half. Unlike the other two, he was by profession a builder of accelerators. For him the problem with using a cyclotron to reach his target was that it would be impossibly massive and costly. Hence the need for a new approach.

We have also addressed the issue of whether Oliphant understood phase stability, or, to put it another way, whether he was justified in his confidence that the pioneering machine would actually work. The answer appears to be “yes”, even if that may not have been apparent to others, and even if his understanding was more physical than mathematical. He was certainly able to overcome the doubts of his colleagues during the funding process, though that may have been as much by force of personality as by argument. In the long run, he was vindicated. The pioneering accelerator did work, if not as well as he had envisaged.

Overarching these issues is our awareness of how Oliphant was able to spend his “capital”, both innate and acquired, to pursue his goal. He had behind him nearly two decades of experience in the building of particle accelerators, characterised by a search for ever-increasing particle energy. This included awareness of the issues involved in funding such enterprises. His own innate technical ability and imagination made him willing to contemplate challenges that would have daunted many others.

312

Without any other project of similar magnitude to inspire him, he willingly confronted the task of creating the largest and most complex piece of scientific apparatus ever constructed.

His reputation, flowing from his work on the Manhattan Project, gave him status and momentum and he was well in touch with the leading figures who could influence the outcome of his endeavours. He possessed notable doggedness and determination, and a volubility and persuasiveness which enabled him to carry arguments and overcome opposition. We can argue that, taken collectively, these qualities uniquely positioned him at that time to undertake such an endeavour. Put more simply, no one else could have done it or would even have attempted it, certainly not in post-war Britain. We need to look now at how his vision became reality and whether, on summing up the outcomes, the enterprise was worth the effort.

313

CHAPTER SEVEN The Birmingham Proton Synchrotron 1946 to 1953: Design and Construction

We now come to the heart of this investigation. We have examined the circumstances which have led to this pivotal moment in the growth of experimental nuclear physics. We have sought to answer the key questions “why Oliphant, why Birmingham, why now”?

In his overview of the spectacular post-war growth of this discipline, Livingstone wrote511, in reference to the newly-announced principle of phase stability,

Seldom has a new scientific principle been exploited with such promptness. The reason was that it was announced just at the end of World War II when scientists returned to their laboratories from their wartime assignments, eager to resume research activities and equipped with new skills and expertise. Many new technical devices and materials were available, which had been developed during the war in fields such as microwave radar, electronics and nuclear physics. Experience in wartime crash programs was carried over to speed up accelerator developments. But the most significant was the increased national prestige of scientists, which brought prompt and generous support from their governments.

How clearly these considerations applied to Oliphant. While he had not been aware of the phase stability principle in its fullest exposition at the

511 Livingstone (1969), p. 43. 314

time he first proposed to build a proton synchrotron, he fitted the rest of Livingstone’s profile. He was fresh from the war, keen to advance his field, with new technical insights and with experience in pushing ahead boldly into new fields. His prestige and his capacity to inspire had brought the public funding support his project needed.

With funding for his enterprise secured from both the DSIR and the Nuffield Foundation, Oliphant was now ready to start construction. This was to take seven years, if we take the first successful operation of the machine in July 1953 as the conclusion of that phase, though Oliphant himself was involved on site for only the first four years. Of course upgrading and fine-tuning of the synchrotron continued for years following 1953. Oliphant, it will be remembered, had claimed in his application for funding that design and construction would be completed in three years.

Sources

The major purpose of this chapter is to construct a narrative covering the building of the Birmingham proton synchrotron (PS), from the recorded first gathering of the design and construction team in September 1946 through to production of the first beam at full energy in July 1953, and to draw from that narrative a number of conclusions.

As to sources from which we can construct such a chronology, we gain only a little insight from Oliphant’s own 1967 memoir512 on the building of the machine. Of its brief seven pages, less than one is devoted to the design and construction phase. Being short, that section is worth quoting in full, as it serves to introduce some of the difficulties the construction team faced.

512 Oliphant (1967), Part 2. 315

The subsequent period of design and construction proved longer than we had anticipated. Unexpected problems and delays occurred. I was disappointed that the electronic art at the time was unable to cope with the production of equipment varying the frequency of the oscillator in step with the rising magnetic field. It was able to provide only restricted trimming of a pre-determined function timed mechanically, with mechanical operation of a tuning inductance necessary to maintain a reasonable matching of the load to the frequency. The absence of straight sections made the design of the accelerating resonant cavities very difficult, while the necessity to use insulating porcelain sections from which to build a vacuum chamber introduced formidable problems. The engineer insisted upon the use of special stranded copper conductors for winding the magnet, instead of copper strip that I had wished to use. Though this was easier to wind, as it was flexible and could be made in long lengths, supply was much delayed, and supporting the flimsy material against electromagnetic forces proved difficult. The method adopted for clamping the pole-tips to the yokes was not sufficiently strong, and gave way on initial tests of the magnet at full field513.

Here Oliphant was apparently endeavouring to briefly explain why the completion of the project took so long, more than six years rather than the initially suggested two or three. Of the five problems Oliphant recalled, at least one, the absence of straight sections in the path taken by the protons, was a consequence of the design, which had been set early. As we shall see in a later chapter, the value of having straight sections was quite quickly realised by Oliphant and others and was discussed with a US visitor in 1947, but it was too late to make any

513Ibid. pp. 6-7. 316

change in the Birmingham machine. The inability of the electronics of the day to generate the needed variations in the accelerating field was also realised early and a workable solution found. This did not appear to impede progress. We will return several times to the issue of the vacuum system, which was one of the last challenges to be resolved, but it too was ready by the time the other elements were in place.

The other issues related to the construction of the magnet, to its design and to the materials from which it was built. He referred to two specific problems, the so-called “Great Pole Pieces Disaster”, which occurred in 1951, and a problem in the following year which Oliphant attributed to the choice of material for the coils. As we shall see, we can attribute some of the delays the project encountered to these issues, which can be put down to lack of experience in constructing a magnet of this size. Even the construction of the magnet itself took much longer than anticipated.

A well-documented and comprehensive account of this endeavour has not previously been written. We are able to draw on the relevant sections of John Lawson’s 1997 monograph on early British synchrotrons514, which he describes as “an informal history”, but it is relatively brief, about ten pages. Its writing did generate some interesting and quite detailed information in the form of letters to Lawson from pioneering participants such as John Symonds515 and Len Hibbard516.

We also have the centenary history of the Department (1880 to 1980) by Moon and Ibbs517. This devotes several pages to the synchrotron, with a

514 Lawson (1997), pp. 24-34 515 Reference needed. 516 Reference needed. 517 Moon and Ibbs (1981). 317

perhaps inevitable emphasis on the problems that arose. This is coloured by Moon’s personal experience, having taken charge of the enterprise from mid-1950 following Oliphant’s departure.

Other sources include the proceedings518 of the symposium held to mark the 40th anniversary of the first operation of the machine. This has some very useful recollections and insights, but it is fragmentary, and very few of the contributors had personal involvement with the design or building phases of the project; most were involved with it in later years as experimenters. We also have brief reports gleaned from official university publications and annual reports and from a small number of letters that refer to the project.

The Minutes

The most comprehensive and continuous, if not fully complete, record of progress on the machine is found in the Minutes of the regular meetings of the research team. These are contained in two regular-sized exercise books, together with a third smaller notebook used by the minute taker to record notes from which the full Minutes were later written up. These records are held in the University of Birmingham Library Special Collection519. The bulk of the Minutes were taken by Dennis Bracher, though others took that responsibility if he was absent.

The Minutes are a good guide to the membership of the synchrotron team, if we work on the assumption that all members would have attended at least some meetings. The meetings were never very large, reflecting the size of the team. Over the 31 meetings and two and half years of the first incarnation of the committee, attendance averaged

518 Rolph (1995) 519 Minute Book of the Synchrotron Committee (Book 1). Box 22X/8 (31) UBSC. 318

about nine, never more than fourteen, sometimes dropping as low as six. Some attendees could almost be taken for granted: Oliphant, Gooden, Symonds and Bracher, with Taylour and Hibbard once they arrived in Birmingham. Peckover and Robertson were early strong attendees but dropped out later when they moved away, perhaps after completing their degrees. Hide did likewise.

Many names come and go in the Minutes, often after a single meeting. Moon attended the first meeting but only one thereafter up until the hiatus in the Minutes from 1949 to 1951. The attendees were not exclusively researchers. From around April 1947, “Mick” Cornick, leader of the technical team (and know to some as “The Gaffer”) was a frequent attendee, and MP (“Jimmy”) Edwards, a technician who had come with Oliphant from the Cavendish and would later go with him to Canberra, was present at meetings from late in 1947. Two other technicians, Bull and Bowyer, came to most meetings over about three months from March 1947, most likely because the magnet steel was being erected at the time.

The enterprise recast

Barely had the funding been secured before Oliphant began to make substantial changes to the proposal which had secured it. These proposed changes are outlined in a document held in the University of Birmingham archives520. The existing copy, the latter part of which survives only as a hand-written draft, does not clearly indicate the purpose for which it was written, but it appears to be a report to the Advisory Committee on Atomic Energy which had recommended financial support.

520 Memorandum Changes in the Birmingham Program. Undated. Box 22X/8B, UBSC. The surviving pages appear to be part of a larger document. 319

Though the document is undated, we can guess at its timing. The opening paragraph refers to funding having been approved by the Nuclear Physics Subcommittee and the Advisory Committee on Atomic Energy, so it cannot have originated before February 1946. As we shall see, design of the machine following the amended specification was underway by September 1946. So the document must have been written some time in the intervening six months, more likely toward the end of that period than towards its beginning. The opening paragraph also comments that “construction has commenced”, but that, in Oliphant’s usage, could mean that it was soon to commence.

It appears that Oliphant and his team had made a reality check on the costs of available technology. The plan to use an air-cored magnet energised by heavy currents from a bank of condensers had been based on estimates provided by Philips. They had quoted £80 000 for a suite of capacitors meeting certain requirements. Oliphant had used that figure in his submission to DSIR. Now, the document claimed, Philips wanted an astounding £350 000 for the same hardware. Quotations from British firms were better but still daunting, ranging from £179 000 down to £140 000. Oliphant had only £120 000 to build the whole machine521.

The answer to the dilemma was, ironically, to turn back to the proposal of January 1945, largely abandoning changes made since. The magnet would now contain iron and the field split into two parts. This would halve the energy needed to energise it but require some 500 tons of high-grade laminated iron. The laminations were needed to reduce the “eddy currents” which would otherwise limit how fast the magnetic field

521 A letter from Moon to Oliphant dated 5 June 1946 referred to quotes of the provision of banks of condensers, including one from the firm T.C.C. for £114 100 for 20 000 microfarads. See F 22, Moon Papers, UBSC. 320

could be increased. It was doubtful if such a mass of quality steel was available. “We were advised that the provision of this iron would be a difficult and probably an impossible task as electrical steel is already a bottleneck in the electrical industry.”

Here we have one reason, among many that might be quoted, why the times were far from propitious for a venture such as Oliphant was undertaking, and why the rival American enterprises, not limited by being immersed in an economy recovering from the devastation of war, had such an advantage.

Determined to press ahead, the ever-inventive Oliphant had other options. The high quality “electrical steel”, enriched with silicon, could be replaced by plates of cheaper low-carbon mild steel. More steel would be needed to produce the required magnetic field, 1000 tons instead of 500 tons, and the rise-time of the magnetic field would have to be stretched out to a full second instead of the 1/25th second then being considered (the June 1945 proposal talked of a 1/100th of a second rise time). But the costly banks of capacitors could be replaced by a much cheaper motor-driven generator producing direct current at 10 000 amperes and 1000 volts.

Oliphant already had a possible supplier of such a generator, the Newcastle firm of CA Parsons, who could have one ready for installation in 18 months at a cost of £40 000. He had already ordered the needed steel from Colvilles in Glasgow (the firm which had supplied the steel for the cyclotron). He relied on DSIR, which he described as being “extremely cooperative”, to secure the allocation of steel (another indication of the supply challenges of the times).

321

In terms of costs, there were swings and roundabouts. The extra expense of the iron-cored magnet over the air-cored coil it replaced would be off-set by savings on the source of the current to power it. The total cost would still fall within the original estimate. No change would be required in the ancillary equipment. Furthermore “There is no change whatever in the principle of operation or in the results expected, except that we now feel even more confident that we shall reach energies well above 109 e.V and perhaps as high as 1.4 x 109 e.V.”

It is not quite true that there was to be “no change in the ancillary equipment”. There was a major change that meshed with those already discussed. With the much longer rise-time of the field, particles could now be injected at much lower energies; this would make the proposed use of a cyclotron as an injector irrelevant. It may be, as some commentators have claimed, that the use of the cyclotron as an injector had already been seen as quite impractical, once the decision was made to locate the new machine on the edge of the existing cyclotron pit. How would the particles produced there be injected into the synchrotron five meters higher?

In closing, Oliphant, while acknowledging the great support provided by DSIR, alluded to the ongoing difficulty in sourcing supplies in the current climate, especially as regards “smaller equipment”, without being able “to quote a contract number”. For example, a large sheet of copper could not be secured in less than nine months “in the absence of priority”. Would it be possible somehow to gain such priority without going case by case to DSIR or the Ministry of Supply? It is not known what answer he received.

Scoping the task

322

It is hard to exaggerate the magnitude of the challenge Oliphant and his team had chosen. We can argue that no machine combining such size with such complexity had ever been built, or even proposed, certainly not for purely scientific purposes. The largest optical telescopes, such as the 5 meter Hale reflector then being built at Mt Palomar in California, might match it for tonnage, as would Lawrence’s still incomplete 184 inch cyclotron. Yet while those had to be built with comparable precision, neither would make the same demand on their operators. Even large telescopes can be largely guided by hand if need be. The Birmingham machine would need to be controlled by systems that once set moving would initiate a complex series of pre-programmed events within a second. Perhaps the nearest equivalent would be Charles Babbage’s never-built 19th century “analytical engine”.

The proton synchrotron, when complete, would comprise at least six interlocking systems:

 a massive annular magnet, to be built up from more than 800 tonnes of mild steel, in the form of 2400 one- centimetre thick plates arranged in a ring, and wound with thirty tons of copper wire.  a ring-shaped tube through which the particles would circulate and which had to be so emptied of air that its internal pressure would be only one billionth of a normal atmosphere (with its supporting pumps this would be known as the “vacuum system”.;  a device (“the ion source”) for generating a flow of protons and accelerating a beam of them to several hundred KeV to be fed into the machine (a decade before, this would have been judged a significant achievement in itself.

323

 technology to generate an electric current that alternated in the megacycle range (“the radio-frequency” or “RF” system) and which would be used to accelerate the circulating protons. The greater challenge would be to tune this circuit so that the frequency could be varied with precision over a 40 to 1 range within a second.  apparatus to produce electric currents in excess of 10 000 amperes to energise the magnet and so generate the annulus of magnetic field to hold the protons in their narrow orbits. This would be made to rise from zero to peak in a second, with the cycle repeated every 10 seconds.  most crucially, a control system so that the accelerating electric field and the guiding magnetic field could be kept in step as the protons were accelerated, so that the protons could be confined to a evacuated space only a few centimetres wide.

To those would be added in time, a range of devices to collect and analyse the particles produced by interaction of the beam with targets of various kinds. The totality would enable a continuation of the way nuclear physics had been done since the days of Rutherford, though the bombarding particles would be a thousand times more energetic than he had employed, and used with much greater precision and control.

1946: Getting going

The first recorded meeting of the synchrotron team was held on 17 September 1946 and was numbered, as were all the subsequent meetings until #31 on 18 May 1949. Initially the meetings were held weekly, but within a year they had become much less frequent, with

324

months passing between meetings. This appears simply to indicate that in the early days of the project designs were very fluid, and everyone could make an input to the discussion, whether of the magnet design, the high-tension system or the vacuum system. As the months passed, designs became more settled, and responsibility passed into the hands of individuals or small groups. There was then less need for freewheeling discussion.

As the leaders of the enterprise, Oliphant and Gooden were almost always present (Gooden missed no meetings in this period, Oliphant only one) but they did not necessarily set the pace. Once the meetings became less frequent, they appear to have been commonly called to discuss some particular issue. The Minutes indicate that the member who had called them together kicked off the discussion. However they also suggest a significant, perhaps dominant, role for Oliphant in these meetings as a source of ideas for others to discuss, as a commentator on ideas that others advanced, and as always being hard to convince.

It was certainly Oliphant who set the tone at the start. He was keen to move ahead. The Minutes reported “Professor Oliphant urged that the design of some of the larger items be pressed forward, in order that materials may be obtained. This was especially necessary in the case of the magnet.” Indeed discussion at the early meetings focused on the issue of the magnet. It would not prove the most expensive component of the PS but it was the most complex and ultimately troublesome.

A week after the first meeting of the committee, Oliphant wrote to Lawrence522. The content of the letter shows how some of the sensitive political considerations of the time were forming a significant impediment to rapid progress.

522 Oliphant to Lawrence, 24 September 1946. LPBL. 325

I understand that some hitch has arisen over visitors to your laboratory.523 I’m very anxious that our synchrotron and cyclotron boys should have an opportunity of talking over some of the fundamental problem with McMillan and others in your laboratory. Are you able to give me any idea of when would be an appropriate time for them to come? If there are difficulties in the way at the moment please do not hesitate to say so.

We’re almost finished the design of the synchrotron and hope to get started on construction immediately. Gooden has written up some of his material on the theoretical side, and will be sending copies to you for McMillan. It would be very helpful all round if he could have the opportunity to discuss these fundamentals and make sure we are in agreement.

The letter has a number of points of interest. We see evidence of Oliphant’s unquenchable optimism or, more brutally, his lack of realism. With only one meeting of the team behind him, and with many issues yet to be addressed, let alone solved, he could declare the design of the machine “almost finished”. The start of construction was still at least some months away, rather than happening “immediately”. We also see that Oliphant was keen to interact with Lawrence and his team on the new hardware. As we shall see, security considerations would make that difficult for some time to come.

As early as the third meeting of the synchrotron group, talk had turned to the challenge of the RF system; this would generate an electric field,

523 Oliphant was here referring to some of the (supposedly unintended) consequences of the McMahon Act, recently passed through the US Congress to control access to information about atomic energy and impacting on interaction with previously collaborating nations like the UK. These implications were later softened. Gooden (and John Fremlin, whose interest was in the cyclotron) were able to visit Lawrence some years later. 326

oscillating at radio frequencies (several megahertz), to accelerate the particles, as was the case in the cyclotron. To keep the protons in their narrow orbit, this field had to be locked into the rising strength of the magnetic field. But how was this to be done? Oliphant at once began to envisage various forms of variable inductance.

On 15 October (meeting #4), Oliphant raised the need to produce a paper describing the work done to date. This was “overdue”, he said. He envisaged the paper starting with the reasons for wanting to accelerate particles to high energies, the limitations of other methods, what was proposed, the issues of orbital and phase stability, injection and extraction of the beam, the design of the magnet, the RF system and the vacuum system, though the latter was yet to be settled. It would be a mistake, Oliphant commented, to say too much about matters not yet settled. Section Ten of the paper was to cover “expected performance”. The Minutes reported that no-one had been allocated to the “dangerous field” covered by this item. By November the one planned paper had become two, divided between the physical and the mathematical. Neither was to discuss the RF system.

At the fifth meeting on 21 October, Oliphant produced figures to show he was already aware of the limitations of the new approach. The proton synchrotron was to produce a beam of 1000 MeV (1 GeV or 109 eV) protons. With the magnetic field strengths currently available, this required a magnet 30ft. in diameter, weighing about 1000 tons and costing about £100 000.

These various dimensions would rise more or less linearly with the energy produced. So a 10 GeV machine would be 300 feet across, with a magnet of some 12 000 tons and costing more than a million pounds. These figures would be ten times larger again for an accelerator to

327

deliver a hundred GeV. Oliphant concluded that “thus an energy of 1010 eV may be possible with a synchrotron but not 1011 eV.524, 525 The Minutes also report a comment, probably by Oliphant, that someone in the United States (it is not stated who) had estimated that “a plant” for producing 1010 eV would cost around £25 million in British money, but their costs were “perhaps four times ours”.

November saw discussion focus on the method of energising the synchrotron magnet. Oliphant was now proposing to use a “super exciter”, a 1000 horsepower motor- generator, in place of the banks of capacitors as proposed in the application to DSIR. As we have already noted, this was a significant shift in the design, driven by the unacceptably high cost of the capacitors. He had already been in discussion with the firm of Parsons in Newcastle, and was keen to accept their quotation, though English Electric apparently was still a possible supplier. The generator would not be ready to run for eighteen months. The set constituted a significant technical challenge, requiring a current to rise from zero to more than 10 000 amperes in a second. Gooden reported that a couple of firms were still interested in making the capacitors, should the project revert to that approach. By now the first steps had been taken to engage the Glasgow firm of Colvilles who had provided the cyclotron magnet, to fabricate the 800 tonnes of steel plate needed for the synchrotron magnet.

1947: Construction begins

524 Modern synchrotrons have of course vastly exceeded these limits, reaching beyond 1000 GeV. The availability of much greater financial resources has been one contributing factor, together with two significant technical developments: the use of the “strong focusing” principle which narrows the beam and allows the use of smaller, lighter and cheaper magnets, and the advent of super-conducting magnets to produce much stronger fields. 525 Oliphant was to quote similar figures in Oliphant, Gooden and Hide (1947). 328

When the group reconvened in early February, after the Christmas holidays, the Minutes report an “argument” between Oliphant and HH [“Bob”] Taylour, a newly appointed Nuffield Fellow whose background was in engineering rather than physics. Taylour had been engaged specifically to work on the magnet. It was one of a number of disagreements between the two recorded either in the Minutes or in the memories of participants in the project and reflects, on Oliphant’s often- reported attitude to engineers.

The subject on this occasion was the material from which the vacuum system (the evacuated pipe through which the particles would orbit) would be made, or more precisely the thickness of the material, since Oliphant was at the time insisting on a metal tube. Taylour, relying on engineering data, was concerned about differential expansion of rings 30 feet across. Oliphant was as always more intuitive, declaring that “thin material will do”.

As spring progressed, the issue of the RF system, or, more precisely, its tuning, came in sight of a solution. Since, at that time, the particles were to be injected at 250 KeV and then accelerated to forty times that energy, the RF field being fed into the accelerating tube known as the “cee”526, would need to be varied in frequency by the same forty-to-one ratio in a second. That was a severe challenge. The Minutes report that Oliphant wanted something simple, and that John Gooden responded by proposing that a coil of wire be dipped into a pot of mercury. The turns of wire, wound on a club-shaped former, would be successively shorted out as the mercury covered them, so varying the inductance of the turning circuit and so the frequency of the RF field. The initial plans

526 The name was given presumably by analogy with the “dees” found in cyclotrons. The accelerating tube would surround the orbit of the particles for some fraction of the circle, and was therefore shaped like a letter “C”. 329

called for the coil to be held fixed and the pot of mercury to be raised and lowered; later these movements were reversed.

Robinson, who was involved with Walter Styles in the refining of this piece of technology, later confirmed Gooden as the source of this concept and added some details to the record, at the same time hinting at some of the dangers of such research.

Oliphant blew into the lab and clasping his fat fingers said “This is the way to tune the cee, with a variable inductance made of intersecting fingers.” Then John Gooden came up with the idea of an inductance dipping into mercury and we took it from there.

We all should have had mercury poisoning. We had a mercury still in the same room….At one stage I went to the university medico about it. He said “Are your teeth getting loose?” I said “no”. “Ah well, that’s OK”.

I passed through Birmingham in 1954 and couldn’t believe it. There was the mercury inductance dipping away. No one had bothered about giving it a nitrogen atmosphere and it worked. Amazing. Later on David Caro designed a ferrite inductance. We had no ferrite in my day527.”

By the end of April 1947, we see the first appearance in the Minutes of the name of Len Hibbard528, who was to play such a significant role in the enterprise. The 30 April Minutes also contained a report from John Gooden that “the condenser people had recently visited him and had

527 Robinson to Lawson, 11 October 1995. Moon Papers B29. UBSC. 528 Leonard Hibbard (1914-2006), Australian-born physicist and engineer, arrived to join the team in April 1947, after involvement in the development of radar. He would follow Oliphant to Canberra in 1953. For some background see http://radarreturns.net.au/index.php?content=tribute&obituary=49 330

made a pitiful attempt to get us to change back to the early method of producing the field. They wanted £300 000 for the necessary equipment.” The Minutes continued “a lighthearted discussion of the relative merits of condensers and machines followed, in which Professor Oliphant (condensers) and Taylour (machines) were prominent and a good time was enjoyed by all”.

At the May meeting, progress on the construction was indicated by a reference to the need for 160 bolt holes to be drilled into the concrete floor on which the synchrotron would stand. Some of the preliminaries to this can be gleaned from Bracher.

One day in 1947 (or so) the inhabitants of the Nuffield building heard a noise like pile-driving echoing though every room; it was pile-driving, in fact, in preparation for the magnet of the synchrotron.

It was necessary, this pile-driving, because the top soil was largely sand, approximately down to the cyclotron pit floor level. Therefore the piles had to be driven until each one reached solid ground; then they were filled with concrete.

A raft was constructed from concrete on top of the piles, approximately at floor level, and rails were fitted on the raft to carry the magnet yoke plates. This raft had air passages left in the surface to cool the magnet’s coils529.

It was into this raft that holes were to be drilled. Further details of the foundations were later provided by Hibbard530. He reported that the

529 Bracher DF “The Birmingham Proton Synchrotron”. Moon Papers B22, UBSC. 530 Hibbard in Inall (2000). 331

concrete raft sat on top of 30 piles driven down to bedrock. The rails on which the magnet rested allowed the yoke plates to slide in and out as the metal expanded and contracted.

Writing to his former superior EG Bowen at the CSIRO Division of Radiophysics on 29 April 1947, John Gooden summed up the state of affairs as follows:

Professor Oliphant no doubt told you in some detail some of the problems we are tackling with the synchrotron. The stage is now reached when we feel that most of the fundamental paperwork is completed. We expect the steel for the magnet to begin arriving in a matter of weeks and then assembly should take about 6 months. Experimental work on an R.F system will begin very shortly. Hibbard has settled in and is working with me on the synchrotron. He is tackling the R.F. system531.

With the meetings less frequent from mid 1947, it is harder to keep a detailed track of progress. The meeting in September was called to discuss the erection of the steel for the magnet, then imminent, with a particular need to stop the thin plates buckling as they were transported into place.

The magnet parameters were later provided by Hibbard532. As finally built, the 810 ton magnet consisted of some 2400 low carbon steel plates, each cut in a C-shaped profile with the gap outwards. The flat plates, one half inch (approximately one centimetre) thick touched at their inner edge. To give the required annular shape to the magnet, wedge plates were inserted from the outside after each pair of plates.

531 Gooden to Bowen, 27 April 1947. EBA. 532 Hibbard (1950), p. 34. 332

The wedge plates opened up gaps between the main plates further in, and cooling air was to be drawn through these gaps from the cavities in the raft at the rate of 15 000 cubic feet a minute by a large central fan in the low roof of the room. The plates were to be welded together on the inside, the outside and the upper surface.

Details of the assembly of the magnet were later provided by Bracher. The magnet yoke plates were transported one at a time by a modified fork-lift truck across a bridge constructed over one corner of the cyclotron pit to give access to the synchrotron site. They were then hoisted and lowered into place using a trolley that ran along a girder between a post set in the middle of the magnet space and an A-frame on wheels able to move azimuthally around the magnet space. Once into place, each yoke plate, standing vertically, was welded to its neighhour and/or to the rail imbedded into the concrete below.

The yoke of the magnet was constructed in two parts, one opposite each other, the two sections having approximately equal plates at all times, in order not to strain the concrete raft unduly; this posed the problem of how to arrange that the gaps of the two sections of the yoke met satisfactorily when the magnet yoke was complete. This was solved by a long rubber tube and glass end- pieces, filled with water, and two tele-microscopes533. This apparatus was in fairly constant use in building the two halves of the yoke534.

Here is an early example of the sort of logistic and constructional ingenuity which would be needed at every step of this pioneering project. There were very few rules or precedents.

533 According to several sources, this technique was devised by Taylour. 534 Bracher (1995).. 333

1948: Fits and Starts

The synchrotron group meeting on 19 January 1948 was told by Gooden that steel for the magnet was being erected at the rate of “five circumferential feet per week”; at that rate erection would be completed in April or May535. For reasons which are not clear, the task took considerably longer. According to Symonds536, based on photographic evidence, the last of the yoke plates was not in place until late in September. The Minutes suggest completion was a few months earlier than that. On other matters, Hibbard reported that the dipping mercury inductance was “working much better than expected”, while Symonds stated that progress was being made on the proton source. By now the debate over the vacuum system had been closed with a decision to use electrical porcelain for the piping.

Six months later, on 13 July 1948, the magnet steel being in place, the coils were being wound and attention turning to the making of the pole tips and discussion on how they should be clamped to the magnet yoke. The porcelain pipes intended for the vacuum system had stood up to initial vacuum testing and were now going elsewhere to be tested to destruction. Oliphant reported that the motor-generator set was now ready to be tested at the Parsons factory. Delivery was still some time away, but Taylour could not be sure the building to house it would be ready in time. Oliphant also had a request, that the buttons and meters on the control panel should be big. He maintained that those on the cyclotron control desk were too small and too hard to read.

535 In a letter to Lawrence dated 20 January, Gooden stated the he hoped the magnet yoke would be completed “some time in the early summer”. Gooden to Lawrence, 20 January 1948. LPBL. 536 Symonds provided this information, and copies of the relevant photographs, to the author during an interview in 2010. Notes of interview in EBA. 334

The Minutes also reported one of the not-infrequent disputes involving the engineer Taylour, in this case with Hibbard rather than Oliphant. “A battle between Hibbard and Taylour on the question of armature reaction was stopped by Professor Oliphant just as the contestants were warming up.”

The 29th meeting of the group was 1 October 1948, more than two years after construction began. It is clear from the Minutes that Oliphant, and perhaps others, were growing anxious about the time taken to complete the magnet. The “Prof”, as the Minutes designated him, asked Taylour how long the task would take. Taylour’s response was that “it was impossible to say”. Pressed to set some limits on the timeframe, he declined to do even that. The issue seemed not to be one of manpower. Four men were available to work on the job. There was even some lively discussion as to whether or not they had enough to do. Progress was evident on other fronts. Symonds was now getting one milliamp of protons from the ion source. Bracher was talking to members of the team about the controls and instruments needed for the control panel, so that he could come up with a design for discussion.

News of the progress of the project was passed to interested parties in the general public through the University’s annual reports. The report of the Joint Standing Committee for Research for 1947-48537 was succinct. “[The project to build] a synchrotron to accelerate protons to energies of over 1300 MeV has passed from design to construction. The main steelwork for the magnet was erected during the year”.

We should note the report on progress on the cyclotron project, running essentially in parallel with the synchrotron, though started nearly a

537 Report of the Joint Standing Committee on Research, University of Birmingham, 1947-48. p. 9. [Hereafter this source will be cited as “JSCR” for the particular year.] 335

decade earlier. “The cyclotron has been in operation with a peak amplitude of 300 KV between the Dees and a circulating beam of hydrogen ions of energy 26 MeV, enough to melt a 1/8th inch silicon plate with the production of a substantial amount of radioactive silicon. Work is proceeding on auxiliary apparatus needed to bring the beam out of the vacuum tank.”

Oliphant’s future

Another development in 1948 was to have a profound influence on the progress of the project, both short term and long term. In his Annual Report for 1948, Raymond Priestley, the Vice-Chancellor at Birmingham, reported that earlier in the year Oliphant had taken up the post of Vice-Principal of the University. He had replaced Norman Howarth the Mason Professor of Chemistry, who had retired at age 65. Priestley was very happy with the new appointment. “In [Oliphant], again, the university secured for its Vice-Principal a scientist of outstanding ability and a man of quite exceptional personality”538. He went on

To those not in the know it was something of a shock when Professor Oliphant’s acceptance of the Vice-Principalship was followed almost at once by the public announcement from Australian government sources that he would be shortly leaving for Canberra to take over the Directorship of a Research Institute in the new Australian National University. Personal knowledge of Canberra’s wide empty spaces and of the sort of equipment and buildings required for modern nuclear physics research gives me some assurance that we in Birmingham are likely to be able to rely for some years yet on his efficient direction of our own

538 Report of the Vice-Chancellor, University of Birmingham, 1948, p. 3. BUA. 336

Physics Department and Nuclear Physics laboratories, and upon his wise counsel in university affairs. For the rest, he is an Australian and we must accept the justice of his people’s claim, while thanking our stars that Cambridge and Birmingham have been able to retain him for many of the most productive years of his working life.

So Oliphant was to leave Birmingham within a few years, with the possibility that the project which he had initiated and directed with such vigour would be incomplete at his departure.

We need now to fill in some of the background to this startling announcement. Construction of the synchrotron was far from being Oliphant’s whole universe. He had a department to run, appointments to make539, lectures to give. In May 1946 he went to New York to serve as an advisor on atomic energy matters to the Australian delegation, led by Foreign Minister HV “Doc” Evatt, to a session of the fledgling United Nations540. In 1948, working with Peierls, he mounted in Birmingham a major international conference on problems in nuclear physics, attracting the cream of researchers in the field including international stars such as Oppenheimer, Bohr, Teller and Fermi541. According to Cockburn and Ellyard, he was much in demand as a lecturer for audiences around the country, particularly on what had become a favourite topic, how the peaceful use of atomic energy would transform the lives of ordinary people542. He continued to attend meetings of government committees, including the nuclear physics subcommittee and its successors, and to promote, without success, the

539 For example, the appointment of Moon to a newly-created chair (later called the Oliver Lodge chair) in 1946 and Sayers to a Chair in Electron Physics in 1948. 540 Cockburn and Ellyard (1981), p. 131. 541 Relevant documents can be found in Box 22X/4 BUSC. 542 See for example. Picture Post 27 October 1945. The article is entitled An Atomic Plan for Britain and based on an interview with Oliphant. Box X 22/7 SCUB. 337

electromagnetic separation method as the best way for Britain to secure the fissile material needed for its own nuclear weapons.

The most substantial, and ultimately most influential, of these many calls on this time concerned the efforts of the Australian government to establish a new university in Canberra, ultimately to be called the Australian National University (ANU), with a strong emphasis on research rather than on teaching. The origins of the proposal lay in the closing years of the war. By general agreement, the major credit for the initiative went to the leading Australian bureaucrat Dr HC (“Nugget”) Coombs, the Director of Post-war Reconstruction in the Australian government. According Coombs' own recollections543, it was he, working with the Australian Prime Minister , who decided that the involvement of leading antipodean academics now based in Britain was a key element of the development of the new university. Those would include the medical researcher , who had just won the Nobel Prize for his work on penicillin and who, like Oliphant, was originally from Adelaide, the historian Keith Hancock, the New Zealand- born anthropologist , the physicist Harrie Massey, and of course Oliphant.

According to Coombs, Oliphant met with him and Chifley in London during the Commonwealth Prime Ministers conference (presumably in May 1946) and was introduced to the idea. He (Oliphant) was reportedly very enthusiastic. It is not clear at what stage it was suggested he might be part of the enterprise. However, early in 1947, following a visit by Oliphant to Australia to advise the Interim Council of the new university and, remarkably, to address a meeting of the Cabinet, an agreement had been reached. Oliphant would return to head up the proposed

543 These developments, including the involvement of Coombs and of Oliphant are well recounted in Foster and Varghese (1996), which draws strongly on Coombs own recollections. . See chapter 1 for Coombs and chapter 2 for Coombs and Oliphant. 338

Research School of Physical Sciences and a sum of £500 000 would be made available to establish and equip the School, an increase of £380 000 over the sum initially proposed. Oliphant had apparently made the greatly increased allocation, to be directed specifically to nuclear physics research, the price of his coming544.

The same day as Cabinet agreed to Oliphant’s terms [6 February 1947] Chifley wrote to the British Prime Minister . That letter has not been sighted by the author and may not have survived545, but Attlee’s reply on 4 March 1947 is available. Chifley had apparently informed Attlee of the plans for the new university, and of the hope that Oliphant could be involved. Attlee warmly endorsed the first, offering any personal help he could with”your proposal”. As for releasing Oliphant to play a role, that would be more difficult, at least in the short term.

We should, of course, be very sorry to lose Professor Oliphant at present, since he is in charge of important fundamental research at Birmingham University in nuclear physics. The government felt justified in making a substantial grant towards the cost of the special apparatus needed for this work, largely because of the confidence we feel in Professor Oliphant’s scientific ability and leadership.

I am glad therefore that there is no question of his leaving before the program of work had reached a stage where it could be continued without his supervision.

544 There is a clear parallel with the conditions Oliphant had placed on his acceptance of the Chair in Birmingham in 1936. Oliphant clearly had the capacity, as a vital part of his “capital’, to convince potential employers that “he was the man they wanted”, and, essentially, to bargain to secure what he judged to be appropriate resources to undertake his planned research. 545 The relevant archives have not been specifically searched for this document. In their history of the ANU Foster and Varghese do not refer to this correspondence. 339

The letter explicitly suggested a timeframe of three years for that stage to be reached. If at that time, Oliphant was to leave the UK “it would be with the greatest goodwill on both his fellow scientists and of the British government, whose debt of gratitude for his brilliant work must certainly exceed that for which he feels to them”.

Chifley wrote back on 15 March, welcoming the spirit of cooperation which Atlee had expressed, and underlining his view that building of first class research facilities in Australia would also serve the interests of the “British Commonwealth.” Attlee had commented the Oliphant’s “brilliant work” had mostly been undertaken in the context of the development of the atomic bomb, and around such research much secrecy still clung. That would limit the capacity of the UK government to cooperate in such areas even with other Commonwealth countries. Provided Oliphant was engaged in “fundamental physical research of a non-secret character”, full consultation and collaboration with UK scientists should be possible.

Now is not the place to deal in any detail with Oliphant’s interactions with the fledgling university through the late nineteen forties. We can be sure that that, according to the surviving record, the four UK-based “academic advisers” (Oliphant, Florey, Hancock and Firth), as they were formally designated from early 1948, were required to meet often to deal with a wide range of matters, and to visit Canberra at Easter 1948 to view the site first-hand. Clearly those commitments took time and effort. In addition, there were ongoing tensions. Of the four, only Oliphant had made any real commitment to head up one of the research schools at the new university, and in the long run, he was the only one who did so. Even he had thoughts of backing out, late in 1949, largely over a proposal that Harrie Massey be appointed to the ANU on terms and conditions similar to his own.

340

Of more concern was the impact of planning for the proposed new Research School, and of the equipment it would house, on Oliphant’s ability to focus on the barely-begun proton synchrotron, and the many issues remaining to be solved. He already had in mind for Canberra a radically new type of accelerator which he dubbed a “cyclo-synchrotron” and which would be able to reach 10 or even 20 GeV The biggest accelerator planned at the time was the 6 GeV Bevatron. Oliphant once again wanted to build the biggest.

It was in early 1948, probably at the time of the gathering of the academic advisors in Canberra and about a year after the agreement mentioned above, that the Australian authorities announced Oliphant’s is appointment to the ANU. Priestley implied the announcement was not a surprise to everyone, but it is not clear from the surviving record who within the physics department, or more broadly within the university, knew of his (relatively) imminent departure and who did not. The Minutes of the Synchrotron Committee, which did not meet between 19 January and 13 July, carry no mention of the fact. We are left to speculate on the reaction of his close colleagues to the announcement and their judgement of his actions. Perhaps, like Priestley, they were, firstly, hopeful that he would not be leaving soon, and, secondly, accepting that he had every right to go.

As for Oliphant’s motivations, he expressed them in a letter to Chadwick late in 1948. “My decision to go to Canberra in a year to two has not been without a great deal of heart-searching, but I believe it is right to endeavour to create in Australia conditions for work in science that will enable at least some of the good men to return without losing all the advantages of the scientific way of life in England. I shall finish off the

341

job here and hope to have a year of use of the equipment before I leave it to others546.”

Two aspects of Oliphant’s character are on show here. On the one hand we see again his almost absurd optimism, or lack of realism. Did he seriously think the machine, then barely begun, would be ready in a year, so he could fit in a year’s use before he departed in two? On the other, we see his higher aims, his willingness to put his own career, then at its peak, on the line for the sake of the scientific future of his homeland.

1949: Still Much to Do

When the synchrotron team next met formally on January 1949, the Minutes noted that a lot remained to be done: the control desk, the vacuum system, the sequence timer. Oliphant asked for a big effort to be made on these. Taylour reported that the magnet was “nearly finished”, but it is not clear what that meant. According to the Minutes, the “coil winding machine” would be on site until the middle of March. Steel for the pole tips had been obtained and preparations made for cutting them to shape.

Five months later (meeting #31, 18 May, 1949), Gooden reported there was still little progress on the systems for injection of ions and extraction of the beam, though there had been talk in the Minutes of work on this more than two years earlier. Oliphant’s response was, essentially, “not to worry”. He argued the team should get on with completing the magnet and control systems, and be initially content with an injector good enough to get a circulating beam, if only of a few particles. It appears Oliphant was growing anxious to reach some significant milestone,

546 Oliphant to Chadwick, 3 November 1948, CHAD II 1/10, CPCC. 342

before his now looming departure to Australia. It seemed he would be satisfied with a “proof of concept” demonstration at that stage.

Oliphant was able to report that the generator should be erected in a week or so, but the news from the vacuum system was not so cheerful. Satisfactory porcelain sections were being produced by a firm called Hathware but months would be needed to grind the ends of the sections so that joints could be made sufficiently airtight. There was little to report on the RF system; the ion source, according to Symonds, would be ready in August. Le Cairne, a new member of the team, was working on ways to monitor the beam when there was one.

Progress thereafter becomes less certain, for the Minutes cease. According to the pattern by now established, we would have expected another meeting of the team towards October 1949. But none is recorded in the Minute book, which has only blank pages after the Minutes from 18 May. It is possible of course that a meeting of the group was held and notes taken but detailed minutes not written up. The small rough note book does not support that hypothesis. It too displays a break in recorded continuity from May 1949. Thereafter we find some undated jottings, which appear to be mostly in Moon’s hand but are all but unreadable. The first subsequent date to appear is 3 November 1952.

Where the detailed sources fail, others less comprehensive can fill the gap, at least partly. The report of the Joint Standing Committee on Research for 1949 sums up the year’s relevant activities in the Department of Physics as follows.

The Department’s most ambitious undertaking, the construction of a synchrotron for the acceleration of protons to energies

343

corresponding to about 1 200 000 volts has made good progress during the year. The 800 ton magnet body is complete, its copper winding is in place and has been tested, and the motor- generator unit, which provides the pulses of current through the winding, has been installed by the makers, as has most of the ancillary switch gear. The radio frequency system is in an advanced state of design and parts of it have already been constructed and tested; devices for maintaining its frequency in the correct relation to the changing magnetic fields have been actively worked upon. The porcelain vacuum system has presented special difficulties owing to the limited accuracy with which ceramic tubes can be manufactured and these limitations are still a matter for concern547.

Reports of progress were also reaching the wider public. The October 1949 issue of the popular illustrated magazine Picture Post carried an article by Maurice Goldsmith under the banner The Biggest Magnet in the World548. This reported that “mighty, powerful machines that will help to find out what goes on inside a microscopic grain of sugar are being constructed at the university under the direction of Professor ML Oliphant. These are big and ugly and cost many thousands of pounds. “

According to the article, the synchrotron was not only the largest such machine in the world but also different from any other in that it was to accelerate protons. Construction of the magnet had apparently begun in August 1947. All its impressive dimensions were quoted; 810 tonnes, a radius of 16 feet, 2400 laminations, energies to 1300 MeV. Images accompanying the article showed groups of up to eight people needed to carry out the “exhausting operation” of winding the copper wire coils

547 JSCR 1949-48, p. 8. 548 Picture Post, October 1949. Clipping in EBA. 344

around the magnet. This involved “much straining and pulling”, with 24 turns to each winding, requiring a thousand feet of wire.

1950: A change in leadership

Despite some evidence of progress in 1949, we are still left with the break in the written record, beginning in May 1949 and extending for more than two years. Does this hiatus indicate some crisis in the progress of the enterprise? And, if so, can we identify any significant triggers to such a crisis? One clue may be found by considering events in the lives of the project’s two leading figures, Oliphant and Gooden.

We have already noted the precursors to Oliphant’s departure from Birmingham in July 1950. Whatever loss his immediate colleagues felt, it seems that the university as a whole felt it no less keenly. Writing in his Annual Report for 1950, Vice-Chancellor Priestley commented that

At midsummer, Professor Marcus Oliphant left us for the National University of Australia. This move has been long foreshadowed, but the university’s loss is none the less for that. In him we have lost not only one of the world’s outstanding nuclear physicists, but a colleague who was big in every way. For years he exercised a decisive influence on many aspects of university policy and administration. He, perhaps more than any other professor, can take credit for the good feeling and the friendly atmosphere that is characteristic of the Faculty of Science today.

The University marked Oliphant’s departure and honoured his contribution with the award of the degree of Doctor of Science (honoris causa). Moon, his soon-to-be-successor, later commented on Oliphant’s contribution in the following terms.

345

Apart from his physics, he had been a notable and well-loved figure in the University. Never a “committee-man”, though Vice Principal for a brief period, he was refreshingly outspoken and full of ideas, some of which were much to the general benefit. One that was not was the creation of three separate Boards of Studies for physical, biological and engineering sciences, but the establishment of an Executive Committee to steer the already overlarge Senate was of lasting benefit…. The Public Orator was able to say that Oliphant, though not mathematically inclined, had “multiplied his students by ten, divided this Faculty into three, rationalised the Senate and occasionally squared the Council549"

Mid-1950 was a time of departure in another sense, with the death of John Gooden at the age of 30. It is not clear when the serious nature of his illness became known, though it appears his health had always been fragile, and the disease was at least in part hereditary. In May 1950, Oliphant arranged for the repatriation of Gooden and his family to his home town of Adelaide, where he died on 9 June550.

That these two events, coming so close together, had a major impact on the progress of the project cannot be doubted. The two men had been central to the enterprise, one the visionary, the other the hard-working practical leader. Reporting on progress in the Department of Physics in 1950, the university’s Joint Standing Committee for Research expressed the impact as follows; “The resignation of Professor Oliphant and the untimely death of Dr JS Gooden were heavy and almost simultaneous blows to the synchrotron team, coming towards the end of a year of steady but not spectacular progress. There remains much to be done

549 Moon and Ibbs (1981), p. 26. 550 Gooden died from Bright’s Disease. For an obituary of Gooden, see Smith (1950). 346

before the synchrotron will deliver its first burst of thousand million volt protons551.”

This assessment was in bleak contrast with the good news about the cyclotron. That had come on-line with an external bean within a few weeks of Oliphant’s departure (though more than a decade after it had been commenced), able to deliver 10 MeV protons, 20 MeV deuterons and 40 MeV alphas, and ready to generate radio-isotopes for research. The one concern was with the shielding against the generated X- radiation, which was judged “barely adequate”.

Oliphant’s Chair was quickly filled by Moon, who already held the junior professorship and whose academic interests Oliphant had consistently promoted for more than a decade. Lawson summed up the impact of the change of leadership from Oliphant to Moon as follows.

With Oliphant’s departure at the beginning of July [1950] responsibility for completing the synchrotron fell on Moon, soon to be appointed Poynting Professor. Neither particle accelerators nor high energy physics were close to his current interests and although he was not happy to be “landed” with the project, he tackled it conscientiously and with vigour. It was a difficult year and some of the problems were proving less tractable than anticipated. Furthermore, lack of technical support was causing some of the installation work to move more slowly than planned. Indeed, the original hope of completion by 1950 now clearly could not be realised552.

551 JSCR 1950-1, p. 9. 552 Lawson (1997), p. 28. 347

Lawson can be interpreted here as suggesting that, of itself, the change of leadership made little difference to the progress of the project. Yet there is evidence of Moon’s lack of genuine (let alone Oliphant-esque) enthusiasm for the enterprise. As a Fellow of the Royal Society, he was required to prepare a biographical statement, the final version of which and an earlier draft are preserved among the Moon Papers. This includes the following statement “When in 1938, I left London to join [Oliphant] in Birmingham, he was generous as regards my progress up the university ladder, but I was standing scientifically on my own two feet [and I was privately critical of his ambitious technical program, though I felt bound to support them especially after his departure}553.

Perhaps significantly, the phrase in square brackets does not appear in the final version of the statement. Perhaps Moon himself recognised that the statement sat oddly with the apparent enthusiasm with which he had greeted Oliphant’s earliest ideas on the project. Elsewhere in the final version we read

Mark Oliphant appears several times in this document. He was over-ambitious with synchrotrons, not only in Birmingham but in Canberra; he wrote later “I was an ass to get mixed up in that complex engineering field554.” By contrast, his cyclotron in Birmingham (a variation on Lawrence’s 60 inch design) was basically excellent, justifying much improvement and still in active use 50 years [later].555

Moon also had nothing like Oliphant’s depth of experience in managing projects of this nature, or indeed of general administrative duties, in the years prior to his appointment to head the Department and the project.

553 Moon Papers, A8. SCUB. 554 This quotation is yet to be sourced or verified. 555 Moon Papers, A9. SCUB. 348

As he wrote in his memoir, “The years from 1946 to 1950, when I was a professor with a technical assistant and no administrative responsibility, were my best opportunity, and I believe I used it well”.

We can assume it would have taken Moon some time to get up to speed with the project, and its many complexities, as he had attended only two of the synchrotron group meetings on record to that time. He later recalled his involvement in the synchrotron with affection and for “giving me so many friends”556. Yet at the time, it is doubtful that he was enthusiastic about the job or even the right man for it. For example, it might have been thought appropriate that once Oliphant’s intentions were known Moon might have become more engaged in the projects which he would soon be leading. He might have come to some of the committee meetings. There is no evidence that he did so.

He did however make some necessary changes in arrangements once he took over the role, though perhaps not immediately. By his own later recollection557 he was “apprehensive” on seeing off Oliphant at Snow Hill railway station in the company of Ibbs, having just been appointed to succeed to the Poynting chair, and therefore, by implication, to oversight of the synchrotron. He also admitted that he was “not well trained” for the role. Yet it seems that not until the appointment of WE Burcham as Oliver Lodge professor, as Moon’s previous personal chair had been renamed, in 1951 did Moon seek relief from other duties so that he could concentrate on the machine. Burcham took charge of nuclear physics except the construction of the proton synchrotron. Ibbs was asked by the Dean and Vice-Chancellor to defer his retirement and to take professorial charge of teaching and general administration.

556 See Moon’s comments in Rolph (1995), p. 120. 557 Moon and Ibbs (1981), p. 27. 349

These new arrangements were, in Moon’s words, the consequence of “realising the magnitude of the task” and apparently lasted for the two years it took to make the synchrotron operational. Since that event occurred in mid-1953, the account appeared to confirm the response was not put in place until six months or more after Oliphant had departed, and several years after Moon must have realised that the role would ultimately fall to him.

As we have noted, with detailed reporting of the synchrotron project ceasing in May 1949, and not recommencing until December 1951, a gap of more than 30 months splits the record in the Minute books. This lack of reported progress does not necessarily mean that work ceased on the various elements and subsystems of the machine, but it is very likely that the pace slowed with the dynamism of Oliphant and Gooden withdrawn. This might have been equivalent to several months at least of time lost. Certainly it appears that the synchrotron team did not meet during this period, which would have meant the loss of the collegial spirit previously so evident.

The state of play

In October 1950, Hibbard published the first comprehensive account of the synchrotron558 as its construction stood at that time (or at least at the time Hibbard was writing a few months earlier), and the first since the Oliphant, Gooden and Hide paper of three years before. After four years of design and building, the machine remained a work in progress. There remained, for example, issues with the vacuum system; the porcelain sections had proved sufficiently strong but they were unacceptably leaky. Regarding this system, Hibbard had most to say

558 Hibbard (1950). The subheading of this paper states “Designed to accelerate protons to an energy of 1 BeV, this machine has been under construction for over three years. It is thoroughly described here”. 350

here about the need to line the interior of the vacuum box with metal to prevent build-up of static electricity that had ruined experiments with other accelerators.

He also discussed how about a quarter of the circuit would be turned into the “cee” drift tube. The particles would be accelerated as they entered and exited this section by subjecting them to an alternating voltage which increased in frequency to keep up with them. There was little to say about extracting the beam, since it was envisaged that initial experiments would take place with an internal beam. The brief consideration of the ion source and injection system commented on how much work remained to be done to get the particles into the machine with the required accuracy.

Hibbard devoted more than half the review to the generation of the rising magnetic field that guided the particles, the varying radio frequency voltage that accelerated them, and in particular to how those two could be kept very accurately in step. This was a unique challenge, one which McMillan and Veksler had not seriously contemplated since they had not planned to accelerate protons. Hibbard was not speaking specifically of this problem when he wrote “.... the proton synchrotron presents technical difficulties which either do not exist or are not nearly so serious in other types of accelerators”, but the assessment nonetheless applied particularly here.

The technicalities in the solutions implemented were complex, and stand as evidence of much hard work, thought and attention to detail by the participating physicists and their technical assistants. Indeed they were too complex to go into in detail here, but a synopsis may serve as a guide to their intricacies. It has already been pointed out that control of the inductance of the circuit in generating the varying radio frequency

351

voltage into the “cee” was to be achieved by use of a coil dipping into a pot of mercury. The rate of movement of the coil (and therefore the rate of change of the inductance and so of the frequency) was in turn controlled by a frequency signal generated by beating together the output of two oscillators. One was fixed in frequency and 35MHz, the other varied between 34.66MHz and 25.4MHz, so that the resulting difference frequency varied as required from 0.33MHz to 9.6MHz.

In the not-fixed oscillator, the variation in frequency was achieved by a variable capacitor with a rotating vane, in the form of a carefully-shaped aluminium disc 60 centimetres in diameter which could rotate completely in 1.2 seconds. To lock together the accelerating frequency and the rising magnetic field, the rotation of the vane was in turn controlled by a servo motor which could speed it up or retard it as required. The speed was set by reference to the rising voltage in the magnet windings, since that could not be finely controlled and had to serve as the ultimate reference.

To make the system work it was necessary to know how fast the vane was turning and whether it needed to change its speed to keep pace with the rising field. That was achieved by attaching 120 small mirrors to the rim of the capacitor lander and reflecting a beam of light from those areas into a device to measure the time of arrival. Therefore securing the necessary synchrony between the accelerating voltage and the magnetic field involved a number of steps, and the proof of the soundness of such a complex system came with the ultimate successful operation of the accelerator.

The accuracy required of this equipment, and therefore the tolerances which governed its construction, was never less than impressive, and at times must have seemed daunting. For example the mirrors on the

352

capacitor vane were positioned to within two seconds of arc, allowing time measurement to two microseconds as was required. In much of the equipment, components such as the pole pieces had to be cut, shaped and fitted with tolerances of a thousandth of an inch (0.04 mm) or less.

To avoid problems with particles oscillating around the desired orbits it was preferable that they be injected into a cone only a twentieth of a degree wide and with a beam width of only 1 mm. The stability of oscillators had to exceed one part in 40 000. The error in accelerating frequency at the time of injection could be 0.25% at most, if loss of particles by collision with the walls of the vacuum system was to be kept below 10%. To avoid problems created by particles colliding with molecules of gas in the vacuum system, pressure in that system had to be kept below 10-6 mm of mercury, or one billionth of typical atmospheric pressure.

This precision stands in contrast to the gross parameters of the machine; a magnet weighing 800 tonnes and some ten metres in diameter, a motor-generator set rated at 1500 horsepower with a 36 tonne flywheel and able to produce the rapidly changing currents of 12 000 amperes at 1100 volts, cooling air flow rates of 16 000 cubic feet a minute. This unprecedented combination of massiveness and accuracy was needed if the equally unprecedented target energy of 1000 MeV was to be achieved. Indeed the Birmingham machine and the others being built or planned at that time were setting a pattern for accelerator development through to the present day and into the future. Today that same precision has to be maintained in equipment thousands of times more massive or extensive than Oliphant’s, in quest of accelerated particles with energies more than a thousand times greater than he and his team sought.

353

In the same month that Hibbard’s paper was published, an article appeared in the Birmingham Gazette559 implying that the project was running behind time. The source quoted was Fremlin, who was not personally involved (that is, his name does not occur in the committee Minutes), and he was more generally mentioned as being associated with the cyclotron). According to Fremlin, a lack of willingness by the Government to pay for staff was hampering progress. The remaining mechanical work would take at least another six months to complete, he suggested. Fremlin commented on the climate of competition in which he perceived the Birmingham enterprise was being advanced.

A synchrotron three times the size of the Birmingham plant is being constructed in New York. Its construction may catch up with progress of the British project because a greater number of people are employed on it. Another is being erected in California but it is not at so advanced a stage. As the Russian Professor Veksler was the first to publish details of the synchrotron, it is thought very likely the Russians are also building one.

In the last statement Fremlin was mistaken. The Russians were not in the proton synchrotron race at this time, though they built a synchro- cyclotron (“phasatron”). Nonetheless, the impression was given that the lead Birmingham had enjoyed was fast disappearing.

1951: “The Great Pole Pieces Disaster”

One of the biggest problems with the building of the machine, and at least a partial reason that its completion was so much delayed, occurred during the period the Minutes do not cover. There is no official record of

559 Birmingham Gazette, 10 October 1950. Clipping in Box 22X/6, UBSC. 354

this incident, though it was later recalled by a number of participants, such as Symonds, Hibbard and Moon. The following, for example, was provided by Symonds560, who placed the incident in late 1950 or early 1951, after Oliphant had left to return to Australia, and much of 1951 was taken up in responding to it.

It occurred in the early stages of testing of the magnet. “Mac” Cassidy, ex Royal Navy CPO, experienced in running the generators on battleships, was now our white boiler-suited chief engineer in the motor-generator room. He was able to manipulate the equipment to produce just one pulse of current at a time.

I was standing close to the machine at the time and had just given the nod to Cassidy to send through a slightly higher pulse of current. Suddenly a group of pole pieces561 broke away from the upper yoke of the magnet only a few metres from where I stood and slammed into the lower pole pieces with a tremendous noise. The collapse had affected 3 metres of the circumference of the machine and had been so violent as to shear off some of the bolts holding the supporting brackets. Fortunately the vacuum tube had not yet been put in place; otherwise it would have been badly damaged.

Hibbard supported this account of the incident and amplified the reason for what was soon dubbed “the great pole pieces disaster”.

560 This account is derived from interviews with Symonds in 1980 and 2010, and from a letter which Symonds wrote to Lawson in 1995 (see Moon Papers B27 UBSC). While there are some small variations in details between the various versions over the 30 year period, the essential features of the account are consistent. 561 The “pole pieces” were pieces of mild steel attached top and bottom to the laminations which made up the magnet yoke, that is, above and below the space to be filled in due course by the vacuum system. They had been carefully shaped to generate the needed radial gradient in the magnet field. 355

The first time the magnet was powered to a full field, a batch of these thin steel plates broke away and shot across the gap to hit the plates on the other side. The plates as a whole are held on firmly by the effect of magnetic fringe fields, so a strong mechanical support is not required, however the forces on individual plates are sensitive to local irregularities in the tiny gap between the pole plates and the main thick plates of the magnet yoke. Those small variations cause the magnetic flux to divert quite disproportionately away from a pole plate in which the gap is larger than it is for its neighbours, and the hold-on force for that plate is reduced. The result of this is to increase the resultant force trying to pull the plate across the accelerator gap. If a single plate or small batch moves as a result of this increased force a runaway situation develops, leading to immediate local failure. The answer to this was equally simple, though requiring considerable time to implement. Increase the tiny gap to a small gap and the problem goes away. Excusable lack of experience with magnets was a reason for this mistake in design562.

Hibbard clearly ascribed the problem to the design of the magnet, or to the accuracy with which the yoke plates had been manufactured, though he calls the error “excusable” given the lack of relevant experience. Symonds continued the story.

Hibbard had a solution almost at once, such was his grasp of the fundamentals. He proposed to lay an additional layer of non- magnetic material, such as Bakelite, to reduce the effect of the gaps and smooth out the magnetic field. This was simple to

562 Inall EK, Some of What Mark Oliphant Did and Others Who Worked With Him. EBA. A comprehensive account prepared by Inall following Oliphant’s death in 2000 and revised in 2001. In the document (p. 3) Inall ascribed the material dealing with the Birmingham proton synchrotron to Hibbard. This extract is from p. 7. 356

propose but it would be a major enterprise to implement, requiring the removal of the hundreds of pole pieces, many of which would need to be refurbished, and the removal and replacement (including welding) of the holding bolts and brackets.

Moon’s recollections563 put the event into a broader context.

The building of the synchrotron had been a long and hard struggle, because of two constructional errors of design564. The yoke was one inch steel plates but the pole-pieces were of 1/8 in. stampings ….calculated to give the desired field distribution across the gap. When current was passed through the windings, blocks of pole lamination broke away from their relatively light securing brackets and crashed against the opposing pole565.

As Symonds recalled the aftermath of the “disaster”, the attitude of Moon was now a decisive factor. Moon was, he recalled, a cautious man, and therefore reticent to push straight on with the repairs. He insisted that the opportunity be taken to study further the behaviour of the magnet at low current levels before taking off the pole pieces so that repairs could begin. Symonds acknowledged that many of the tests Moon suggested were sound and valuable, but when one lot was completed he always seemed to have some others he thought should be done. Temporary repairs were made to the damaged area to allow the tests to be run. After a number of months, Symonds and the other team members believed they knew what had gone wrong and what should be done. Pressure was put on Moon to authorise the full repairs.

563 Rolph (1995), p. 120. 564 The second error led to the “shorting-out” problem of the following year, to be discussed later. 565 Moon appears unreliable on details in this matter. By other accounts, the yoke plates were half-inch steel and the pole pieces quarter-inch. In another part of his account, he indicated that the “vacuum-vessel” was already in place at the time of the incident, but “not so deep” as to interfere with repairs. This is at variance with both the Minute book and other accounts. 357

By calling a meeting of the senior team to discuss the issue only 10 minutes before Moon had to be elsewhere, a decision was squeezed out of him. He agreed that there was nothing more to be gained by further testing.

We took him at his word. Once he left the meeting, I immediately gathered together whoever I could and we set to work. By three am we had removed and carefully stacked one third of the pole pieces, with pieces of paper between them, and it was too late for Moon to change his mind. I sent the men home for a wash and some breakfast and we all then returned to work as if nothing had happened.

When I saw Moon in the corridor the next morning, he tried to suggest another test for us to do. I said that it was now too late for that. On viewing, at my suggestion, the partly dismantled magnet, with eight circumferential feet of the yoke now bare of pole tips, he was initially shocked and returned to his office pale- faced. Later in the day he was able to laugh as heartily as anyone, and said “Those Aussies! I might have guessed! I have always known they are a menace to have around the place and now I have demonstrated it”. We could see he was in fact very glad the decision had been taken out of his hands566.

Moon had a last comment on the incident. He endorsed the Hibbard solution as a “theoretically simple remedy”, inserting a few millimetres of non-magnetic material between the yoke-plates and the pole laminations so that the distortions of the field would be less. However, that required the right materials, and time. “Through a friend in the “Bakelite’ industry, he obtained what was then very scarce material and

566 Symonds, op cit. 358

it was not too difficult to redesign the fixing brackets; but the task of removing and replacing thousands of laminations was formidable and took many months.”567. He did not comment on how many of the months were taken up with the testing which he reportedly required before repairs could begin.

The issues faced and overcome during 1951 are briefly summarised in the Report of the joint Standing Committee for Research for 1950-51. After noting that the cyclotron, in its first full year of operation, had “established itself as one of the country’s prime sources of radioactive materials”, the report went on “The building of the proton synchrotron continued; modification of the original construction proved necessary, imposing heavy burdens on the staff and resources of the laboratory, but progress became rapid toward the end of the year”568.

1952: Momentum resumes

It is not certain when the repairs required after the pole pieces incident were completed and the way opened toward the commissioning of the machine, but it was not until the end of 1951 that regular meetings of the synchrotron group recommenced, at least according to the surviving documentary record. On 10 December 1951, more than five years since the first recorded gathering of the team, Phillip Moon opened a fresh exercise book569 and wrote “As a preliminary to the new series of synchrotron group meetings, brief descriptions of the generator, magnet, vacuum system and frequency control system were given, and were followed by a general discussion”.

567 Moon and Ibbs. p. 29. 568 JSCR 1950-51, p. 9. 569 Minute Book of the Synchrotron Committee (Book 2). Box 22X/8 (31) UBSC. 359

It is hard to avoid the implication that some considerable time had elapsed since the team had last met, and that everyone needed reminding where others were up to. Over the next month or two, Len Hibbard reviewed the state of the frequency control system, John Symonds discussed methods of measuring the magnetic field, Riddiford reported some issues with the vacuum system; some units had cracked on firing. A week later, Colin Ramm reported progress on the ion source and injection system. This latter discussion must have been extensive; the Minutes run to nine pages.

After a few meetings Bracher resumed his Minute-taking role. On 18 February 1952, the meeting discussed “.the characteristics of the motor- generators, but as the Secretary understood practically nothing of what was said he did not take any notes”. Bracher had shown a tendency to light-heartedness before (as well as towards long-windedness) in his Minute taking, but this comment may reflect something more. Perhaps the team had not yet regained that sense of collective responsibility it had displayed in the early days.

Lightheartedness was not always appropriate. Impediments continued to appear. A meeting on 29 March 1952 (a Saturday) was convened by Moon, as the Minutes reported, “to discuss the magnet situation”. That situation was summarised by Symonds. During tests on 17 March, “an earth fault had developed late in the day”. It appeared that the coils of the magnet had been pulled radially outward by a few millimeters by pulses of current, permitting shorting-out by contact with a metal stud supporting a baffle. Moon asked about the possible impact of any long delay on the rest of the work. It is not recorded what answer he was given.

On this holdup, Hibbard commented

360

A second570, less excusable, error led to the failure of insulation of the magnet windings, which occurred after a substantial number of pulsed operations of the magnet had occurred. This was due to a method of holding the flexible coils to wooden battens resulting in a polygon coil. The hand-wound coils had a small amount of slack, which was no problem, but became important when it migrated around the coil and accumulated at one point, allowing the coil to impact a steel stud on the magnet yoke. A continuously-powered coil might have gone for many years before this occurred but repetitive pulsing produced the effect much more quickly.

We should recall that Oliphant identified this failure as one of the four or five issues that led to severe delay in the project. He argued that had more robust copper strip been used in place of the stranded copper wire that the engineer Taylour had insisted on, the problem would never have arisen.

On 22 April, Symonds outlined how the issue could be addressed. A strong loop of string (‘whipcord’ was Taylour’s preferred terminology) could hold each coil firmly in place and so prevent the problem occurring. On 5 May he estimated that it would take about a fortnight for the current team of four men to tie up the coils, but that things needed to move faster if the installation of the vacuum system was not to be delayed. “The present team is working hard and some were getting sore fingers as a result of straining up the whipcord. Professor Moon was in favour of getting help from the research staff, but both Symonds and Taylour thought that the period of training would be too long to make this practicable.”

570 The first being the issue with the pole pieces. 361

Moon’s later recollections of the event added some information but also some variation.

It was not until the vacuum-ring had been successfully installed and tested and protons injected into the ring that the second disaster struck us. The flexible winding was primarily supported against inward movement by being wound around stout posts, thus being a polygon and not strictly a circle. After a few weeks of magnetic pulsing, short-circuits developed between the winding and the yoke. It had been recognized that additional restraint was needed, and drawings showed springs intermediate between the posts. Unknown to those of us who remained on the staff, these springs had been found difficult to incorporate and had been omitted. The solution, once again, was simple but laborious: stiffen the sides of the polygon by lashing insulating bars to the winding. Again there was just enough room, but it took weeks to thread many turns of cord through the gaps which were not only narrow but invisible because of the pole pieces. Of course the vacuum ring had to be removed and replaced571.

In another account572 he recalled

Again Bakelite was the solution -- in the form of rods which had to be strapped to the windings in between the vertical rods around which they were wound. This was done without taking out the many laminations, but with the aid of stiff wires and a little bit of mirror work, a little bit of feeling, and so on and I suppose some fisherman’s techniques, it was possible to tie the cord around the

571 Moon and Ibbs (1981), p. 31. 572 Rolph (1995), p. 120. 362

windings and the Bakelite rod at a couple of places along the length of each rod and that was alright.

Moon’s recollections, though vivid, appear to be inaccurate in some details. As we shall see, the installation and testing of the vacuum ring and the first circulation of the beam, at least according to the Minutes, was yet to come. The immediate issue was dealt with quite quickly, but according to Hibbard there were wider implications.

The above faults and delays produced some deterioration of relations in the department. Taylour had a nervous breakdown and was confined to his home for several months, possibly because the magnet faults were more his responsibility than that of anyone else. He had heated arguments with Moon and is said to have sworn at him. For some weeks Symonds acted as a go- between liaising between Moon and Taylour, until Taylour suspected that his advice was not being properly attributed to him. He appealed to me as a neutral observer to intervene on his behalf with Moon… I recognised that Taylour had some cause for grievance and made representations on his behalf. It was a peculiar situation because I myself had had fairly strained relationships with Taylour for years573.

An indication of the timing of Taylour’s absence comes from the Minutes. These show that, having missed only one meeting out of ten from January 1952, he did not attend three in a row in September through December. He may have missed more, but from the December meeting onward, Moon did not record attendance.

573 Hibbard in Inall op cit. p. 7. 363

At the meeting on 21 May 1952, the key issue was the vacuum system. Riddiford, who was in charge of this element of the project, took the Minutes. Unusually high, and indeed unacceptable, losses were occurring in the manufacture of the units of ceramic pipe which were to make up the ring through which the protons would circulate. These had to be vacuum-tested before the ends were ground and the units metalised. Riddiford hoped that these issues would not hold up the assembly of the complete system.

Raising the pace

On 14 June 1952, some six months after the meetings of the synchrotron team had been revived, Moon called a special meeting “to discuss the possibility of a special laboratory effort during the second half of 1952 towards the completion of the synchrotron. “ The Minutes, taken by Bracher, record

Professor Moon said that the scientific and technical staff of the synchrotron constituted about one third of the laboratory. The dependence on outside firms might end soon and the completion of the machine would depend entirely on our own efforts. He proposed to invite members of the laboratory to join in a special effort for a limited time on the understanding that

1. The cyclotron would be kept in operation. 2. The essential needs of the teaching classes would be fully met. 3. Final year research students would not be affected

Subject to the above, it was proposed to give the synchrotron overriding priority for technical assistance. The present meeting

364

was preliminary to a decision which would be reached in about two weeks.

The nature of the forthcoming decision is not clear. It may be that Moon was contemplating canceling the whole enterprise if the way could not be cleared to a successful conclusion in a reasonable time. The reason for convening the meeting was clear enough. Progress over recent months had been impeded by the troubles recalled above, but there had been some, and with the expected completion time now having been almost doubled, a major push to completion seemed called for.

The Minutes do not contain a list of attendees at this meeting, but it is implied that at least most of the Department was present. Certainly the relatively new WE (Bill) Burcham was there, as was John Fremlin, whose particular concern was the cyclotron, as both their names are mentioned. They had not attended before, as far as can be ascertained.

At the meeting, Hibbard, Aughtie, Symonds, Ramm and Riddiford outlined ways in which their sections of the work could be helped. Assistance was needed at the high power end of the RF system; Cassidy could do with help with circuits involving the generator house and control room. At the technical level, much wiring remained to be done and an extra electrician was needed. The workshop under Robertson would see all essential tasks through.

Riddiford and Burcham advocated “group attacks” as had been done with the tying of the magnet coils. Symonds was concerned that “a drive for a limited time might leave some jobs incomplete”. In response, Moon said that any temporary workers would take on only those tasks which would clearly be finished by the end of 1952. Fremlin asked about the attitude of DSIR to the taking on of extra staff. The Minutes report that

365

Moon gave some confidential information on this point. He felt that in any case we “should see a strong team in the next session; several more research workers, two of them specifically interested in the synchrotron, were expected to be in the laboratory.

More than three months later, the team met again (29 September 1952). Moon took the Minutes, as he was to do from then on. Clearly the vacuum system was now in place, as Riddiford reported that the pressure in the ring could be reduced to 2.6 x 10-6 mm after two days of pumping. Ramm reported that the HT set was complete, though work was needed on the control desk; according to Hibbard, the injector was almost ready to go. From October, the meetings were to be weekly and short. The meeting on 3 November was told that the generator had been run up to full power for several hours on 31 October, pulsing every 9 seconds, with peak current steady within 50 A. The HT set which was to serve as the injector had been brought to 500 kV.

By 1 December the magnet was reportedly ready to be run; by 15 December it was being pulsed most days. Vacuum could reach the needed 10-6 mm after 24 hours of pumping; injector current was up to 150 μA. The bad news was scum reported on the mercury in the variable inductance; this needed frequent clearing. The concern was that the mercury might begin to “wet” the copper and adhere even when the coil was lifted clear. Attendance at the meeting, the last time it was recorded, was Burcham, Hibbard, Symonds, Riddiford, Aughtie, Coe and March, plus Moon of course. The meeting in October had been attended by 17 people, indicating how the team had grown.

With more progress to report, the relevant section of the Joint Standing Committee on Research was much larger in 1952574 than in 1951,

574 JSCR 1951-52, p. 9. 366

though it covers developments only up until around September. In particular, it gives insight into the challenge of the vacuum system.

The synchrotron has taken up an increasing share of the Laboratory’s effort and very substantial progress can be reported. The vacuum system, a hollow ring made of sixty sections of electrical porcelain, proved virtually free of leaks at first assembly and by September 1952 had been pumped out to less than 10-5 mm of mercury. Each section had to be metallised internally, and some sections required an external electrically-conducting coating as well. This was done by our own staff with the aid of a large oven for the use of which we owe thanks to the Department of Industrial Metallurgy; each batch of from three to six sections had to be in the oven for about 60 hours, with continuous watch on the temperature.

Another view of the history of the vacuum system comes from Moon.

The chemical stoneware vacuum sections obtained by Oliphant, when installed by Len Riddiford, proved to be porous. Len took the plunge -- remember he was only about 22 at the time -- and decided to scrap Oliphant’s chemical stoneware and go for electrical porcelain. He did this, he organised it, he had the slip transferred from one place for another for firing, he supervised firing, he had sections brought back in Birmingham, he painted them with the appropriate oil to metallise them, finishing the metallising process, he baked the sections in an oven at the Industrial Metallurgy Department, and came out of the end with a truly splendid vacuum system that did all that was asked of it right from start to finish575.

575 Rolph (1995), p. 119. 367

The report from the Joint Standing Committee for the year concluded

Modifications in the 800 ton magnet have been successfully made, but we await completion of adjustments to the generators that energise it -- a substantial task in which the efforts of members of our own staff and of the manufacturers have been assisted by consultation with past and present members of the staff of the Department of Electrical Engineering. The radio frequency and control systems of the cyclotron are both virtually complete and we are approaching the stage at which these various sections will be joined into a complex system which has to be made to work as a single organism.

The choice of imagery in the final sentence is striking.

1953: triumph at last

By 5 January, measurements showed that the peak field in the magnet could reach the targeted 15 000 gauss with a current of 6000 amperes. The injector was now able to deliver 350 μA through a ¼ inch aperture. Riddiford reported on the current state of the rival Cosmotron project, which by now was delivering a beam; according to the Minutes “his written notes would be available later”. These have not survived. The reliability of the vacuum system had a boost on 26 February, with the report to the meeting that two spares of each kind of component were now available. Things were picking up. The Minutes report that by early April the injector was operational and protons were circulating.

By now, Moon’s Minutes were falling well short of the standards set by Bracher. They are noticeably lacking in information as the project

368

reached its long-awaited climax. On 11 July, Moon reported briefly that the machine had achieved its first operation at 250 MeV “at the start of the month” but that this had not been previously reported as there had been no meeting on 4 July.

To mark the achievement of 250 MeV, Moon drafted some notes as if he planned a press release. Certainly an attached page lists contacts in various newspapers and other media outlets. Dated 7 July 1953576, these notes survive. Moon’s notes list some earlier milestones, not recorded in the Minutes; 1 MeV (only twice the injection energy and therefore little more than a circulating beam) reached in early May, 18 MeV in June, now 250 MeV in early July. Ever cautious, he thought the future was still “unpredictable”; the machine was still undergoing “tests”.

Then something more upbeat was ventured. While the machine would not be successful until it reached something near 1000 MeV, it was already “one of Europe’s leading nuclear machines through the energy of its beam”. At the same time, it was to be complementary to, rather than a competitor for, Britain’s other accelerators. As for its purpose, the machine would be used for research into nuclear physics and “especially for reproducing in the laboratory some of the processes which go on in cosmic radiation”. The words were appropriate; in his proposal for the machine some eight years earlier, Oliphant had spoken of producing “artificial cosmic rays”. Moon was keen to see the enterprise as collaborative, without a focus on personalities; no names, no individual photographs. He also insisted that the funding agencies that had made the venture possible, the Nuffield Foundation and the DSIR, be properly acknowledged.

576 Relevant papers are held in Box 22X/6, UBSC. 369

There is no record of such a press release being issued. In any case it would have been overtaken by events. On 16 July, pulses of particles were accelerated to over 950 MeV, very much the target that Oliphant had set in his first musings on the concept (being the rest mass of a proton) nearly a decade before. It seems that the committee did not meet to discuss that success; certainly there is nothing about it in the Minutes. News did however get out and into the press. Reports spoke of the machine having been under construction for seven years and under trial for two months.

The report of the JSC for 1952-53, which celebrated the first successful operation of the synchrotron, looked both back and forward.

The proton synchrotron was brought into experimental operation at full energy during July 1953, providing nuclear particles with energies six times greater than were previously available in a British laboratory, though far below those of the more energetic cosmic rays. It is pleasant to record that this machine was conceived by Professor ML Oliphant, designed in the Nuffield Laboratory by a largely Australian team, and constructed by the staff of the Department with financial support from the Department of Scientific and Industrial Research. Several major components were supplied by industrial firms, but the specifications were in most cases so unusual that members of the laboratory were intimately involved in the technical problems that arose in making them. This detailed knowledge of the machine should be most valuable in its future use.

Research is already commencing with the fast protons and with the mesons they generate in passing through matter, but its scope will be limited at first by the need to secure the best and

370

most reliable performance of the machine; in any event, only brief periods of operation at full energy can be allowed before massive protective shielding has been installed577.

The last meeting recorded in the Minute book was on 8 August 1953578, almost seven years after the first such gathering. With the machine now operational for almost a month, Moon proposed that a new pattern of meetings “was appropriate”. Henceforth, colloquia on high energy physics would be organised by RH Dahlitz and WO Lock, Symonds would convene technical meetings on the machine itself, while meetings under Ramm would coordinate the experimental program. To keep others up to date, Ramm and Symonds would report developments in their fields to the colloquia. It was time to draw on the fruits of years of intense effort.

Seven years instead of three

Though it was the first proton synchrotron to be proposed, funded, designed and commenced, the Birmingham machine was not the first to generate a beam in the desired high energy range. It was beaten to that goal by about a year by the Brookhaven Cosmotron. Whatever priority Oliphant may be entitled to in the conception of the synchrotron principle, he failed to be the first to get his machine on-line, despite being willing at times to take some short cuts to enhance that possibility. This second placing was largely the consequence of the blowout in time for completion; what was initially judged likely to take three years took almost seven. Even a year saved would have seen the Birmingham machine first across the line and into history.

577 JSCR 1952-53, p. 8. 578 The small note book has some all but unreadable jottings in Moon’s hand from a meeting on 21 December 1953, but those were not written up. 371

In assessing the reasons for such a delay, we can sum up some broad possibilities.

 That from the start, the timelines set for the project were unrealistic, especially given the pioneering nature of the technology and the many challenges that had to be faced.  That the economically-straitened circumstances that existed in the UK immediately after the war, with industry just recovering from the damage and shortages of the war, were unfavourable for a project of this scale and complexity, making delays inevitable.  That the project was significantly under-funded, and that as a result, Oliphant and his team were short of human and material resources, resulting in delays.  That the designs for the synchrotron contained some basic flaws, and/or that poor decisions were made regarding materials or processes, resulting in problems that caused significant delays.  That the team relied too much on graduate students who came and went and as a result vital continuity of personnel was lacking.  That the project did not have sustained and consistent leadership.

The narrative of the project, as we have set it out here, enables us to make some judgement as to the importance of these various factors.

Many would agree that Oliphant was by nature a man of positive outlook, of "crash through or crash” style. This may well have made him over-optimistic when judging how long a project would take. His first such endeavour, the basement accelerator of the early 1930s, was up and running in a few months, but everything that followed, the HT set at the Cavendish, the Nuffield cyclotron, the proton

372

synchrotron, the cyclo-synchrotron in Canberra, all took much longer to finish than planned (the last was abandoned well before completion). It is not fair to say that Oliphant thought that everything would go smoothly, that no problems would arise, but he had a profound faith in ingenuity and in “the fire in the belly” to throw solutions (often more than one) to any challenge. Nonetheless he was entering a new realm of science and technology, uncertain when he set out quite what he would find or how he would get there. Undeniably, he was impatient. A little more time in planning and design may well have got him there sooner.

It cannot be denied that immediate postwar Britain was far from the ideal environment for an enterprise as Oliphant was undertaking. Industry was only just recovering from the stresses of war, repairing the damage from bombing, shifting its focus from military to civilian activities. Restoring the nation’s fractured infrastructure, for example the electricity supply, was a priority. We have seen references to the shortage of materials, the need for the DSIR bureaucracy to give Oliphant priority if he was to secure his steel, Moon having to call on a friendly contact to get enough scarce Bakelite for repairs after the pole pieces affair. That Parsons took 18 months to deliver the generator sets was at least in part the consequence of electrical machinery being in high demand elsewhere. It was significant that food rationing, symbolic of an economy under strain, was still in place when Oliphant and Peierls held their international nuclear physics conference in Birmingham in the autumn of 1948.

Equally there can be little doubt that a large-scale project of this kind would have moved ahead faster in a less stressed climate, without the endemic shortages, as indeed similar projects did in the US. Some commentators have thought it more remarkable that the

373

Birmingham enterprise succeeded at all, rather than that it took longer than planned. We shall return to that issue in the next chapter.

There is little doubt that the Birmingham enterprise was significantly less well-funded than the US enterprises with which it was contemporaneous, and to some extent, in competition, the Cosmotron and the Bevatron. Analysis of expenditures suggests that on a dollars per MeV basis, the UK enterprise gave two or three times more value for money than its rivals, though the comparison is complicated by considerations of beam intensity and repetition rate.

Does that mean the enterprise was underfunded? If so, did that contribute to the machine taking longer to complete? We should note that Oliphant gained all the funding he sought, and that even when the design was changed in 1946, he was able to fit the new layout under the old budget cap. We find in the Minutes no reference to money being short or being a constraint on progress. It is likely therefore than the greater cost-effectiveness of the Birmingham machine was more a consequence of the different style of working, with more of the work being done in-house by graduate students on their meagre salaries, rather then by higher-paid engineers or farmed out to industry. In Birmingham, the major outlays were for purchase from industry of bulk materials, such as the steel and copper for the magnet, or of large scale but relatively well-proven technology such as the generator sets and switchgear. The real value-adding, represented by the many complex, indeed pioneering, systems the enterprise needed, was done within the laboratory walls.

Design and construction flaws in the synchrotron have been acknowledged by several of the participants, including Oliphant,

374

Moon and Hibbard. In terms of impeding progress, those of most importance concerned the magnet. The problems with the pole pieces and later with the magnet windings, which together set the project back perhaps a year579, have both been attributed to design flaws and those flaws to lack of experience, particularly that of Taylour, who carried major responsibility for this element of the project. Hibbard and Symonds have conceded that the first problem could perhaps have been foreseen with a little thought, and the second appeared to be largely due to the omission of some components during construction, and/or, as Oliphant argued, to the choice of materials for the windings, a choice he ascribed to Taylour.

The initial construction of the magnet also seems to have run late, leading Oliphant at one point to question Taylour when it would be finished, a question Taylour declined to answer. Dates reportedly set for commencement and completion were regularly not met, with the windings not finally in place until well into 1949. The reasons for such delays are not clear. Lack of staff may have been a factor, though apparently teams of six or eight could be assembled to carry out the windings, and at other times there was discussion as to whether the magnet builders had enough to do.

The vacuum system was the last element to be completed, though it was ready in time. Delays there were due in part to changes to the materials chosen, firstly metal sheeting (as of early 1947), then chemical stoneware, and when that proved too porous, electrical porcelain.

579 For example Symonds later commented, with regard to the pole pieces affair, “we lost the best part of a year as a result. Without that delay, our machine may well have beaten the Cosmotron and been the first synchrotron to produce a beam”. Symonds op cit. (footnote 536??)

375

There is no doubt that the project relied heavily on the efforts of younger people, the larger proportion of them Australian. These were in the main post-graduates working towards PhDs, some of them Nuffield Fellows or junior lecturers. In the main they stayed long enough to complete their dissertations and then moved on. This led to a continual change in the composition of the synchrotron team, with only a few, such as Symonds, seeing the task through from start to finish. It also meant that responsibility for various elements of the machine changed over time, those who undertook the initial work not always being those who completed the job.

Did such an arrangement contribute to the project running late? We could frame arguments both ways. On the one hand, less-than- efficient handovers might have slowed progress, and there is also the issue that graduate students usually needed some experimental results around which to frame their thesis. That may have taken time away from working on their element of the machine. On the other hand, many of the students did end up with PhDs based on their work as machine builders, rather than as experimentalists. They were building careers, as well as an accelerator. That was a new development, which we examine further in the next chapter.

At the same time, there is little doubt that the students were generally enthusiastic (sufficiently so, if some recollections are to be believed, to give their labour to jack-hammering holes in the concrete raft so that the magnet foundations could be created), and they responded well to Oliphant’s leadership style. They may therefore have been more productive than professional engineers on a salary.

376

That leadership of the project was passed on midstream from Oliphant to Moon is a matter of record, as is the co-incidence of that transfer with the loss through death of Gooden, the second vital figure in the enterprise. Oliphant had known for about three years before his departure that he would be leaving, though when others knew is not so clear, and planning for his new laboratory in Canberra inevitably took some space among the many other issues in his life, including the synchrotron. Given Oliphant’s much-commented-on energy and inventiveness, we cannot say that he devoted significantly less time to the synchrotron as a result. He attended all but one of the committee meetings before his departure and always had plenty to say, as the Minutes attested.

We have noted the comments about the “double blow” to the project through the almost simultaneous loss of Oliphant and Gooden, and looked at evidence that Moon, who admitted being ill-prepared to lead the project, took a number of months at least to firmly take the reins and to clear his desk of other duties so he could give the task his full attention. The failure of the synchrotron committee to meet for nearly two and half years before and after the loss of Oliphant and Gooden surely attests to some loss of momentum. We can add to that the assertion, by Symonds at least, that Moon’s “caution” added to the time needed to recover from the pole pieces fiasco. To put a time-penalty on these changes is not easy, but they must have set the project back months at least. Finally we are left to ponder how differently the project might have progressed had Oliphant’s charismatic leadership and Gooden’s efficient management continued to be available.

What can we say then in conclusion to this analysis of factors that impeded the completion of the Birmingham synchrotron? If we try to

377

place the factors in some order of importance, we should begin with the errors in design and construction; the issues that they generated, together with the general delays in completing the magnet, added as much as a year to the completion time. The leadership changes may well have added almost as much. There can be little doubt that the times were not optimal, and the initial expectations were unrealistic given the newness of the technology. Even with its advantages the Cosmotron took four years to complete, the Bevatron longer.

So much for the construction of the synchrotron. How useful a machine resulted? Did it live up to expectations? What impact did it have on the understanding of nuclear physics, on the technology of particle acceleration, on the training of physicists? How did it compare with its American rivals? That assessment is the subject of the next chapter.

378

CHAPTER EIGHT The Birmingham Proton Synchrotron 1953 to 1967: Operation and Impact

The building of the Birmingham proton synchrotron had taken almost seven years from the first meetings of the team to the first runs at full design energy. The operational lifetime of the machine would double that, with the synchrotron not closed down until 1967. That total of 14 years use was similar to the lifetimes of the other early first-generation machines. During that time substantial upgrades were undertaken, with primitive systems replaced with more modern equipment.

How successful or valuable was the bold Birmingham enterprise? Was it a worthwhile expenditure of time, effort and resources? That question has a number of dimensions, some of them drawing on the reasons originally advanced for building the machine. We can lay out three key questions relating to outcomes; technological, scientific and educational (or more broadly “national capacity-building”). In each of these enquiries, we must go as far as the evidence does, though it may be that no firm conclusion can be reached.

In assessing the impact of the Birmingham enterprise, we will pursue two further lines of inquiry. The first will explore the broader context within which the enterprise sat. At the same time as the Birmingham proton synchrotron was being built, two other similar machines were under construction, the 6 BeV Bevatron at the University of California at Berkeley, and the 3 BeV Cosmotron at Brookhaven on Long Island, NY. We need to explore the interaction, and possible mutual influences, between the Birmingham venture and the other two, including before they were commenced.

379

The second additional avenue takes us to the views on the venture, and on Oliphant, by people in a position to comment from personal experience. This inquiry will take us far beyond the brief assessments collated in our earlier attributional survey, which drew on overviews on the field, including some by people with little personal involvement.

Technological outcomes: selling the machine.

Building the Birmingham accelerator was firstly a technological enterprise. Plans had been drawn, funds secured and a team assembled to build a machine (albeit a machine of almost unparalleled size and complexity) to perform a prescribed task, the acceleration of protons to a hoped-for energy of 1000 MeV. Until that machine was built and operating, none of the other objectives, exploring new physical phenomena or training physicists, could be pursued. We should therefore begin by assessing how well the machine worked by its own terms, how well it lived up to the vision of Oliphant and others.

By mid 1955, two years after the first circulating beam and ten years after Oliphant’s proposals to Akers to secure funds, enough had been learned about the behaviour of the machine for Moon, Riddiford and Symonds to describe that behaviour in a paper intended to guide experimentalists interested in using it580 581 The beam energy, which

580 Moon et al (1955). 581 Over the previous few years, a series of papers had been published describing in detail the construction and operation of the various elements of the complex apparatus; these included Riddiford (1951) on the vacuum system, Hibbard and Caro (1953) on the production of stable sources of high voltage, Hibbard, Randorf and Riddiford (1953) on the high power RF amplifier, Hibbard (1954) on the radio-frequency system, Hibbard, Caro and Freeman (1954) on the servo mechanism to accurately control the variation of the frequency of the accelerating field, Fuller and Hibbard (1954) on accurate measurement of the rising magnetic field, and Ramm (1955) on the injection system. The many appearances of Hibbard’s name indicate how central he had become to the enterprise. Over the same Several team members produced papers which placed the enterprise in a wider context, such as Riddiford (1953) on the importance of 380

had always been a key driver for development, depended on both the mean radius of the particle orbit (450 cm) and the peak field (12 500 gauss). Hopes of creating fields approaching 15 000 gauss, as originally planned, and therefore energies as high as 1300 MeV, had by now been thwarted by the discovery that the magnet “saturated” at around 12 500 gauss, so that increasing the current in the coils did not further strengthen the field. Those key figures indicated a beam energy of around 990 MeV, just below the 1 GeV threshold (though above the rest mass of the proton which had guided Oliphant’s early vision). The beam at full energy had been measured to be only about 2 cm across in both dimensions. The charge carried by a pulse of particles represented, according to electrostatic measurements, some three billion (3 x 109) protons. Pulses could be fired every 10 seconds, the acceleration lasting about a second.

By appropriate finessing (mostly of the precise time in the magnet cycle at which the accelerating field was turned off) the orbit could be driven slightly inwards or outwards (though still within the vacuum chamber). Smaller orbits meant lower energy, larger orbits higher energy, and it was by such variations, Moon et al reported, that the 1000 MeV border could be crossed, if not by much. More commonly, this technique was used to reduce the beam energy. Maximum energy was not always the goal. Moving the beam inwards or outwards also allowed it to be brought into contact with targets attached to the inner or outer wall of the vacuum chamber, making experimental investigations possible. The paper devoted much space to the fine details of such matters, since they would be crucial in designing experiments.

high vacua in high energy physics and Symonds (1955) on the measurement of strong magnetic fields.t 381

While the ultimate goal was to produce an external beam to be fed into targets or measuring equipment outside the synchrotron, techniques were already available to secure a “scattered beam” by driving the beam outwards so that it struck the edge of a guiding plate. This scattered about one proton in every ten thousand outside the machine, a relatively meagre return, but a substantial external beam had not been an early priority for the synchrotron. Much significant experimental work could be undertaken with an internal beam.

The machine could boast some other merits. A beam substantially homogenous in energy was one of those. The energy spread of the beam was less than 1 MeV at 1000 MeV, quite adequate for any experiments being planned and providing “very high energy resolution”. Knowing with precision the energy of the particles being used (as well as knowing their nature and when they would arrive) was a key requirement for much experimentation and had been one of the significant ways that, in Oliphant’s original vision, artificial cosmic rays would have advantages for research over the naturally-occurring form.

Significant defects remained. A number of contributors to the 1993 Symposium commented on the impacts of one of the machine’s major design faults, the presence of a large “fringing” magnetic field extending well beyond the limits of the synchrotron machine and frustratingly evident whenever it was in operation. This field resulted directly from the outward-pointing C-shaped magnet design, though it could have been minimised by wrapping the magnet coils on both sides of the vacuum chamber, as was done with the Cosmotron. This was presumably not recognised in early design or not thought an important issue. Here then was another consequence of poor magnet design582.

582 It could be argued (and maybe was) that the magnet design used made the vacuum system readily available for maintenance or even replacement. Perhaps we should note that the vacuum system was perhaps the only element of the machine that never gave trouble. 382

Van der Ray commented that this field not only caused any magnetic object in the vicinity to move583, but “also interacted, with disastrous consequences, with the electronics such as photo multipliers and thermionic valves, as the advent of the transistorised electronics was yet to come.” This required, for example, all such electronics, including particle detectors, to be magnetically shielded. Shaylor584 recalled that the “magnetic field had a very large external component which made itself felt over virtually the whole of the campus”. In a nearby workshop “small tools would stand up on the bench and sort of wave at you, and the metal dustbin in the corner of the room would lift its head as if it was greeting you”585. More frustrating was the effect on any wrist or pocket watch unwisely worn in the vicinity; the affected timepiece would require a visit to the Department’s demagnetiser.

Impacts extended to the TV set in the control room of the cyclotron, then on the same level as the synchrotron. The regular shrinking, expanding and reversing of the TV picture as the magnet went through its ten second cycle made watching the cricket, a popular pastime with the many Australians in the group, an exercise in patience. As was later discovered, imbedded magnetism, strong enough to affect the operation of an oscilloscope, lingered in the area which had housed the synchrotron long after the machine itself had been dismantled and removed.586

583 This could have serious consequences. Whitaker recalled an occasion when Hibbard, against the rules, was attempting to move a steel trolley laden with lead blocks when the machine was pulsing. He misjudged the arrival of the next pulse and lost control of the trolley as it moved with the pulse. The trolley toppled and the load of lead nearly fell on an assistant who just managed to jump out of the way. Rolph (1995), p. 43. 584 Rolph (1995), p. 107. 585 Whitaker recalled that a pair of razor blades on his desk stood up when the magnet pulsed, so he knew the machine was working satisfactorily. Rolph (1995) p. 43. 586 The source of the lingering magnetism proved to be the steel piers driven down to bed rock to support the mass of the magnet. These had simply been cut off at floor level when the synchrotron was dismantled. 383

Continuous improvement.

Oliphant’s family motto was Altiora peto (I seek higher things). This had guided his life, always in search of better ways to achieve his aims. The conception and building of the Birmingham synchrotron can be seen as an embodiment of that motto, the quest for higher energies to advance the cause of knowledge of nuclear reactions. That quest for improvement outlasted his personal involvement with the machine. Over the 14 years of its operational life, little about the synchrotron machine persisted unchanged following its first firing up in July 1953. Upgrades and enhancements were undertaken with just about every supporting system, providing capabilities not initially present. It is not necessary to draw on Oliphant’s motto to justify that; experimental apparatus in science is constantly under development, though commonly in that development one form of apparatus is rendered obsolete by later advances. In Oliphant’s own career, he had moved successively from the “basement accelerator” to the HT sets, thence to a cyclotron and finally to the synchrotron. He would attempt even grander achievements in years to come587.

The synchrotron, whatever debt its construction owed to the “sealing wax and string” tradition, was a massive investment of funds and effort in a substantial piece of hardware and could not be lightly discarded for something better. The massive magnet, once built, was in place to stay, and its inbuilt faults had to be endured or worked around. The machine’s performance could however be fine-tuned, and even substantially upgraded, by changes in the auxiliary systems. That indeed is what

587 As part of his plans for the ANU Research School of Physical Sciences, of which he took command after leaving Birmingham in 1950, Oliphant planned to build a 10 GeV accelerator of (once again) radically new design. The machine ran into such difficulty that is was dubbed by critics “The White Oliphant” and was never completed. 384

happened, and as a result its working life extended for a decade and a half, a long time when compared with earlier generations of nuclear physics apparatus588.

We can construct a chronology for that path of improvement by drawing on the Annual Reports of the Department of Physics, as included in the annually-published reports of the university’s Joint Standing Committee on Research. Those accounts can be supported and embellished by reference to the recollections of participants in the 40th anniversary symposium, held in 1993.

Even thought the first circulating beam had been achieved in July 1953, the synchrotron could not be regarded as complete. The process of design and construction of the Birmingham synchrotron, with its emphasis on speed, ingenuity and a minimum of planning, had left a legacy of systems whose performance, though initially adequate, soon began to show their limitations. As a result (and as might be expected with any technology so new and complex), ways were soon being sought to upgrade the machine and enhance its performance. These improvements resulted from an intermeshing of factors, including the arrival of forms of technology not previously available, such as ferrites and transistors, the availability of time to devise better solutions to challenges that had been solved in an interim manner in the drive to get a beam, and experience with operating the machine which both illuminated deficiencies and raised possibilities.

In the middle of the 1950s, following several years of operation, the machine’s handlers could report that it was working reliably and well, especially following the replacement of the dipping mercury inductance

588 For example, the working life of the basement accelerator at the Cavendish was only three years. 385

with one based on ferrites. Proton beams were available ranging from 200 to 1000 MeV. Additional experimental facilities were in place to increase access to the scattered beam by experimental teams, and as a consequence, more shielding had been installed “to give satisfactory protection to all populated areas”589. The concrete used had been thickened with iron pyrites to double its density. A development that “held much promise” was the availability of beams of protons which had been polarised (that is, sorted by spin) through shallow angle collisions with a carbon target. The issue in question was the effect of polarisation on interactions with other nucleons.

To analyse the outcomes of collisions of the beam with targets, that is, the variety and behaviour of the resulting particles, a range of techniques were on hand, including the traditional particle counters, photographic films (“nuclear emulsions”) and diffusion cloud chambers filled with hydrogen, deuterium, helium and nitrogen590. In the latter two of these devices, the particles left visible tracks as the result of ionisation of the medium. The results of such data-gathering added to knowledge; for example, studies using cloud chambers filled with deuterium and helium showed that under high-energy proton bombardment, the protons and neutrons in deuterium behaved essentially as independent particles but they were much less independent in helium.

The first steps were being taken to “semi-automate” the analysis of the records, notably using electronic techniques to measure the direction of the tracks of the particles in three dimensions, which helped to determine their energy. These methods, which involved devices such as

589 JSCR 1956/57, p. 10. 590 Unlike the early cloud chambers pioneered by Blackett at the Cavendish, following the work of CTR Wilson, diffusion cloud chambers were always ready to respond to any influx of charged particles and did not need to be “triggered”. 386

photocells, promised to reduce the amount of work involved in such analysis and to increase its accuracy. This was the beginning of a major development in technique which would ultimately see Birmingham as a leading center of activity in this field.

The machine was overhauled in the summer of 1957591, after four years of operation, part of a program of continuous improvement to produce more intense beams and steadier operation. A range of issues was addressed and corrected; changes in the layout of the laboratory allowed beams of scattered particles to be directed into detection and measuring devices some distance from the machine. In consequence, results were being produced “more rapidly than ever before”.

But not all was well. A ghost from the past reappeared in April 1958592. For a second time the coils powering the magnet developed a short- circuit to the magnet yoke, the result of nearly five years of flexing under pulsing. Repairs took six months; the windings had to be stretched, stiffened and bound together in bundles, and separated from the yoke by insulation. To facilitate this work, the vacuum system, which had never given any trouble and had never needed repair or upgrading, was removed and its gaskets renewed.

Waiting for the machine to return to service was a new generation of detecting apparatus, a five litre “bubble chamber”593 containing liquid hydrogen (a two litre bubble chamber filled with propane had come into use just before the breakdown). By using liquids rather than gases, these devices presented many more targets to the beam than did cloud

591 JSCR 1956/57, p. 10. 592 JSCR 1957/58 p. 9. 593 The bubble chamber had been developed by the American physicist Donald Glaser in the early 1950s. For some background see Glaser (1952). 387

chambers, generating more collisions (and therefore many more tracks to analyse).

Also waiting the reactivation of the machine was a new system installed in the winter to extract an external beam for the first time. A current pulse of 40 kA sent through a single turn copper coil inserted around the circulating beam redirected up to 10% of the particles to a coherent external beam that could be directed across the laboratory. The beam was 300 times more intense than had been previously available594, and more remodelling of the laboratory layout and the placing of more concrete shielding was needed to protect those working nearby. By the end of the decade, the availability of an intense external beam, married to the bubble chamber which could produce thousands of images of particles for analysis, generated a new program of work and significant results. For example, the scattering of protons by protons, and protons by neutrons, using both hydrogen and deuterium in the bubble chamber, generated results which were consistent with the “charge independence”595 of the .

The “last dinosaur”

Upgrades continued, mostly at the front end of the machine; a new ion source, a new accelerator tube, a new system for the guiding beams into the accelerator. In a major development, a new capacity to vary the relationship between the RF accelerating field and the magnetic field, raised expectations of more reliable and flexible operation, and brought some important new options. The new system was trialled in 1960/1 and permanently installed in 1961/2596 .

594 That is, obtained by scattering the internal beam. 595 This concept asserts that the force of attraction operating between nucleons (later known as the “strong nuclear force”) was independent of the charges on the particles. 596 JSCR 1961/62, p. 10. 388

Whitaker described the existing system as the “last electromechanical dinosaur to be eliminated”597. It was dispatched to history by the intervention of an electrical engineering graduate, recently recruited to the team. The complex “sequence timer”, which had controlled the operation of the generators powering the magnetic field, was to be replaced by “a box of transistors”, much to the chagrin of Cassidy who had overseen the operation of the electromechanical timer for a decade. He was consoled, Whitaker recalled, by being allowed to keep the old system on standby in case the new electronics failed.

The new controller allowed the repetition period to be reduced over time from ten seconds to four seconds, effectively increasing the beam current two and a half times. That development pleased the experimenters, with more data to work with but had a knock-on effect through overheating of the magnet coils, since there was less time between cycles for them to cool. That was solved by a bigger fan and larger ducting to remove the heat, which in turn required action to minimise noise, since teaching was going on in neighbouring buildings. That was a constraint not suffered by the Cosmotron and Bevatron in their purpose-built environments. The move intensified the frustrating interference with other work, such as the operation of the cyclotron nearby, through the more frequent pulsing of the fringing magnetic field.

The most important new capability provided by the new system was for the acceleration of deuterons. Beams of deuterons (108 per pulse) at 650 MeV constituted “the highest beam energy of these particles available for experimentation”598. This new capacity again generated a new program of work; for example, the first D/D collisions at such high

597 Rolph (1995), p. 45. 598 JSCR 1960/61, p. 10. 389

energies, research undertaken in collaboration with Harwell and yielding data to further refine the theory of nucleon interactions and the production of pi mesons. There were the first glimpses of a reaction that produced two at once, apparently the result of the decay of a yet-unknown particle created in the collision but lasting only a fleeting moment (as little as 20-20 second). This was among the first “nuclear resonance” to be sighted, and study of the phenomenon was central to the work in the laboratory for some years to come.

Into the new decade, with a new control room, simplified operation and ongoing refurbishment of the injector, the Birmingham synchrotron entered a new period of productiveness. A second beam line, made possible by a grant from the National Institute for Research in Nuclear Science (NIRNS)599, allowed the laboratory to offer beam time to other users, including researchers from the Cavendish and the Clarendon and from the NIRNS itself. The JSCR report for 1962-63 commented “This collaboration is particularly welcome in view of the build-up of university teams for the accelerator program to be centred on the Rutherford Laboratory Accelerator (NIMROD) which is now in operation.”600 The collaboration was soon flowing the other way as well, as the Birmingham physicists extended their interests to higher energies by making use of NIMROD in collaboration with teams from Harwell and Bristol.

The process of enhancement was ongoing. As late as the last year of operation, there were plans to make the extracted pulse ten times longer (another benefit for the experimenters). This would have pushed existing technology to its limits by creating an immense current demand, some 35 000 amperes for a tenth of a second. Indeed, as Whitaker recalled, the University electrical engineering department said that it

599 The NIRNS had been formed in 1957 to operate the Rutherford High Energy Laboratory (later the Rutherford Appleton Laboratory or RAL) established next to Harwell. 600 JCSR 1962/3, p. 9. 390

could not be done, but “physicists are made of sterner stuff and such a comment could only make me more determined to have a go”601. Oliphant would have approved such a point of view. Initial development of the system was promising but a decision had already been made to close down the synchrotron, and the idea was not taken further.

The end of the road

The Annual Report of the Department in 1964/5602 presented two contrasting messages. On the one hand, the machine was at the height of its powers, with a beam intensity that had been rising over the years, pulses now available 15 times a minute and a flexible relationship available between the magnetic field and the accelerating frequency, all increasing the flow of data. As we have seen, access to the machine was much sought after, and from the early 1960s, it was being run around the clock to meet demand.

At the same time, there were signs it would soon be all over. “The synchrotron program is being slowly reduced to permit a greater emphasis on other aspects of the Department’s work in high energy physics.” In veiled terms, this comment indicated that a decision had already been made to shut the machine down. The Birmingham proton synchrotron was finally closed in August 1967, after fourteen years of operation.

Many of the synchrotron team, now “demobbed”, and much of their equipment, were diverted to an experiment to be undertaken at

601 Rolph (1995), p. 46. 602 JCSR 1964/5, p.10. 391

NIMROD. This would be in the new glamour area of “nuclear resonances”, now forming a growing family of unstable particles (particles with very short lifetimes), related to the nucleons and the known mesons such as the , but in ways not yet understood. The “double pion” resonance seen at Birmingham was one of the first of these to be found, but producing others required energies beyond those achievable at Birmingham. NIMROD with its available 7 GeV was big enough to take part in that game, and the new work had been stimulated by the glimpsing there of an excited “omega minus” resonance.

The closure left the machine itself to be dealt with. Technically it belonged to the DSIR who had paid for it over the previous 20 years; those rights were now held by the Science Research Council. As Moon and Ibbs recalled the matter603, Treasury rules did not allow the machine to be given to the university. The Department, without any reference to the university authorities, therefore offered to buy it for £3000, all it could afford. The sum was less than 1% of the material cost of the hardware, let alone the expenditure of labour and mental effort, but it was accepted. Ultimately various parts of the machine were sold for considerably more, so covering the cost of reconfiguring the space in which the synchrotron had stood, and providing “a useful balance for the Department’s general benefit”.

At the 1993 symposium, Whitaker604 recalled his “final, sad task” to arrange for the dismantling of the synchrotron and the sale of any components for which a buyer could be found (for example, the motor- generator set with its 30 tonne flywheel had been bought by the Culham laboratory), with the rest to go for scrap. The most valuable materials were the 30 tonnes of copper windings for the coils, and after that, the

603 Moon and Ibbs (1981), p. 32. 604 Rolph (1995), p. 109. 392

800 tonnes of laminated steel plates, in particular the 1/8th inch pole pieces. Many of the latter were retained and used by the workshop for several years as a source of very flat steel plates.

Whitaker recalled the pain with which he and others who had worked with the synchrotron over a number of years watched its breaking down by workmen who (understandably) had no such emotional attachment. Meticulously aligned steel plates were quickly cut away with blow torches and stacked up in heaps for disposal, the vacuum system which had served so well became merely a pile of porcelain bits and pieces, the magnet coils which had given such trouble on several occasions were reduced to a growing pile of chopped up lengths of copper conductors. As the dismantling continued, the measures which had been taken to deal with magnet failures in 1952 and 1958 were revealed.

There was however a legacy, even after the machine itself was but a memory. The dominance of the machine in the overall Birmingham nuclear physics program had been declining for several years, since the decision taken around 1964 not to extend its life. Resources had already been diverted elsewhere, particularly into the team whose special expertise lay in analysing the visual records of nuclear reactions as captured by films, bubble chambers, spark chambers and similar instruments. The group had been formed initially to process the results of experiments undertaken with the Birmingham machine, but had soon extended its reach beyond the university. By 1966, the Birmingham group was part of a major national effort to analyse the torrent of data pouring from the 60” British National Bubble Chamber installed at CERN605.

605 The 300 tonne magnet for this device had been built in the Birmingham laboratory and was completed in 1960/1. 393

Over the years, the work of the “visual techniques” group had loomed increasingly large in the program of the laboratory; from 1962/3 it had merited its own section in the annual report of the Department. The report for that year commented on the group’s blooming reputation. Later reports chronicled the steady upgrading of the group’s facilities, such as; semi-automatic track measuring machines with increasingly powerful support from IBM and KDF9 computers, first in London, later on campus.

The 1966/7 annual report noted that the group was now one of the major centres of activity in this field in the country, with an output of 600 analysed events a week. The status of the group had been recognized during the year by a grant from the Science Research Council for new automatic measuring equipment and computer facilities that would increase its weekly capacity to 4000 events. The group was thus able to shrug off the demise of the machine which had given it birth. Indeed it benefited from still more resources, human and material, following the closure of the synchrotron, making this line of work one of that machine’s enduring legacies.

Scientific outcomes: what did it discover?

Our second question in this assessment deals with the impact of the Birmingham proton synchrotron on nuclear physics. Did it contribute significantly to the growth of new knowledge of nuclear reactions at very high energies, leading to a breakthrough in methods for the release of nuclear energy? This had been the prime justification advanced by Oliphant for the enterprise, anticipating that important new phenomena would be revealed.

394

The technological purpose of the synchrotron was to generate a beam of very high energy particles. That it clearly did. Its usefulness in scientific terms depended on what experimenters did with that beam. In general terms they continued to follow a well trodden path, first explored decades before by Rutherford and his coworkers. The particles were fired at a range of targets in order to initiate nuclear reactions and interactions which could then be studied. Even with an extracted beam, no experimental work could be done without auxiliary apparatus to detect and characterise the particles produced by such nuclear events. Over the decade and a half lifetime of the Birmingham accelerator, a number of techniques were employed for this purpose, starting with electronic counters and nuclear emulsions (photographic plates), the latter until around 1958, and, once a beam could be extracted, moving to cloud chambers, bubble chambers and spark chambers which could graphically represent the tracks of both the particles and the products of their interactions606.

Almost up till its shutdown, the Birmingham machine had continued to toil at a task to which it was well suited, and to which it had been almost exclusively devoted, the study of the interaction at very high energy between one proton or neutron and another. One particle would be in a beam, the other in a target. The beam nucleon would be scattered by the target nucleons either “elastically” (without any loss of energy) or “inelastically”, in which case some of the energy was transformed into matter, in particular releasing or creating the pi mesons which now by general agreement were recognised as carrying the “nuclear force” that

606 All these systems could provide a visual representation in two and three dimensions of the paths of the interacting particles and of any reaction products. The passage of charged particles generated ionisation of the medium along their paths, then made visible by the formation of trails of water droplets (in cloud chambers), gaseous bubbles (in bubble chambers) or streaks of sparks (in spark chambers), each marking out the otherwise invisible particle tracks and able to be photographed. Analysis of the number and direction of tracks emerging from an interaction could yield information about the nature of the interaction and the energies involved. 395

controlled the reaction between nucleons (and bound them together in an atomic nucleus).

Such scattering experiments were therefore of fundamental importance, probing in detail the operation of this basic force, describing for example the effective size of the proton at such energies (around 10-15 m) and the influence of the polarisation (“spin”) of the proton on the interactions. The Birmingham machine briefly enjoyed some monopoly on this work, at least in Europe, since the only other accelerator able to undertake such studies at 1 BeV was at Brookhaven. No similar research were possible elsewhere in Europe until the 3 BeV machine at Salcay in France came on line in 1958. At the time of its inauguration, Birmingham was, in the recollections of Kinson, "the place to be in Europe for high energy physics"607.

The 1966-7 report of the JSCSR, written after the machine has been put to sleep, provided the following assessment of the outcomes of its 14 years of “useful life”.

In retrospect the achievements of the experimental groups working with the machine are seen mainly as increasing our understanding of the very short range force which acts between two colliding protons. This is a basic problem in nuclear physics and the synchrotron was oriented to this topic almost exclusively, even during the last year of its life, when measurements of very small angle scattering were undertaken to give further technical details of this force.

We have already noted that in proposing the building of an accelerator to reach 1000 MeV (above the rest mass of the nucleus), Oliphant had

607 Rolph (1995), p. 73. 396

expected (or perhaps merely hoped) that some new fundamental understanding would arise, transforming the technology for the release of nuclear energy. That did not happen, not in his machine, not in that energy range. In fact it has not happened even with the vastly higher energies we have available today. Our basic understanding of the release of energy by nuclear fission or remains as it was in 1940, though the techniques for exploiting it have greatly advanced. Even exploring the new world of the “resonances” which could be glimpsed from Birmingham but which mostly lay over the energy horizon, did not add knowledge of practical value, though it did lead to a deeper understanding of the arrangement of matter at a fundamental level.

At the same time, we must acknowledge that across the spectrum of science such cataclysmic breakthroughs are relatively rare. Nearly all advances in science come incrementally through the ever-more precise measurement and characterisation of physical phenomena. That the operation of the Birmingham synchrotron did not turn nuclear physics on its head cannot be taken to imply that it failed to make significant contributions to the step-by-step advance of the discipline. The machine was kept running, and as we have seen, productively so, for fourteen years after it first generated a beam. Clearly some value for science was being extracted.

Without a more detailed examination, it is not possible to say how sound or penetrating was the research done, though the value of the work was constrained to some extent by the relatively low intensity of the beam, which limited the production of useful beams of “secondary particles” such as pions. Likewise the relatively low beam energy (though unprecedented when compared with the previous generation of accelerators) prevented a discovery of the magnitude of the first

397

production of anti-protons by the Bevatron. We must note that the target energy of that latter machine had been set with such a goal in mind. Even so the discovery was not easily made and depended much on a sophisticated detection strategy.

We can at least put some numbers on the volume of research by quoting figures cited by Kinson608. From 1953 to 1964, 26 papers were published on work done in Birmingham using cloud and bubble chambers, which covered the major fields of experimental use. Sixteen papers dealt with equipment, 10 described physics results. In addition 13 post-graduates gained PhDs. Kinson also helped place the Birmingham contribution in its context by noting that the major developments resulting from the use of bubble chambers over the first 20 years dealt with the discovery of resonances and the first insights into the mechanism of exchanges between “hadrons”609. “Birmingham data played an important part in the study of the first examples of both of these”.

It may be at once thought that such numbers are low, given the magnitude of the investment, and that on that reckoning the machine would have the judged a “white elephant”. However Kinson seems to have not counted many other publications that do not comply with his narrow field of interest. Published papers were reported annually in the report of the Joint Standing Committee of Scientific Research. Scanning the titles of publications is not an infallible guide to the source of the research data being published (for example, some data may have come from the cyclotron which was operating at the same time), but the following table, I suggest, gives a conservative indication of the

608 Rolph (1995), p. 84. 609 “Hadron” is the current terminology for a sub-atomic particle that is subject to the strong nuclear force. It includes protons, neutrons and pi mesons. 398

published output the Birmingham-based research teams610. The listing is by year, and distinguishes between papers which reported technical developments in the machine itself and those which reported the outcome of research undertaken using the synchrotron.

Technical Scientific 1952-3 4 0 1953-4 5 0 1954-5 2 2 1955-6 2 3 1956-7 3 10 1957-8 2 4 1958-9 0 7 1959-60 2 6 1960-1 1 4 1961-2 0 3 1962-3 1 1 1963-4 0 6 1964-5 0 3 1965-6 0 2 Totals 22 49

Summing the two types of papers, the total comes to more than 70 over 14 years, nearly three times the figure Kinson quotes611, and an average of five a year. Whether or not this is a notable figure or merely an acceptable figure for such a pioneering facility could be judged only by comparing it with the published output of the Bevatron or the Cosmotron, a task yet to be undertaken. It is clear that the number of scientific papers substantially outweighs those dealing with technical matters, especially from 1954-5 where experimental results began to flow, and that productivity peaked in the late 1950’s. The decline in the early to mid-1960s no doubt reflects the transfer of resources to other activities once the fate of the machine had been determined.

610 It should be recalled that researchers from other institution were using the Birmingham machine from the early 1960’s, and the resulting publications may not be captured in the Birmingham listings. 611 We can note that the number of successful PhD candidates, as reported elsewhere, is also about three times Kinson’s figure. 399

In advance of any further analysis, perhaps we can take Lawson’s brief assessment as a summary of the scientific contribution of the Birmingham enterprise: “From 1963, the Birmingham Synchrotron program began to transfer to the new NIMROD synchrotron at the Rutherford Appleton Laboratory and then to CERN. In the preceding decade the machine had made possible a useful though not spectacular contribution to a specific field, and its existence had led to the emergence of a strong and experienced research group with potential for future work612.” This is no ringing endorsement of the value of the machine, but given its limitations perhaps none can be expected. However it stands well above the assessments of some (such of Alvarez that we shall consider later) that as a tool of research the machine was a waste of effort.

Technological outcomes: showing the way?

Beyond judging how well the Birmingham machine fulfilled its brief, there is a broader consideration. Did it contribute to the development of the technology of particle acceleration, showing the way to later generations of machines? Clearly, the successful operation of the first-generation machines, including that in Birmingham, validated the synchrotron principle, stimulating the building of later generations, even if there were significant changes in design details. Whether or not the Birmingham machine had any particular impact is harder to ascertain.

Oliphant had seen the application of the synchrotron concept as the obvious way forward to achieving very high energies. Yet, given that the synchrotron had been designed to overcome limitations in existing technologies, he must have foreseen that it too would in time (and perhaps quite quickly) be overtaken by newer developments. The

612 Lawson (1997), p. 34. 400

development which was ultimately to consign all the first-generation proton synchrotrons to obsolescence arose in the early 1950s in Brookhaven. As Courant later recounted events613, the stimulus towards a newer and more powerful type of proton synchrotron came from Europe, especially from the formation of the pan-European nuclear physics laboratory to be known as CERN614. Being able to draw on the resources of several governments, CERN proposed to build a machine even larger than the Bevatron, with a peak energy of 10 GeV (the target proposed in the original Bevatron design studies).

Preparing to receive a delegation from CERN for an exchange of ideas, Livingstone and his American colleagues undertook a review of their programs, stimulating a chain of investigation which generated a most important refinement of the synchrotron principle. This required the magnets generating the guiding magnetic field to face alternately inwards and outwards, that is, the radial gradient of the field was alternatively towards the centre of the machine and away from it. While not the original rationale, this arrangement was found to greatly enhance the focusing of the beam, containing it within an annulus only a few centimetres thick. This allowed the vacuum tube to be similarly reduced in cross-section, with a corresponding slimming-down of the cross- section of the magnet. The weight and cost of the magnet for a given energy could therefore be greatly reduced, or alternatively, a much higher energy be reached for a magnet of a given weight and cost.

“Strong focusing” as it was dubbed615, immediately raised the bar for new synchrotrons, with 25 GeV or more becoming a realistic

613 Courant (2003), p. 11. 614 For background on the early history of CERN see Pestre and Krige (1992). 615 The adoption of this terminology resulted in the first generation machines being dubbed “weak focusing” machines. 401

prospect616. The first synchrotron to go on-line embodying the new ideas was the CERN “PS”, which first generated a beam in 1959. It was closely followed by Brookhaven’s “alternating gradient synchrotron” (AGS), proposed in 1952, approved in 1954 and achieving operational status in 1960.

Can it then be said that the Birmingham venture influenced the development of the synchrotrons that came after? It had proved that such a machine could be built and do useful work, and that of course was a very considerable achievement. It had demonstrated that the operating principle was sound. In later machines, even of the same “weak focusing” first generation, the technical solutions adopted were different, just as the Birmingham machine itself was upgraded and older systems were supplanted. The NIMROD synchrotron more closely resembled the American machines than that in Birmingham (higher energy, housed in a purpose-built facility, straight sections in the particle path). By then of course the technology had spread far beyond the groups where it had first taken root.

Capacity-building outcomes: people and skills.

The answer to our third question begins with a consideration of educational outcomes. Did the building and operation of the Birmingham machine contribute significantly to the training of young physicists, equipping them with skills they could use with profit elsewhere, for example in the new facility at Harwell? This had been Oliphant’s secondary motivation, though it had loomed larger in the minds of some of those who had recommended funding of the venture.

616 As was later discovered, the concept of strong focusing had already been developed several years earlier by the Greek physicist Nicholas Christofilos, but his ideas, contained in a letter to Berkeley, had been ignored at the time. For background see http://hifweb.lbl.gov/public/Sharp/HIF_documents/Melissinos-Christofilos+physics.pdf

402

This concern can be broadened. In line with one of the characteristics assigned to Big Science617, namely, alignment with wider social, political or economic objectives beyond the venture itself or the science it sought to illuminate, did the operation of the Birmingham synchrotron contribute to British expertise in the exploitation of nuclear fission, in both the civilian and military spheres? Did it, in short, help build “national capacity”? As we have seen, the need for such national capacity was the reason that public (and to some extent private) investment in infrastructure for science was so strongly skewed toward nuclear physics in the decade following the war.

One way in which this could be assessed would be to track the later careers of students and staff who spent formative years working with the Birmingham synchrotron. Should we find that a significant number of these went to work in research and development in the applications of nuclear energy, we may be able to assert that the Birmingham enterprise did indeed help built national capacity. This has not yet been methodically done, though we do know that some Birmingham men, such as Lock and Ramm, went on to significant careers at CERN.

We can however put some numbers to the impact of the machine on the training of those fledgling nuclear physicists. A list of the PhDs awarded on the basis of work done with the Birmingham synchrotron has been compiled and is attached as Appendix A. This shows that from 1954-5 (the year from which comprehensive data is available) to 1967-8 (the years after the machine closed) 45 PhDs ascribable to involvement with the synchrotron were awarded. A few were for new technical developments, but the large majority were based on experimental work. The list does not include the PhDs earned by pioneering team members

617 These issues will be dealt with in Chapter 9. 403

such as Hibbard, Riddiford, Symonds, Caro and Ramm, all of which has been awarded by 1954, almost exclusively for their contributions to design and construction. In summary, the building and operation of the machine generated around 50 PhDs in a decade and a half, mostly to men (only one or two women) who went on to work elsewhere. This is one measurable contribution by the Birmingham enterprise to the “national capacity”.

There is strong evidence that the machine was much in demand, not only by the home team but by visitors, including researchers from the Atomic Energy Research Establishment (Harwell), close to the new 7 GeV NIMROD accelerator618. Galbraith, reminiscing at the 1993 symposium619, reported that the interaction between Birmingham and AERE extended to interchange of equipment when specific needs arose. Some of the equipment being developed for NIMROD was tested in Birmingham.

As a consequence of this demand, the Birmingham synchrotron was kept running from 0800 to midnight, and later, at Galbraith’s instigation, around the clock. He had amusing tales of what such night work involved, including that visitors would seek accommodation locales willing to provide “breakfast and bed”, rather than the other way round. While such evidence of use cannot speak for the quality of the physics being done, the machine was clearly not lying idle, even a decade after it was first fired up, and when it had been surpassed in energy and intensity by several other accelerators.

Around the time of which Galbraith was speaking, debate was beginning as to the long-term future of the machine, or indeed if it had one. The

618 Named after an Old Testament figure, renowned as a mighty hunter. 619 Rolph (1995), p. 125. 404

debate was to some extent driven by bureaucratic requirements. A decision was needed as to whether funding to operate and maintain the synchrotron should continue into the next quinquennium of public outlays, from 1967 to 1972. Obviously the decision would depend largely on whether the machine was continuing to fulfill the basic purposes, educational and scientific, for which it had been constructed.

Galbraith, responding in 1964620 to a letter from Burcham seeking advice on the matter, stated that, in his opinion, the strong support for keeping the Birmingham synchrotron running came mostly from the numerous Birmingham graduates who had learned their high energy physics using that synchrotron. Contrasting their experience with that that might have been gained by a young researcher using the contemporary machines at Brookhaven and Berkeley, he commented that (in Birmingham)

…you get involved in the whole experience from ion source to beam stop, whereas, with these larger monsters, the machine is a thing rather apart from the physicists. So as a physicist training machine, the Birmingham synchrotron is probably better than many of its younger, if bigger, brothers.

In his opinion, therefore, the Birmingham enterprise could be judged a significant success, at least in terms of its attainment of educational objectives.

What strikes me now, about the days in Birmingham, is the fact that such a complex machine could be built within a university laboratory and work so successfully over a period of many years. It had dedicated staff who kept the many complex pieces of

620 Galbraith to Burcham, 22 April 1964. Box 22X/8B, UBSC 405

electronics, which go to make up the synchrotron, operating and it produced a breed of experimentalists who now hold posts of responsibility in [high energy physics] around the world621.

The wider influence of the Birmingham enterprise

So far we have assessed the Birmingham synchrotron only in terms of its own objectives, technological, scientific and capacity-building. There is however a wider context which we should address. Given that the machine was the first of a new breed of particle accelerators, operating at unprecedented energies and embodying revolutionary principles, to what extent did it affect other similar enterprises, both those contemporaneous and those to come?

We must begin by assessing to what extent the nature of the Birmingham endeavour was known beyond the walls of that university. Here the circumstances of the time are important. In 1945, with the war still underway (until May in Europe, until August in the Pacific) and with Oliphant’s plans deriving so directly from the top-secret activity to build the “weapon”, it is reasonable to think that everything would be played close to the chest. As we noted earlier, such security considerations have been used by some to suggest that no-one outside Birmingham or official British government circles knew what Oliphant was planning until the appearance of the first description of the machine in the scientific press in March 1947.

An examination of the evidence suggests otherwise. In particular, those with whom Oliphant had been in contact in the immediate post-war period, Ernest Lawrence and some of his co-workers at Berkeley, seem

621 Ibid. 406

to have been in the know. By his own account, Oliphant had been keen to make them so. In his 1967 recollections622, he stated

So far I had said nothing to American colleagues about my ideas, but after I received Akers’ reply I discussed the principles involved, and the problems of construction, with Lawrence. He was kind, as always, but clearly quite uninterested, pinning his faith to larger cyclotrons. However, when, in 1945, he drove me to the airport in San Francisco and, as we said goodbye, he recalled my remarks and encouraged me to go ahead623.

These recollections were set down more than two decades after the event, but they are supported by a letter Oliphant wrote to Lawrence on 3 July 1945, less than four months after his departure from Berkeley624.

I have not yet been able to make more than paper progress on my own proposals for the production of artificial cosmic rays625. The more I think of it the more convinced I am that the suggestions which I discussed with you will work. The only large and awkward part of the equipment will be the power source necessary for pulsing the magnetic field. This will require a short circuit generator, or a condenser bank, to deliver about 2.6 MW.second of energy. From the engineering point of view there is no difficulty whatever in this last factor, but from the financial point of view and from the priority point of view even more, it represents a grave difficulty.

622 Oliphant (1967) Part 2, p. 4. 623 Oliphant made a similar assertion about Lawrence’s awareness of his plans in his 1951 letter to Mann, discussed in Chapter Six. 624 Oliphant to Lawrence, 3 July 1945. LPBL. 625 Oliphant did not use the term “synchrotron” to describe the machine, as that name had not yet been devised by McMillan. 407

In correspondence with the author in November 1980626, McMillan said of this letter “The last paragraph on the second page [the paragraph quoted above] is the relevant part in our present context. It doesn’t tell me much about the design, and I don’t think it did to Lawrence either, and neither of us were (sic) very clear about what Oliphant planned.” McMillan thereby indicated that he and Lawrence had discussed the letter at the time and that it made little sense to either of them.

The paragraph quoted clearly required prior knowledge to be intelligible. Brief as it is, it contained two significant technical details; the project was intended to produce particles of extremely high energy (“artificial cosmic rays”) and that is would involve a pulsed magnetic field. It is clear that Oliphant believed he had discussed the matter with Lawrence, and that therefore Lawrence would understand what he was alluding to. He did not need to give the background to the particular issue he raised. McMillan would not have had that background, as Oliphant never claimed to have talked with McMillan about his proposal. Indeed, as we have seen, he would have had little opportunity to do so since McMillan had been based at Los Alamos throughout 1944, while Oliphant moved between Berkeley and Oak Ridge.

Of perhaps even more significance is the following quotation, already cited, from Oliphant’s January 1945 letter to Akers627, written while he was still in the United States.

I must say I have discussed [the proposed accelerator] with Lawrence on two occasions. He does not commit himself. But he says he can see no reason why it should not work. Of course he is very occupied at present with other things – as no doubt I

626 McMillan to Ellyard (1980), 5 November 1980. EBA. 627 Oliphant to Akers, 17 January 1945. EBA. 408

should be also – and it is clear he has given the scheme little attention. But I do not think he would let it pass if there were a fundamental flaw in the argument. His instinct in these matters is uncanny.

Is there contrary evidence in this matter? The project was not mentioned in Oliphant’s “farewell” letter to Lawrence628, though that was written prior to his departure (and therefore before the two reputedly discussed the matter on the way to the airport), or in Lawrence’s reply on 20 March629. Given the nature and purpose of those letters, that is perhaps not surprising.

Of more impact is a letter from Lawrence to Oliphant in April 1946630. This says, inter alia, “During the latter days at Los Alamos, McMillan devised the synchrotron which you will doubtless read about in the Physical Review and in other ways”. The implications of this letter may well have shocked Oliphant. If Lawrence had known of Oliphant’s plans a year earlier, it appears he had forgotten or that he did not recall them as being commensurate with McMillan’s proposals. An alternative interpretation could suggest that perhaps Lawrence did not consider Oliphant’s proposal fitted the definition of a synchrotron,

Perhaps the crucial evidence in this debate was provided by Lawrence in the letter in 1952 to Wilfrid Mann of the US National Bureau of Standards, already cited. In December 1951, Mann was seeking to add an epilogue to his monograph on the cyclotron (then in its fourth edition) and wrote to Lawrence631 to seek his help. In the course of his research he had come across references to Oliphant’s 1943 proposal, and

628 Oliphant to Lawrence, 16 March 1945. LPBL 629 Lawrence to Oliphant, 20 March 1945. LPBL. 630 Lawrence to Oliphant, 20 April 1946. LPBL . 631 Mann to Lawrence, 19 December 1951. Attachment K to McMillan to Ellyard 8 November 1980. EBA. 409

wondered how that fitted into the work of McMillan, Veksler and others. In his reply Lawrence wrote

I rather vaguely remember Oliphant telling me about 1943 about pulsing a current through some coils and at the same time pulsing a frequency-modulated RF accelerator and, if I remember correctly, I commented at the time that the idea certainly looked feasible in principle. His idea seemed more akin to an air-cored synchrotron, though I don’t recall his mentioning the utilisation of phase stability in achieving the acceleration. At least I did not appreciate the point at the time632.

The key point here was that Oliphant was, in Lawrence’s recollection, proposing to pulse both the magnetic field and the accelerating voltage simultaneously, as would be needed to confine the particles to an orbit of fixed radius (though Lawrence does not make the last connection). On balance therefore, and despite McMillan’s scepticism, it appears likely that Lawrence did know something of Oliphant’s proposed accelerator, though perhaps not much beyond basic design concepts and operating principles. Given his work load and his many other concerns, it was, not surprisingly, a matter that did not engage him much and which he could apparently forget when not reminded of it for a year or so.

By August 1946, there was no doubt in Lawrence’s mind that Oliphant had conceived a synchrotron. He credited him with as much in his review of current and potential accelerator technologies633, already cited in Chapter 2. “The synchrotron which was conceived independently here

632 Lawrence to Mann, 11 January 1952. Attachment M to McMillan to Ellyard 8 November 1980. EBA. 633 Lawrence EO. Experimental Methods in Nuclear Physics, 20 August 1946. EOL 72/117C/40/28. LPBL. 410

[i.e. Berkeley], in England and in Russia offers an attractive approach to even higher electron energies…. .” This was most likely the first time independent discovery was publicly ascribed to McMillan, Oliphant and Veksler. The synchrotron was described by Lawrence in his review only as an accelerator of electrons, though he saw the synchro-cyclotron as a way of accelerating protons to hundreds of millions of eV.

Through further analysis of correspondence between Berkeley and Birmingham we can see how the awareness of the Birmingham enterprise grew among Lawrence, McMillan and their colleagues. In the chronicle of correspondence provided to the author by McMillan in 1980 the next letter cited is dated 20 November 1946634. Oliphant wrote

I am sending herewith some laboratory reports on the synchrotron. There may be some ideas in them which could be useful to you. I hear that you hope to be in operation in January. We are making good progress but do not anticipate having any beam for about eighteen months. I hope that it won’t be very long before I am able to discuss these things with you in person again.

Again the letter seems to assume that the nature of the Birmingham enterprise was known to the Berkeley team, and indeed to McMillan himself. The last paragraph could be read to imply that Oliphant had previously talked to McMillan about it, but as noted above, this is unlikely. Perhaps Oliphant assumed that by now Lawrence would have spread his knowledge of his plans to other Berkeley colleagues

634 Oliphant to McMillan, 20 November 1946. Copy provided to author by McMillan in 1980. EBA. 411

McMillan replied on 30 December 1946635, beginning his letter “Dear Professor Oliphant” and so eschewing the more familiar manner of address that Oliphant had used to him. He continued “The papers that you sent came some days ago and I have found them very interesting. Dr Serber is looking them over and I had planned to send you his comments, but it is taking so long to get them (the theorists are a pretty busy group these days) that I decided not to delay my reply any longer.”

After commenting that the January start date (for the electron synchrotron) had slipped to June, McMillan went on “I would like to hear more about the machine you are building. Is it for protons or electrons; iron or air core; what is the field strength, radius, aperture etc.?” Commenting in 1980 on Oliphant’s letter and his reply, McMillan remarked “The “laboratory reports on the synchrotron” seem to be lost and I can’t remember them, but the third paragraph of my reply indicates that they did not include design data.”

Though McMillan said he found the laboratory reports “very interesting”, he did not offer any comment on them himself. He planned to include comments from a theoretician, comments that may never have eventuated and were apparently not recorded. Instead he claimed almost complete ignorance of the nature of Oliphant’s machine, asking quite basic questions.

In January 1947, McMillan received a letter from F K Goward636, working at the Telecommunications Research Establishment (TRE) which formed part of the British Ministry of Supply. During the war TRE had been a centre of activity for radar development. The letter was in response to one from McMillan, dated 15 November 1946, and can be

635 McMillan to Oliphant, 30 December 1946. Copy provided to author by McMillan in 1980. EBA. 636 Goward to McMillan, 13 January 1947. Copy provided to author by McMillan in 1980. EBA 412

read to imply that McMillan was seeking information about synchrotron development in Britain. After describing the progress of his own work to convert a 4 MeV betatron into a 16 MeV synchrotron, and plans for a 30 MeV machine, Goward went on “In collaboration with Metropolitan Vickers Co, we are making a 300 MeV machine for use at Glasgow University. This is only at the design stage637.” The details provided here suggest that this is the machine for which funds were recommended by the Advisory Committee on Atomic Energy (ACAE) in March 1946 though at that time the machine was described as a betatron.

Goward’s next comment provided, according to McMillan’s recollections in 1980, “the first substantive information that I got from any source regarding Oliphant’s machine”.

Prof. Oliphant at Birmingham University is constructing a synchrotron to produce about 1300 MeV protons. This is in about the same stage as the 300 MeV machine, perhaps a little ahead. It will use quite a slow magnetic field change (about 1 sec. time of rise) and will change radio frequency as well as magnetic field. Some of the parameters are rather frightening, e.g. the radio frequency bandwidth required (~30/1) and the close tolerances on it, but everyone is enthusiastic and hopeful.

For a further perspective, we can cite correspondence from 1974 between Oliphant and Paul Forman, Curator, Department of Science and Technology at the Smithsonian Institution638. Forman had written

637 This is the machine proposed by Dee and funded in parallel with the Birmingham project. 638 Forman to Oliphant and reply, March 1974. EBA. 413

In his 1951 Nobel Lecture, ETS Walton credited you, along with McMillan and Veksler, with conceiving the principle of phase stability for the acceleration of particles to relativistic energies. He cites a proposal submitted by you to the Directorate of Atomic Energy in 1943.

I would be grateful if you could tell me how I might obtain a copy of that proposal, for my responsibilities here centre on the development of the tools of nuclear physics.

Oliphant replied

Yes, a proposal for a proton synchrotron was formulated and sent to the British “Tube Alloys” chairman Sir Wallace Akers, while I was working in the United States. I also discussed this with EO Lawrence in Berkeley. I received the answer I should have expected at the time – that consideration of a research project must wait till after the war. However, through the cautious habit of those hectic days, the proposal was classified secret, and along with some other letters and documents, could not be found when the files were handed over to the Atomic Energy Authority after that was established.

My proposal was in very general terms, phase stability being established by considering what happened to a single proton as it drifted in phase. There was no consideration of the problem with the cyclotron, so that I missed the synchrocyclotron idea. Moreover, I made no exhaustive analysis, as did McMillan and Veksler, so Walton’s reference to me was an over-generous gesture.

414

We have another source of perspective on these developments, namely the correspondence in 1951/2 by various parties with Wilfrid Mann, the circumstances of which have been described above. Lawrence’s letter to Mann, acknowledging some early awareness of Oliphant’s plans, has already been quoted.

Oliphant’s response to Mann’s enquiries639, which preceded Mann’s letter to Lawrence, contained some familiar material; mention of the first submission to Akers (“late in 1943”) and Akers’ purported “get on with the war” response, a brief description of the proposed machine, a comment on how elementary was his treatment of phase stability (and that it did not proceed from McMillan’s idea of stationary orbits). He then reported (tantalizingly, in the current context) “I had some correspondence with MacMillan (sic) about the idea [of orbital stability] before he published his papers (author’s emphasis) because Titterton had told him I was working on the same problem”. We currently have no copies of such correspondence, and McMillan later strongly denied any such early exchange on such matters, but Oliphant’s claim remains to be refuted.

Having been shown the correspondence by Lawrence, McMillan replied to Mann on 3 January 1952640, offering “to tell you what I can of the Oliphant story”. He stated that the concept of phase stability had come to him in late June or early July of 1945, and continued

After explaining this to many people I began to hear rumours of other accelerators of similar kinds. I believe it was Peierls, not

639 Oliphant to Mann, 11 December 1951. Attachment J to McMillan to Ellyard 8 November 1980. EBA. 640 McMillan to Mann, 3 January 1952. Attachment L to McMillan to Ellyard 8 November 1980. EBA. 415

Titterton641, who first told me that Oliphant had a similar idea, but he was rather vague about its exact nature and agreed to write to Oliphant for more details. I don’t believe I ever wrote Oliphant directly or ever got a letter from him at that time, but Peierls did get some sort of communication (I think from a staff member at Birmingham, not from Oliphant himself?) and I think it was probably from that that I got my idea of Oliphant’s original proposal.

The field was to be provided by air-cored coils, and was to rise very rapidly; acceleration was to be provided by a ring of accelerating electrodes, driven by a polyphase oscillator; the reason for the polyphase acceleration was that the energy gain per turn was so great that it would disturb the orbit if applied at one point; no mention was made of phase stability.

At the time of the 1947 papers there was clearly no need for a rapid rise of magnetic field and the published design had an iron core magnet. In 1946, I asked Gooden (then in Berkeley) whether the 1943 proposal involved phase stability; he said no, they had got that from my 1945 letter in the Phys. Rev. In 1949, I saw Oliphant in Birmingham and inquired about the 1943 proposal, which I wanted to see; he said he could not find it. Of course in 1943 to 1945, people were more interested in other things and it is not strange that poor records were kept and memories are vague.

This commentary raises interesting points, the reference to Peierls as the source of information about Oliphant’s plans being particularly

641 Both and , the former a Birmingham colleague of Oliphant’s, the latter a former student, had been working at Los Alamos (as had McMillan) during the last years of the war. 416

intriguing. That meshes with existing information. As was noted earlier642, it was Peierls who, according to Moon, brought Oliphant’s several-times-promised memorandum on the new accelerator to the USA late in 1945. By that time, as Moon stated in a letter, McMillan had already returned from Los Alamos (where Moon was still stationed) to Berkeley. Did Peierls then visit Berkeley and interact with McMillan? Or was there some other avenue by which information was passed on? We have no sure answer at the present time.

I suggest that it is feasible that some understanding of the Oliphant vision could have been current in Berkeley as soon as early 1946. This is not supported by McMillan, but his recollections need to be viewed with some caution. He was almost certainly incorrect in asserting that Lawrence knew nothing of significance about Oliphant’s plans. His description of what he came to understand of Oliphant’s design, as an air-cored synchrotron with a number of accelerating electrodes (but with no comment about phase stability) reflects the content of the Oliphant Memorandum (and his June 1945 submission to Akers) rather than the machine Oliphant began to build in late 1946 (or the one he proposed in January 1945). That suggests that McMillan’s information came sooner rather than later. McMillan’s memory may have been faulty on some issues. As he had commented, people had their minds on other matters at this time and details could be understandably vague.

The implications of this line of argument, if it can be sustained, are profound. As we shall see, the building of a proton synchrotron at Brookhaven (the Cosmotron) was stimulated by developments at Berkeley, and would, in all probability, not have proceeded without such stimulus. If it can be shown that Oliphant, in whatever way, stimulated events at Berkeley, then his central role in this revolution in particle

642 In Chapter 6. 417

physics will be established beyond doubt. We shall now turn to evidence that such an influence was indeed exerted. It will also serve to explain Lawrence’s change of mind between April and August 1946 as to Oliphant’s role in this development.

William Brobeck and the Bevatron

It is time to consider the origins of the second of the first generation synchrotrons, the Bevatron project undertaken at Berkeley. By general agreement, the driving force behind this initiative to accelerate protons was not McMillan, who was busy with his electron synchrotron, but his colleague William Brobeck643. Certainly, it was Brobeck who first recounted the origins of the Bevatron in an article published in 1957644. He cited the development of the principle of phase stability by Veksler and McMillan as the primary stimulus, leading to its utilisation in the electron synchrotron and the frequency-modulated cyclotron (synchro- cyclotron). He noted that its use raised the energy limit of the machines to about 300 to 500 MeV.

Once the practicality of such machines had been established “by 1947”, attention turned to more complex applications, involving variations in both magnetic field and accelerating frequency, so that protons could be accelerated. Preliminary design studies indicated that the energy limit could thereby be increased tenfold over that for the synchrocyclotron. As published in 1948, these studies indicated the feasibility of a 10 GeV machine operating on such principles.

643 William Brobeck (1908-1998), American physicist and engineer, and a leading figure at the Berkeley Radiation Laboratory for most of his career. Assistant Director and Chief Engineer 1937-1957. Responsible for the design of several generations of cyclotron, including the 184 inch. For background see http://www.lbl.gov/Science-Articles/Archive/bill-brobeck.html 644 Brobeck (1957). 418

Importantly, for our purposes, Brobeck explicitly acknowledged the influence of the Birmingham project on developments at Berkeley. “This work was encouraged by information from the group at Birmingham, England under Professor Oliphant who, it was learned, had been working on the proton synchrotron since before the end of World War II. In this connection, Brobeck cited the March 1947 paper by Oliphant, Gooden and Hide, though this does not exclude the possibility of earlier awareness as indicated above. The choice of the expression “by 1947” indicates that the thought process has already begun. Brobeck also reported that discussion had taken place with researchers at Brookhaven who were similarly interested in a new form of particle accelerator.

This acknowledgement of the impact of awareness of the Birmingham venture is in strong contrast to McMillan’s disavowal of any such influence. Writing in 1980, he commented “The proposal of Oliphant had not been published by the end of 1946 and our knowledge of it at Berkeley was extremely vague during the early design period of the Bevatron, so that I do not recall that it had any influence at all” (author’s emphasis).

Some of the back-story to Brobeck’s account was provided by McMillan in his 1980 commentary. He reported that following the development of the concept of phase stability and the proposal that it be applied to extend the energy limit on the cyclotron, attention turned to “more complicated” schemes in which both the magnetic field and the frequency were varied together. According to McMillan, this was first proposed at Berkeley by Brobeck “sometime in 1946”.

I recall many discussions with him on the topic, but have no notes or documentary evidence of dates, except for a drawing labeled

419

November 1946 which is a conceptual design645 of the Bevatron in considerable detail. Shortly after this drawing was made, II Rabi of Columbia University visited Berkeley and took a copy back with him. This was how the proton synchrotron design got to Brookhaven646.

This account impacts on two elements of the story. Firstly, it helps set the timeframe within which Brobeck was developing his thoughts on the “proton synchrotron”. It indicated that the starting point lay back in 1946, a timing supported by Brobeck’s use of the phrase “by 1947” in the account quoted earlier. Secondly, we have the information that the concept of the proton synchrotron reached Brookhaven sometime later through the agency of Rabi.

Brobeck reinforced and amplified his recollections, including the involvement of Oliphant, in a substantial interview recorded in 1975/6647. By his account, the concept of what later became the Bevatron arose in the context of the transition of the Radiation Laboratory from a wartime to a peacetime orientation648. To bring the wartime intake of engineers, who had been totally involved in the development of the “calutrons” for isotope separation, up to speed in the broader field of particle acceleration, a series of lectures was staged, beginning in January, 1946. Early in these presentations, McMillan outlined his concept of phase stability and a range of possible applications was canvassed. Those included a synchrotron to accelerate electrons (fixed-frequency, varying magnetic field), and the synchrocyclotron (fixed field, variable frequency). To prove the feasibility of the latter, the 37 inch cyclotron

645 This was the one document referred to in this correspondence of which McMillan did not provide a copy to the author, so his assessment of it cannot be endorsed. 646 McMillan to Ellyard, op .cit. 647 Brobeck WM. Interview with Graham Hale conducted in 1975/6. Held in the Neils Bohr Library of the American Institute of Physics. 648 Oliphant’s first thoughts on the synchrotron had of course been driven by a similar consideration of “what to do after the war”. 420

was to be rebuilt as a frequency-modulated cyclotron. Work was soon underway on both those projects.

It was appropriate that the simplest of the three possible combinations of frequency and magnetic field variations were addressed first, but thinking soon turned to the more complex application, with both frequency and field being varied simultaneously, and which could be used, as McMillan had noted briefly in his 1945 paper, to accelerate protons (“heavy ions”). It appears that Brobeck, up until then working on designs for the synchrocyclotron, took the lead in such deliberations, making “sketches and calculations”, which he could date only approximately to “late 1946”. A drawing he made of the proposed machine was used by Lawrence in a lecture, possibly at Yale, and was shown to Rabi when he visited Berkeley for the inauguration of the 37 inch synchrocyclotron649. An enthusiastic Rabi took the idea back east where, as we shall see, it played a key role in the decision to build the Cosmotron.

As Brobeck recalled the circumstances, enthusiasm for the new machine, perhaps including Rabi’s, came from knowing that “Oliphant in England was working on this type of machine”.

During the war, apparently, they had been thinking about it in England, but didn’t have the principle of phase stability in mind, from what I understood. Oliphant had been in Berkeley during the war, went back to England, and then came back and told their thoughts about this machine. This encouraged me quite a bit. When you have an idea, you wonder if it’s a really good one or not, but they had been working along the same lines650.

649 The date of this event has yet to be established. 650 Brobeck(1957), p. 142. 421

According to Brobeck, Oliphant had visited the Radiation Laboratory in 1946. At that time he gave a lecture on the proposed Birmingham machine, that being the first time Brobeck and others had known what was planned. Correspondence confirms that Oliphant did visit Berkeley at the end of May 1946, having come to the U.S. as an adviser on atomic energy matters to the Australian delegation to the fledgling General Assembly. He was accompanied to California by GT Briggs651, another scientist and member of the Australian delegation652. There is no record of Oliphant visiting the US again in 1946.

All this being so, we can feel confident that the Berkeley community was aware of Oliphant’s thoughts on the synchrotron, and of his plans to build one653, by the middle of 1946, contrary to McMillan’s assertions654. This was concurrent with Brobeck’s own musings on the matter, and he later recalled that he gained confidence from that awareness. ”It made it more likely that [such a machine] would succeed, and also the competition was good”. How profound Oliphant’s influence was on the details of the design of the machine that Brobeck was tentatively planning is not clear; his machine was for example for a much higher energy than Oliphant’s, 10 BeV rather than 1 BeV, his was planned to have straight sections where Oliphant’s had none, and so on. However Oliphant’s stimulus to Brobeck’s plans cannot be denied.

651 GT (George) Briggs was then Chief of the CSIR Division of Physics and appropriately chosen by Evatt as his scientific advisor at the UN meeting. He had worked at the Cavendish on two occasions, once during Oliphant’s time. See Sherratt (1993). 652 Oliphant to Lawrence, 22 May 1946, Briggs to Lawrence, 13 June 1946. LPBL. 653 By this time, Oliphant had secured funding for his synchrotron and was reworking the design. The synchrotron committee at Birmingham would meet for the first time only four months later. 654 As we have noted above, Lawrence moved from apparent ignorance in April of Oliphant’s involvement in synchrotron development to an acknowledgement of that involvement in August. We can argue that Oliphant’s visit in late May had changed his mind. 422

The outcome of the deliberations at Berkeley by Brobeck, Lawrence and others was a proposal to the Atomic Energy Commission in February 1948 for funds to build a 6 GeV proton synchrotron, to be located close to the Radiation Laboratory above the Berkeley campus. The energy target was determined by the calculated energy required to generate proton-antiproton pairs (5.6 GeV). This reminds us that Oliphant had set the target energy for the Birmingham machine using a similar physical parameter, namely the rest energy of a proton.

McMillan’s 1980 version of this decision ran as follows.

Sometime later there was a meeting at the AEC headquarters in Washington DC, at which I spoke, along with someone from Brookhaven, Enrico Fermi, and I think some others and the decision was made to support the construction of both a six BeV machine of this kind at Berkeley and a three BeV version in Brookhaven655.

The timeline for the construction of this machine bears interesting comparison with that at Birmingham. Unlike the latter, the machine was to be housed in a purpose-built facility on a site that had to be cleared and leveled. That undoubtedly added time to the construction. Magnet erection did not get under way until 1950; coil winding was not complete until mid 1952. The power supply (provided, as in the Birmingham machine, by motor-generator sets) was tested, without pole pieces in place, in July and August, 1952. Following placement of the pole pieces, the vacuum tank was in place during 1953, with the injector operational in June that year. The first low energy beam was detected in February 1954. This was, as Brobeck noted, almost six years after the approval of funding for the machine. However he claims that most of the

655 McMillan to Ellyard op. cit.. 423

laboratory’s effort in 1951 and 1952 was devoted to unspecified projects of a classified nature. Without this diversion of effort, the completion time might have been reduced to four and a half years.

Brobeck reported that the basic cost of the Bevatron, that is, net of experimental facilities or shielding, was $9 million. This represented some $1.5 million per GeV. The increased cost per GeV brought some benefits. As Brobeck noted, the design of the machine allowed for flexibility; changes could be made to some elements of the facility without major disruption to the rest. For example, any of the 20,000 steel plates in the magnet could be taken apart for, say, re-insulating as needed, since they had been bolted together rather than welded. Assembly was therefore reversible. This flexibility undoubtedly added to the cost. We have noted that lack of such flexibility through appropriate design in the Birmingham enterprise delayed completion, though it can be argued whether or not it increased costs.

Another parameter for resources employed is the number of people involved in the enterprise. Brobeck suggested that “approximately 300 Laboratory people have been deserving of credit in the direction and prosecution of the work”. A listing of the contributors to the Birmingham project would contain perhaps 50 similarly-qualified names, and perhaps as few as 30, given that (no doubt as at Berkeley) some made short- term contributions and did not see the project completed. Comparisons are difficult, but it is hard to avoid the conclusion that in terms of manpower, as well as expenditure, the Birmingham enterprise generated high-energy protons at a much lower cost per GeV than its [almost] contemporary at Berkeley. This is an interesting observation, since the scaling-up of hardware commonly results in lower unit costs.

Stanley Livingstone and the Cosmotron

424

The third of the first-generation proton synchrotrons was the Cosmotron, erected at a new facility established at a former military base on Long Island, New York and later called the Brookhaven National Laboratory (BNL). Its origins were more complex than either the Birmingham machine or the Bevatron. Both of those were the results of initiatives within a single university, and both were designed at the outset as proton synchrotrons. BNL was on the other hand established as a joint venture by a consortium of East Coast universities656 to house facilities such as a nuclear reactor too expensive for any one institution to build. One of these was to be a proton accelerator, but not initially a synchrotron.

Details of what was initially planned can be found in the Minutes of the UK Nuclear Physics Subcommittee in early 1947657, which reproduced a report from the US prepared by Livingston658 from MIT. The report itself is undated but was presumably generated in the closing months of 1946. The report proposed that a synchrocyclotron (an “FM cyclotron”) in the 500 to 600 MeV range be built at Brookhaven, since that was the only proven design available for immediate start in this energy range. The report noted that work was proceeding on other designs for accelerating protons, including synchrotrons and linear accelerators. Those should not be neglected since some other design might be preferable at energies approaching 1 BeV. The report implied that the alternative design would more likely be a linear accelerator.

656 The universities involved were Columbia, Cornell, Harvard, Johns Hopkins, MIT, Princeton, Pennsylvania, Rochester and Yale. The site of the laboratory, formerly Camp Upton, was transferred from the US War Department to the Atomic Energy Commission on 21 March 1947. For background on the development of Brookhaven, see Crease (1999). 657 Document A.C.A.E. (N.P.) (46) 1, 10 January 1946, AB 16/1584, UKNA. This report had been reproduced on the grounds that it was relevant to plans being made in the UK. 658( Milton) Stanley Livingstone (1905-1986), American physicist and accelerator pioneer. He had spent the 1930s at Berkeley working on cyclotrons but moved to MIT in 1938. He later took charge of the design and construction of the Cosmotron. For a biographical memoir, see http://www.nap.edu/html/biomems/mlivingston.pdf 425

Of the possibility of building a synchrotron, Livingston’s’ report noted that

..widely varying designs are now in the planning or construction phase, including an ironless type. The best design for 1 BeV will not be evident for at least a year. Entrance of the National Laboratories in this competition will not materially shorten this time. The most uncertain factor at present is the determination of the aperture ratio for large sizes and hence the magnet iron requirements.

In little more than a year, those plans were overturned. By early 1948, funding was to be available from the Atomic Energy Commission for the building of a proton synchrotron at Brookhaven, with a target energy of 3 GeV, in addition to the 6 GeV machine at Berkeley.

A brief account by Livingston of the change in plans was contained in a letter to McMillan in 1980, in response to one from McMillan stimulated by the author’s enquiries659. The relevant paragraph of his letter is worth quoting in full.

The person most responsible for starting Brookhaven on the proton synchrotron (later known as the cosmotron) was Prof. II Rabi. His arguments to Phil Morse, myself and others at Brookhaven were based on his knowledge of the rapidly growing interests at Berkeley, particularly by you and Bill Brobeck. He did not mention any knowledge of plans going on at Birmingham. At Brookhaven we had originally planned a large synchrocyclotron, but changed our goal to the cosmotron following Rabi’s

659 Livingstone to McMillan, 12 November 1980. EBA. 426

stimulation. My visit to Birmingham in 1947 was following much design study and development and I gained essentially no useful information from their plans.

This statement supports the account given by McMillan above, underlining that Brobeck was the driving force in this matter at Berkeley, and that Rabi had conveyed the concept of the proton synchrotron to Brookhaven. That Rabi made no reference to work at Birmingham, at least in Livingston’s recollection, is perhaps not surprising, if indeed the major source of information about progress at Birmingham was the publication of Oliphant, Gooden and Hide660 in March 1947. That may well have followed the Rabi interaction with Brookhaven.

Blewett later reported661 that Rabi’s chief motivation in promoting the proton synchrotron to Brookhaven was his firm belief that 750 MeV, which appeared to be as high as the largest technically and economically feasible synchrocyclotron could reach, was insufficient for significant progress in nuclear physics662. The target had to be at least 1 billion eV and hopefully more. Of the designs on offer, only the (at that time) unproven proton synchrotron could offer such energies.

Did Birmingham influence Brookhaven?

We have already traced a path of influence on Brookhaven via Rabi back to Berkeley and thence plausibly back to Oliphant. Later influence had also been suggested. In his letter to McMillan of November 1980 Livingston commented

660 Oliphant. Gooden and Hide (1947). 661 Blewett (1987), p. 2. 662 We can see here a parallel with Oliphant’s thinking about the importance of the highest possible energies. 427

We at Brookhaven were aware of the paralleling design and development work going on at Birmingham. Our awareness was reported in a paper … by Livingston, Green, Blewett and Hayworth663 in the introduction. We had heard of Oliphant’s report to the English Directorate of Atomic Energy in 1943 but never saw any written evidence.

The paper referred to by Livingston attributed the first proposal for a proton accelerator using a ring magnet with simultaneous variation in both magnetic field and accelerating electric field to Oliphant, but based that attribution on the purported 1943 proposal referred to in Oliphant, Gooden and Hide. Some contact with the Birmingham project must have been maintained, as Livingston recalled in 1980, since the Livingston et al paper noted that the construction of a 1.3 BeV machine was underway in Birmingham (as of March 1949) and that completion was anticipated in 1950. However, information on progress at Birmingham could have reached Brookhaven from a number of sources. That does not imply any more intimate association, or any particular influence of Birmingham on Brookhaven, though Livingston did acknowledge a visit to Birmingham in 1947 from which he gained “essentially no useful information”.

A different point of view was provided by some of Oliphant’s colleagues, mostly notably Hibbard. In his contribution to the 2000 Inall memoir664, Hibbard recalled the Livingston visit.

During 1946, Dr M Stanley Livingston, who was famous for having successfully operated the first cyclotron in the world, had been commissioned to conduct a series of meetings throughout

663 Livingston, Blewett, Green and Hayworth (1950). 664 Hibbard in Inall (2000), p 7. 428

the USA in order to formulate a set of recommendations for future accelerator development. His report did not mention a proton synchrotron; instead he proposed the construction of a very large cyclotron665. Despite this, having heard of definite moves in Birmingham, he paid the University a visit towards the end of 1947, and on returning to America cancelled his earlier recommendations and made an immediate proposal for a new accelerator larger than the one under construction in Birmingham. This demonstrates that whatever the merits of the Birmingham program, Oliphant’s project guided US policy at the highest level. It was in fact the third time that Oliphant had had a profound influence on US scientific policies, counting his wartime stimuli to the Radar and Atomic Weapons programs.

To say that Oliphant and the Birmingham venture changed the course of accelerator development at Brookhaven, and therefore in the US, is a large claim, especially when Livingston explicitly denied it. Hibbard fleshed out the claim by reference to specific details of design, notably to the merits of having sections of the orbital path of the protons which were not subject to the strong constraining magnetic field (this would require that such parts of the orbit be straight). This would allow the use of another form of acceleration, through the application of an alternating magnetic flux, as in the betatron. Hibbard claimed to have discussed this possibility with Oliphant prior to Livingston’s visit and even gave a date for a meeting with Oliphant, Sayers and Moon at which it was discussed666. The Birmingham machine was, for reasons of space,

665 This was in fact to be a synchro-cyclotron. 666 The date quoted, 28 November 1947, if correct, helps date Livingston’s visit, which must have taken place after that date. That would support Livingstone’s contention that the visit occurred too late to affect the design of the Cosmotron. However, another source points to an earlier date. The Lawrence Papers contain a one-page list of “Electronuclear Machine in England” visited by Livingstone during a visit between 4 July and 3 August. See EOL 72/117/1/8, LPBL. The matter must remain unresolved for the present. 429

already committed to a circular form (indeed, the ring-shaped magnet was already being built). That was not yet so for Brookhaven.

Thus, when Livingston visited us, both Oliphant and I went out of our way to emphasise the desirability of having some part of the particle orbit that was free of magnetic field, to allow for the new development. On his return to the USA, he listed this as a feature of his new proposals667.

These assertions are purely from Hibbard, and as noted above Livingston denied that any significant elements of the design of the Cosmotron resulted from his visit to Birmingham. Nonetheless, the Brookhaven machine as built did contain four straight segments, as Hibbard claimed to have urged on Livingston. For the present, that is as far as the evidence goes.

It is not clear how widespread his view was that developments in Birmingham had influenced the choices made at Brookhaven, but it was certainly current. Moon for example later commented

The Brookhaven synchrotron – started later but with more resources - had of course overtaken the Birmingham machine in energy and, incidentally, in beam current before it reached its own higher design energy. Though I have no documentary evidence, I believe that the Brookhaven machine would have been a synchrocyclotron had they not known of the synchrotron idea, and I suspect this idea came to them more from Oliphant than from Veksler or from McMillan668.

667 Hibbard in Inall op. cit. p. 5. 668 Rolph (1995), p. 120. 430

Progress at Brookhaven

Whatever influence might have been exerted, the builders of the Cosmotron faced the same challenges as the team at Birmingham, though some of their solutions were different. The vacuum system was built of metal, laminated to minimize troublesome eddy currents, after ceramic tubes proved unable to withstand the pressure differential669. The tubes were made airtight by being wrapped in rubber sheeting. The challenge of the radiofrequency system was not as severe as in Birmingham; injection was to be at 4 MeV using a Van de Graaf accelerator and therefore requiring only a ten to one change in accelerating frequency. This could be achieved by the use of the newly developed ferrite materials670.

Arguably, the most serious issue facing the Cosmotron was that of “aperture”. How wide did the vacuum tube (and therefore the ring- shaped hole in the heart of the magnet) need to be in order that the circulating particles in the earliest stages of acceleration did not collide in unacceptable numbers with the walls of the tube and be lost? At Brookhaven, the issue had been tackled mathematically, by calculating the orbits of the particles under the influence of the various forces. Berkeley had taken a more pragmatic route, building a quarter-size scale model, in part to prove that such a machine would work, but also equipping it with adjustable shutters which could control the aperture. In trials, shutters would be slowly closed until the beam disappeared. By late 1949, testing indicated that the smallest acceptable aperture was larger than that calculated as acceptable by Brookhaven. While the Berkeley measurements were later shown to be unreliable, and the

669 In Birmingham, metal tubes were considered first and later replaced by ceramic pipes. 670 The RF tuning system in Birmingham was later rebuilt using such materials. It is not clear why they were not the first choice or at least considered earlier, though this may be an example of “first mover disadvantage”, as the materials had been only very recently developed. 431

aperture provided by the Cosmotron sufficient, it was, according to Blewett “a bad time at Brookhaven”, with suggestions that the machine design was fatally flawed and construction should be stopped

This occurrence points to an enduring issue with such large-scale projects, the importance of careful design. The outcomes of design choices are built into the machine, often in ways that can be difficult to change. As has been pointed out earlier, the Berkeley team had added to costs by making their machine more flexible and able to be modified. They had also used a quarter-scale model to refine design parameters. As we have noted, the Birmingham project suffered from bad design decisions and the nature of the construction made correction of the errors difficult and time-consuming, in some cases impossible.

The sense of competition between Berkeley and Brookhaven was significant. Each group was striving to be the first to produce a 1 GeV proton beam (as was Oliphant’s team, of course). Some at Brookhaven thought that the wealth of talent and resources at Berkeley would see them reach the target first. However, the entire Bevatron was transferred for two years to a top-secret project involving the production of fissile material for the national defence. As a result, Brookhaven was given the chance to be first.671

Oliphant’s optimistic timelines for completion of his machine were not unique to him, and may indeed be endemic to projects of this kind. An early report on progress on the Cosmotron, covering the six months to December 1948, indicated that the “conception stage” of the project had been completed in 1947, the target range of two to three BeV had been set early in 1948 as part of the “design stage” and the project was already moving into the third or “construction stage”. The magnet

671 Blewett (1987), p. 8. 432

foundations were complete and the steel on order, as was the power supply to be built by Westinghouse. According to forward plans, all major components would be in place by the end of 1949. “It is hoped that this culmination of effort [i.e. the first operation of the machine] would occur before the end of 1950672. “

As it turned out that first beam was possible only more than a year after that target date, largely (it appears) due to delays in constructing the magnet, which was two years late in completion. The magnet was finished some time in 1951 and testing started, the vacuum tube was in place by early 1952 and the beam sent circulating. Early attempts to accelerate the beam failed, reviving fears about the aperture, but such concerns proved unfounded. The Cosmotron first accelerated a beam above 1000 MeV (to 1.3 BeV) on 20 May, 1952, a little more than a year before the Birmingham machine achieved almost the same result. That was by far the highest energy attained by artificial accelerated particles to that date. The beam reached 2.3 BeV in June and the design energy of 3.3 BeV the following year.

As already noted, the Cosmotron may not have been first with a 1 BeV beam but for the two-year enforced delay on the Bevatron and, indeed, without the various delays suffered at Birmingham. In the event, all three machines came on line within a two year period, which more or less reflected the spread of their starting dates, ready to make a contribution to experimental nuclear physics.

We have referred to Birmingham, Brookhaven and Berkeley as homes to the first generation machines, as if there were no others. If we define “first generation” as “weak focusing”, there were another four such, at

672 Progress Report July 1- December 3 1948. Brookhaven National Laboratory. Associated Universities Inc., pp. 19-25. 433

locations in the UK (NIMROD at Harwell), France (Saclay), the Soviet Union (Dubna) and the USA (Princeton). The last of these, at Princeton, came into operation in 1964, some years after the first “strong focusing” synchrotrons began producing beams.

Personal assessments of “the man and the machine”

We are ready now for the last of our approaches to the challenge of assessing the Birmingham enterprise. Over the years, many people worked with Oliphant, either specifically on the synchrotron or more generally at Birmingham or the Cavendish. Many others knew of Oliphant’s work or had other associations with him. In the course of those associations, they built up impressions or opinions of his personality and style, and some were able to comment directly on the synchrotron. We have been able to gain access to some of those views, taken from letters, interviews or contributions to commemorative symposia, and they offer the potential of another approach to assessment.

There are of course hazards with such an approach. In most cases such assessments have been made several decades at least after the events recalled. One person’s assessment of another can be coloured by a range of circumstances. Even the willingness of people to openly assess one another can vary widely, especially if the assessment is not favourable. There is a temptation to try to distill a consensus view from such assessment, a collective judgment which most people would agree with in general terms, even if they differ over details. There can be value in such a distillation, even if some of the more significant insights arise from the divergences from that consensus.

434

As with other elements of this overall analysis of the Birmingham adventure, there is value in beginning with Oliphant’s own account. In 1993, unable through his increasing age673 to attend the 40-year Symposium, he wrote a letter which briefly expressed his assessment of the Birmingham enterprise. As quoted in the Proceedings, this read in part

In retrospect one can see how foolish it was to attempt to assemble a team as large and with [sic]674 access to the manufacturing facilities enjoyed by the Americans. During the war, we had been able to get things done with the existing methods of production, and by not sparing ourselves. Tired, and back again to the belief that near enough was good enough, we were slow and fumbled. And we never, with the synchrotron, got over the loss of the deep knowledge and commitment of John Gooden. It was a brave effort in a totally new field. Looking back, I am amazed we ever tried! At the same time I am proud of every member of the team who did such an amazing “string and sealing wax” job675.

Oliphant’s meaning is not uniformly clear through this quotation, but we can extract the following key points.

 The enterprise may be judged “brave” but it was also “foolish”. The basis for both of these judgments requires further examination.  The enterprise suffered not only from the lack of access to the level of industry support available in the USA but also from

673 Oliphant was at this time in his nineties. 674 This is presumably a typographical error, and should have read “without”, but the original letter is currently not available for confirmation. 675 Rolph (1995), p. 113. 435

what we might label “war weariness” and from the philosophy (perhaps a consequence of the war weariness) that “near enough was good enough”.  The loss of Gooden was a decisive factor, presumably implying that the project did not proceed as well as it might have done with him.  Even though this was a “totally new field”, it was tackled using the methods of “sealing wax and string”. This may be a reference to the fact that so much of the apparatus was built in-house and designed by physicists.  The enterprise, and particularly the team that made it possible, was a source of pride for Oliphant.

One of the purposes in pursuing other personal recollections is to see to what extent Oliphant’s own assessments are supported by others. For example, it was Burcham who quoted Oliphant’s letter in his closing address to the anniversary symposium, and he immediately took issue with the sentiments expressed. “I would like just to read a brief extract from his letter because it seems that he himself after nearly 50 years has had some little doubts about the synchrotron which I must confess I do not personally share”. Burcham recalled that even more than a decade after the synchrotron itself ceased operation, the Birmingham laboratories had maintained “a position of quite reasonable eminence in the field of particle physics676.”

That we still have this position is due both to Oliphant and to the many young men, able people, some present, who joined us since the beginning. So I do not go along entirely with Oliphant when he says that it was foolish to undertake this development. To me it has been an exciting and extremely worthwhile

676 This may be a reference to the reputation of the “visual methods” group. 436

undertaking. It has given us a sense of achievement, a sense of excitement and a sense of reality in the increase of knowledge and for this we must be grateful essentially to Mark Oliphant.

Burcham’s comments here raise two considerations. Firstly, like Oliphant’s, they were made several decades after the events being commented on, and in the light of the developments unable to be assessed at the time, or considerations, such as the later reputation of the laboratory, which were not thought important at the time the project was underway.

Secondly, they point up a significant difficulty for us, namely in seeking to separate “the machine from the man”. As will have become apparent, the Birmingham synchrotron was in a very real sense “Oliphant’s machine”, even if he did not see it through to completion and 50 years later was willing to contemplate the wisdom of undertaking it at all. He had conceived it, articulated its principles and its rationale, undertaken the fundamental design, secured the necessary political and financial support, and led and inspired the team of mostly young men who undertook the early stages of its construction. That the machine existed at all was almost entirely Oliphant’s doing, though of course many others contributed. It is difficult to conceive that anyone else could have initiated such an unproven project at such a time and place, war-weary Britain in the immediate postwar years.

That assessment implies that the project also reflected many of his own personal traits. Its positive attributes were largely his positive attributes, vision, boldness, a willingness to confront challenges head on. Its failings were largely his failings, an over-optimism that sometimes verged on lack of reality, an over-reliance on ingenuity and the “fire in the belly” and the parallel downplaying of the importance of planning,

437

the need to have others to winnow the wheat from the chaff of his extraordinarily fertile mind.

The above paragraph summarises, I believe, the collective assessment of the man and the project by those who worked with him or in other contexts knew him well. It is time to look in more detail at that evidence.

A more distant view

Having begun this exercise by recalling the views of Oliphant’s long-time (since the Cavendish) colleague Burcham, we need a note of warning. Many of those whose views are available worked closely with Oliphant, often over a number of years. It is reasonable to think that they would commonly associate the man with the project and therefore with the sense of community the project engendered. Their recollections (as they generally are) are likely to carry a patina of nostalgia. It may be best therefore to start with some whose associations were not so close.

A severe assessment of the project, and to some extent of the man, came from Berkeley physicist Luis Alvarez677. He bluntly described678 the Birmingham synchrotron as “an absolute disaster”, as “an unbelievably bad machine, which never produced any good physics and was known throughout the world as “the White Oliphant”679. He visited Birmingham to view the machine while Oliphant was still there680. He

677 Luiz Alvarez (1908-1998). American physicist, who spent most of his career at Berkeley. Co-winner of the Nobel Prize for Physics in 1968 for his work on bubble chambers. In the field of particle accelerators, his interest was in linear accelerators. For biographical memoir see http://books.nap.edu/html/biomems/lalvarez.pdf 678 This information and quotations are taken from an interview with Alvarez by in April 1980. Transcript in EBA. No recording was made of this interview. 679 He refined this statement to “White Oliphant No 1, to be precise”, implying that No 2 was the machine proposed to be built at the ANU, never completed and to which the “White Oliphant” epithet is more commonly given. 680 This may have been during the 1948 Birmingham conference on nuclear physics, at which time the machine was still in the early stages of construction. 438

recalled saying to himself “what a miserable machine”, noting that there was no room to fit experimental apparatus. Alvarez would of course have had as the basis of comparison the much larger Bevatron then being constructed at Berkeley in its own especially capacious building.

What, in Alvarez’ view, was the reason for such a debacle? Oliphant, he maintained, did not welcome criticism. He did not listen enough to younger colleagues, whose input should be taken seriously, since they “offer you the criticism that only young people can”. Alvarez’ assessment seemed confused. He maintained that “you have to plant your flag in the field and say ‘Rally round, boys, we are going to make it out there. You have to build something big and it’s got to be good’ ”. That would seem a good summation of Oliphant’s ambition. Alvarez was even willing to allow for an occasional failure; in his view Oliphant had too many.

Could such a fierce critique have resulted from some personal antagonism? Not if some comments are taken at face value. “I like and respect Mark as a person. I think he is a fine gentleman and I always enjoy being with him. Despite that, he just didn’t ever make it in the big time in physics. I think that he had the potential but it just didn’t materialise.” There is however some important history to the relationship. Alvarez recalled coming as a fresh post-doctoral student to work at Berkeley with Ernest Lawrence in 1936, claiming that it was

….probably the greatest center of modern nuclear research in the world. People were kind of embarrassed that we missed out on some really big ones, and everyone was aware that Lord Rutherford and others at the Cavendish looked down upon us as a bunch of upstarts who didn’t know any nuclear physics and happened to be good engineers and had this fine apparatus

439

called a cyclotron. There was a sense of insecurity about the place.

Among the “really big ones” missed by Berkeley would have been the understanding of the reactions between deuterons. As we have previously noted, the correct interpretation of that was formulated by the Cavendish, with Oliphant playing a leading role.

The views of Norman Ramsey681 represented a more moderate perspective682. As President of the Universities Research Corporation which oversaw the work at Brookhaven, he was arguably in a good position to judge contributions to the field. He was closer to Oliphant than was Alvarez, and acknowledged a great personal admiration. Their acquaintance stretched back to 1935; as a student at the Cavendish Ramsey took Oliphant’s class on electricity in gases. He regarded Oliphant as a careful and methodical lecturer, one of the better ones, and able through his own enthusiasm to engender enthusiasm in his students. Further interaction with Oliphant, at a greater distance, came during the war when Ramsey was engaged on radar research683, further developing the magnetron which had been invented in Birmingham.

Ramsey recalled Oliphant’s enthusiasm in the immediate post-war years (1945/6) for the synchrotron he planned to build. “To my knowledge he was the first strong advocate of the [notion of using the synchrotron principle to accelerate protons]. He deserves credit for that, even though he wasn’t the first to reach the target. ….Most of the American machine

681 Norman Ramsey (1915- 1911), American physicist. Most of his career was at Harvard. Co- winner of the Nobel Prize for Physics in 1989 for technology of value to the operation of atomic clocks. For background see http://www.nobelprize.org/nobel_prizes/physics/laureates/1989/ramsey-autobio.html 682 This material is taken from an interview with Ramsey by Cockburn. The transcript is undated but the interview was undertaken c1980. EBA. 683 Ramsey was initially (1940/41) working with II Rabi on the development of magnetrons to work at 3 cm. 440

builders thought his ideas were silly, but they changed their minds within a year. In the light of subsequent history, they were not silly”.

Ramsey acknowledged that Oliphant’s courage to go ahead with the project was ”great”, but, though he does not say so explicitly, clearly thought it fell well short of expectations, since he gives any number of reasons for such a failure. Oliphant had “bad luck and perhaps bad skill to a certain extent”. He “wanted to do it all on a shoe-string”. The project was “too big to be undertaken virtually alone by graduate students. He [Oliphant] was still in the grip of the Cavendish system and the pre-war Berkeley system.”

His accelerator at Birmingham had tremendous pluses and minuses. His concept -- the basic idea of omitting in the middle of the ring -- was very important. It was really a major step forward in accelerators. On the other hand, the project was partially a failure in that other machines came along and soon went past him.

Ramsey, in short, thought the venture was (to use Oliphant’s own terminology quoted earlier) “brave” but “foolish” (in terms of the paucity of resources). Oliphant was still in the grip of the “Cavendish system. Oliphant had “bad luck” (in which we might include the loss of Gooden, though Ramsey does not mention that). We can note how closely Ramsey’s judgment paralleled Oliphant’s own, quoted earlier.

The views of the team

While keeping in mind the word of caution given above, we cannot fail to take account of the assessments of the man and the machine given by those who worked in the team. They are not the words of one-eyed

441

sycophants, but rather of men well able to judge the pluses and minuses of Oliphant’s way of working, and their impacts on the machine over which they laboured together.

David Caro684 was only a junior, and rather short-term, participant in the project, being in Birmingham from September 1949 to January 1952. As he saw it, Oliphant’s emphasis was on speed and ingenious improvisation with as little detailed planning as possible. This was probably the only way such a machine could have been built in the time, but it relied much on his co-workers to be effective.

I found Oliphant a fascinating person to work for. I’ve never known anyone who could produce more ideas per week than he did in 1949/50. It was common for there to be meetings every few days in which he would produce a new set of suggestions. Many of these, perhaps most of them, were unworkable but some were good and his rate of production of useful ideas was exceptional. In Birmingham people like Riddiford, Symonds and especially Hibbard were very good at standing up to Oliphant and eliminating the bad ideas rapidly. That made his method of operating possible and exciting.685

Colin Ramm686 had arrived in mid 1947, staying until October 1954, and then to moving to CERN where he stayed for nearly 20 years. He had arrived as a young M.Sc. from Australia expecting to see the machine well advanced, and almost ready for use, but instead found “only a small packet of magnet plates standing forlornly on the concrete base…. I hadn’t yet understood that sometimes constructions of the future were

684 Caro to Lawson, 6 December 1995. Moon Papers B30, UBSC. 685 Caro continued his comments as follows: “The situation at the Australian National University was quite different. He did not have the people there who could separate the good ideas from the bad and the consequences were very sad”. 686 Ramm to Lawson, 9 March 1995. Moon Papers B30. UBSC. 442

described in the present tense, and my friends who had heard Oliphant’s talks in Australia had not understood the timescale when they told me of his project”. Yet his assessment of the enterprise was whole- heartedly positive. “Given the background of manpower and funds available for the project, the successful operation of the synchrotron was an unequivocal triumph of the whole department. Even Faraday could not have done it with less!”687

A later assessment of the machine came from Kinson who, having arrived only in 1953, had no direct involvement in its construction or contact with Oliphant. He reported to the anniversary symposium688

My main memory of working at the Birmingham synchrotron is that it was fun. There were small numbers of people, and the atmosphere was friendly. The budgets were small, and lots of ingenuity was required. However the physics were serious and played an important part in the development of particle physics.

Of Oliphant’s co-workers, Hibbard gave the most extensive assessment689. Its tone was generally upbeat. To choose some sentences: “The proposal to make the machine was courageous and commendable… the Birmingham synchrotron was the high point of Oliphant's career…The fact that the Birmingham machine reached its top energy two months after the Brookhaven Cosmotron was not a reason for shame. We never once considered it as a race”. Then the critique; “To make it on a shoestring with a batch of students as a spearhead was a romantic concept. It was also the end of an era.

687 This is a reference to a comment that Oliphant reportedly made from time to time. Commenting that “Faraday did it on less” focused attention on the need to conserve resources. Oliphant was a great admirer of the nineteenth century English physicist Michael Faraday and planned to use his concept of a homo-polar generator as the power source for his accelerator to be built in Canberra. 688 Rolph (1995), p. 84. 689Hibbard in Inall op. cit.. 443

Further progress in this field had to lie with more professional and cooperative action."

The delays that delivered silver instead of gold had explanations; the disparity in resources that more than made up for the earlier start, the departure of Oliphant (“giving in to moral pressure and his conscience”), the death of Gooden, faults in the motor generator and the magnet, in the fixing of the pole tip plates and winding (“explainable as a consequence of lack of experience in pulsed magnet design”). Only one error was not repairable, namely a failure to account for saturation limited the maximum achievable energy.

The only way Oliphant might be held responsible for some of its failings lay in his constant exhortation to “stick your neck out” or get some fire in your belly”, translated as “trust your instincts”. It might be argued that a premature decision, which led to the ordering of the magnet plates, was a consequence of one neck stuck out too far. Risks are not risks if none of them fail. Oliphant never hesitated to rail at what he believed was excessive caution. However his bark was much worse than his bite and I know of no case of him taking action against anyone for these failings690.

Summing up

The ten pages devoted to the Birmingham machine in Lawson’s overview of British synchrotrons691 constitute the most comprehensive assessment of the Birmingham project before the present exercise. In his comments we can see the influence of his correspondence with Oliphant’s co-workers. Lawson contrasted the style of working in

690 Ibid. 691 Lawson (1997), pp 24-35. 444

Birmingham with “the more conventional and thorough approach to the Cosmotron, with organised engineering support [as being] more likely to be successful in reaching its targets on time”. Yet he pointed out that it would hardly have been possible for Oliphant to set up such a costly organisation in post-war Britain in a university setting.

Of the machine he commented “The enterprise can be seen as a bold and courageous attempt to be first with a one GeV machine”. Of the man: “Though at times irritatingly stubborn, Oliphant was an inspiring leader, with a great faith both in the ‘fire in the belly’ as a receipt for getting things done quickly, and in the rapid emergence of good ideas to circumvent difficulties as they are encountered”. Echoing the thought of others, Lawson observed that Oliphant was fortunate to have colleagues, perhaps most notably the long-serving Hibbard, “able to select from his flood of ideas those which were worth pursuing and strong-minded enough to firmly reject others”. As for the influence of the shift to Moon’s leadership after Oliphant's departure, this was, in his opinion, mixed. Moon had not participated in the detailed design and was not enthusiastic about Oliphant’s style, but once the responsibility was his he “provided continuous encouragement and gave high priority to providing resources and support”.

Of the broader impact of the project, the evidence examined in this chapter enables us to draw certain conclusions. In terms of the technology, there is no doubt that the Birmingham synchrotron performed mostly up to expectations, despite errors in design and construction and suffering "early mover disadvantage". The technological outcome demonstrated that Oliphant's vision was sound, and that the synchrotron was based on sound physical principles. Combined with the successes of the Bevatron and the Cosmotron, the Birmingham enterprise pointed the way to the future in terms of the

445

technology of particle accelerators and their capacity to advance understanding in nuclear physics.

The scientific contribution of the Birmingham machine was limited by the maximum beam energy achievable and the relatively low beam intensity. This excluded Birmingham from some of the emerging fields of research, such as the study of nuclear resonances. On the other hand, the work that it was equipped to do, such as the elucidation of nucleon- nucleon interactions, was valuable. It appears there were sufficient scientific questions lying within the capacity of the machine for it to be in substantial demand. As a result, the synchrotron provided training opportunities for a substantial number of nuclear physicists and made a not-insignificant contribution to the national capacity in this field. It also served as the catalyst for the development of an important centre of skill in Birmingham in the analysis of the records of nuclear interactions, able to be used to analyse the output of a number of other nuclear physics research centres, and resulting in a significant legacy to the University which outlasted the closure of the synchrotron by a number of years.

Perhaps of more importance, and insufficiently recognised up to the present time, was the impact of the Birmingham project beyond its own university walls. We have demonstrated the strong influence which the Birmingham endeavour exerted on the genesis of the parallel project in Berkeley, and from there on the origins of the Brookhaven enterprise. We find very little acknowledgement of such a debt in the commonly- accepted chronology of development in this field.

446

CHAPTER NINE Was the Birmingham enterprise Big Science?

In our documentation of the genesis of the Birmingham proton synchrotron, we have sought to place the enterprise into a number of key contexts. These include the development of the relevant technology as it existed at the time of Oliphant's initiative, the course of Oliphant's own career and the growth of his “capital” and the immediate surroundings of world events, particularly those associated with the development of atomic energy. Another, but rather different, context for assessing the importance and the impact of the Birmingham machine arises from the ongoing discussion of the phenomenon commonly dubbed Big Science692. Experimental nuclear physics is often cited as a prime example of this trend, usually seen as emerging in the years immediately post-World War II, and characterised by very large costs, establishments, hardware and profile.

The question to be addressed here is; what place, if any, does the Birmingham proton synchrotron have in that development? If the Birmingham enterprise fits the definition of Big Science, it may be that useful insights can be gained by a consideration of such broader trends.

The first analysis of Big Science

The earliest recorded discussion of the issue of Big Science was presented in 1961 by Alvin Weinberg, then the Director of the Oak Ridge National Laboratory, itself one of the legacies of the Manhattan Project.

692 See footnote #4 for an explanation of the proposed usage of this term in the following discussion. In quotations, the author’s preferred usage will be employed. 447

In her book Before Big Science693, Nye claimed the commentary was “influential” and that Weinberg had coined the phrase Big Science, writing it with initial capitals. It is therefore the appropriate place to start.

Weinberg’s article, published in Science under the title The Impact of Large Scale Science on the United States694, carried the following summary under the title: “Big Science is here to stay, but we have yet to make the hard financial and educational choices it imposes”. Weinberg reiterated and expanded on this point in the body of the piece.

…. it is fruitless to wring one’s hands over the bad effects of Big Science. Big Science is an inevitable state in the development of science and, for better or for worse, it is here to stay. What we must do is to learn to live with Big Science. We must make Big Science flourish without, at the same time, allowing it to trample on Little Science. That is, we must nurture small-scale excellence as carefully as we lavish gifts on large-scale spectaculars695.

The choice of the term “spectaculars” was typical of the tone of Weinberg’s analysis. In similar vein he labeled the growth of the ranks of the administrators to oversee the spending of public money on such large-scale science as “administritis”, and those participants who seek to inform the voters and tax-payers about the nature and purpose of their Big Science projects as “publicists”.

On the latter point he added

693 Nye (1996). 694 Weinberg (1961), p. 161.

695 Ibid. 448

… since Big Science needs great public support, it thrives on publicity. The inevitable result is the injection of a journalistic flavour into Big Science which is fundamentally in conflict with the scientific method. If the serious writings about Big Science were carefully separated from the journalistic writings, little harm would be done. But they are not so separated. Issues of scientific and technical merit tend to get argued in the popular, not the scientific, press, or in the congressional committee room rather than in the technical-society lecture hall; the spectacular rather than the perceptive becomes the scientific standard. When these trends are added to the enormous proliferation of scientific writing, which remains largely un-read in its original form and must be predigested, one cannot escape the conclusion that the line between journalism and science has become blurred696.

From the perspective of today, there was a disturbingly “elitist” tone in these comments. The implication is that discussion and decision-making about Big Science is best left to those with detailed understanding of the science. Yet Weinberg must have known such a constraint was impossible. If, as he maintained, the rise of Big Science was inevitable, then equally inevitable was the engagement of government as the immediate source of funds and of the tax-payer as the ultimate source.

The challenge for the proponents of Big Science projects was surely to couch their plans in terms both intelligible and attractive to the public, the bureaucracy and the executive, the sort of “publicist’ activity he derides. As we have seen in the case of Oliphant, the success of many Big Science endeavours depended on a style of leadership that drew together skills in science, technology and salesmanship.

696 Ibid. 449

The “monuments of the past”

Weinberg opened his article by comparing the new enterprises of Big Science with grand monuments of the past, the pyramids of Ancient Egypt, lavish royal palaces, medieval cathedrals. Of these he says

Throughout history, societies have expressed their aspirations in large-scale monumental enterprises which, though not necessary for the survival of the societies, have taxed them to their physical and intellectual limits. History often views the monuments as symbolising the societies697.

He went on

When History looks at the 20th century, she will find science and technology as its theme; she will find the monuments to big science …the huge rockets, the high-energy accelerators, the high-flux research reactors….symbols of our time as surely as she finds Notre Dame a symbol of the Middle Ages. She might even see analogies between our motivations for building these tools of giant science and the motivations of the church builders and the pyramid builders. We build our monuments in the name of scientific truth; they built theirs in the name of religious truth; we use Big Science to add to our country’s prestige, they used their churches for their cities’ prestige; we build to placate what ex-President Eisenhower suggested could become a dominant scientific caste, they built to please the priests of Isis and Osiris698.

697 Ibid. 698 Ibid. 450

What did Weinberg mean by Big Science? His definition, though never clearly stated, appeared to hinge on large scale hardware and big budgets. Throughout the article he referred several times to three examples, manned space flight, high-energy physics and (less often) nuclear physics. The first two inevitably involve large investments and large pieces of equipment, the third commonly does, though not always. On the other hand he does not refer to astronomy, the earliest of the ‘big sciences’, with the quest of ever-larger mirrors to see further and in more detail dating back more than 150 years to William Herschel. Yet the then-largest telescope in the world, the 5 metre Hale Telescope at Mt Palomar, had seen first light in the immediate post-World War II years, around the same time as the first proton synchrotrons and research reactors.

The impacts of Big Science

So what are the hazards that Weinberg saw being posed by Big Science? He drew a little on arguments advanced by the British astronomer Fred Hoyle, who was opposing any large-scale involvement by the UK in space research. As Weinberg quoted him, Hoyle was making two points: that the intrinsic scientific interest of space research is not worth the money and manpower that goes into it, and certainly does not justify spending any more on it than goes into any other branch of science, and secondly, that whenever science is fed too much money, it becomes “fat and lazy”. “He claims to see evidence that the tight intellectual discipline necessary for science is, especially in America, being loosened”699.

Weinberg tended to agree with Hoyle about the merits of investment in space, though he spoke mostly of manned space flight. Given the need

699 Ibid. 451

to protect astronauts against the (then still-uncertain) hazards of space flight, he judged the likely needed expenditure as too unsure to be seriously contemplated. These judgments now seem premature. As for the first point, it would have been difficult for Hoyle to foresee that investment in space research would in the long run have an immense payoff for astronomy, with the placing into orbit of a wide range of observatories, including the iconic Hubble Space Telescope and the sending of information and image-gathering space probes all over the solar system. There was a big payoff for the rest of the human race as well. The social return from the capacity to place satellites in orbit for communication, weather forecasting and environmental monitoring has been great, as is the psychological/ inspirational dimension of space enterprise; (that is, statements of the kind “we can put a man on the Moon so why can’t we ……”)

Weinberg parlayed Hoyle’s second point into a more general question. Is Big Science ruining science? In fact he sought to answer a less- strident question; is Big Science likely to ruin science, including Little Science? We have already seen his thoughts on the malign influence of the “publicists” that Big Science projects seem to need. His second concern is for the researchers themselves.

…One sees evidence of scientists spending money rather than thought. This is one of the most insidious effects of large-scale support for science. In the past, two commodities, money and thought, have been hard to come by. Now that money is relatively plentiful but thought is still scarce, there is a natural rush to spend dollars rather than thought, to order a $107 reactor instead of devising a crucial experiment with the reactors at hand or to make additional large-scale computations instead of reducing the problem to tractable dimensions by perceptive physical

452

approximations. The line between spending money and spending thought seems to be blurring700.

This was Hoyle’s “fat and lazy” comment restated. The epigram attributed to Lord Rutherford comes to mind: “We haven’t got the money so we’ve got to think”.701

Weinberg brought forward no evidence that this trend was an increasingly common one, one that might be attributed to the growth of Big Science, though the references in the quotation suggest some cases might have arisen in his own experience at Oak Ridge. On the contrary, it could be argued that even in relatively bountiful times, no-one can get all they want. In most projects corners have to be cut and there is unlikely to be any money to waste on the unnecessary. Hoyle’s comment may have arisen from the frustration of seeing resources directed to space research (of its nature necessarily more expensive) and away from the more traditional astronomy in which he was engaged.

Weinberg’s third concern is the apparently inevitable growth in administrators, there to see that large sums of public money are spent wisely. “The big scientific community tends to acquire more bosses. The Indians with their bellies to the bench are hard to discern for all the chiefs with bellies to mahogany desks. Unfortunately science dominated by administrators is science understood by administrators, and such science quickly becomes attenuated, if not meaningless702.”

This seems rather overstated; some increase in the complexity of administration must accompany the spending of public money for the

700 Ibid p. 162. 701 Quoted by Jones (1962) p. 102. 702 Weinberg (1961) p. 162. 453

government is accountable. In many cases of course, the administrators will come from the ranks of the researchers, as Weinberg did himself.

Weinberg devoted a good space to the threat he saw Big Science posing to the universities, particularly to the extra demands which would be placed on senior staff.

A professor of science is chosen because he is extremely well qualified as a scientist, as a thinker or as a teacher. If he becomes too involved in Big Science, he will have to become a publicist, if not a journalist, an administrator and a spender of big money….Once Big Science invades his precincts and he becomes an operator (even though a very effective one) his students and his intellectual eminence and proficiency are bound to suffer... I do believe that Big Science can ruin our universities by diverting the universities from their primary purpose and by converting university professors into administrators, housekeepers and publicists703.

Weinberg clearly intended these comments as a warning, but they can be taken to indicate something more profound. The nature of a Big Science enterprise demands a new type of scientific leadership. To the university professor’s standard CV of scientific insight, experimental skill and the capacity to teach must be added administrative skills of a high order, the ability to draw together and lead a large and diverse team, and a personality, indeed a charisma, able to inform and inspire colleagues, the general public and a wide range of stakeholders, including potential funders. This confluence of capacities is rare. It will be argued that Oliphant, among a small number of others, possessed it

703 Ibid p. 162. 454

and for that reason, among others, he became a leading player in the early days of Big Science.

Making choices about R&D spending

Weinberg has another concern, though it is not explicitly on his list. To him, Big Science endeavours like manned space flight and high-energy physics, as well as being “wonderfully expensive”, have “remoteness from human affairs”. He would rather put scientific resources into issues “which have more bearing on the world that is part of man’s everyday environment, and more bearing on man’s welfare”, areas like molecular biology, the problem of water, atmospheric pollution or chemical contamination of the biosphere. “Each of these is a technical issue which has a claim on our resources, a claim that will have to be heard when we make choices.”

Nye quoted Daniel Kevles that in this matter Weinberg practiced what he preached. “His own management of the Oak Ridge facility was one that directed some of the laboratory’s research effort to such socially useful concerns as cheap energy sources, desalination and environmental problems.”

Weinberg urged that such choices be matters of high national policy and the subject of extensive public debate. “We cannot allow our overall science strategy, when it involves such large sums, to be settled by default, or to be pre-empted by the group with the most skilful publicity department.” There was a final cautionary note. In his view, we needed to remember the experiences of other civilizations. At times much effort, talent and treasure have been spent on “monuments” which did not really advance human welfare and the societies that built them often soon came undone. “History tells us that the French Revolution was the

455

bitter fruit of Versailles, and that the Roman Coliseum helped not at all in staving off the barbarians. So it is up to us to learn well these lessons of history; we must not allow ourselves, by shortsightedly seeking after fragile monuments of Big Science, to be diverted from our real purpose, which is the enriching a broadening of human life”704.

Derek de Solla Price on Big and Little Science

A year after Weinberg’s seminal writings, physicist turned science historian Derek de Solla Price gave the Pegram Lectures at the Brookhaven National Laboratory. These were published soon after under the title Little Science, Big Science705 While Price adopted the same terminology as Weinberg, his focus is quite different. He was concerned not with the apparently disproportionate growth of certain areas of science but with the burgeoning of the enterprise of science in total. On various measures, the scientific enterprise had grown exponentially in the decades before his analysis, with the volume of scientific publications, scientific manpower and expenditure on research and development doubling in as little as ten or twenty years. Indeed some of these trends could be traced back three centuries representing a million-fold growth.

Projecting some of these trends forward, he maintained, would see the total expenditure on scientific research and development exceed the total national budget or even the gross domestic product in a foreseeable time. In this, he echoed some of Weinberg’s concerns. Clearly such rates of growth could not be sustained, and Price spent time analysing the countervailing trends which could slow the rate of

704 Ibid p. 164. 705 de Solla Price (1963). 456

growth and convert (and indeed did convert) an exponential growth curve into an “S-shaped” one (a “logistic one” in his terminology).

But even the growth so far had produced a form of science that was almost unrecognizable.

Because the science we have now vastly exceeds all that has gone before, we have obviously entered a new age that has been swept clear of all but the basic traditions of the old. Not only are the manifestations of modern scientific hardware so monumental that they have been usefully compared with the pyramids of Egypt and the great cathedrals of medieval Europe but the national expenditure of manpower and money on it have suddenly made science a major segment of the national economy. The large-scale character of modern science, new and shining and all-powerful, is so apparent that the happy term “Big Science” has been coined to describe it. Big Science is so new that many of us can remember its beginnings. Big Science is so large that many of us begin to worry about the sheer mass of the monster we have created. Big Science is so different from the former state of affairs that we can look back, perhaps, nostalgically, at the Little Science that was once our way of life706.

This was obviously a sweep with a broad brush, and Price was quick to moderate his tone. He claimed we need to appreciate the nature of the transition from Little Science to Big Science, while being aware of the danger of dramatizing it. He questioned the perceived suddenness of the transition. Despite obvious examples such the Manhattan Project, Cape Canaveral rocketry, the discovery of penicillin and the invention of radar and electronic computers “The transition from Little Science to Big

706 Ibid p. 2. 457

Science was less dramatic and more gradual then appears at first. For one thing, it is clear that Little Science contains many elements of the grandiose. And, away in some academic corner, modern Big Science probably contains shoestring operations by unknown pioneers who are starting lines of research which will be of decisive interest by 1975 [i.e. in a few years]”707. .

Here Price was touching on two matters to which later commentators would return. The two categories of science are not wholly distinct but can interpenetrate, each able to display some of the characteristics of the other. And the transition from the little to the big is by no means as definite and dramatic as others, notably Weinberg, had claimed. Nor was Big Science a purely modern phenomenon. “Historically there have been numerous big national efforts: the great observatories of Ulugh Beg in Samarkand in the fifteenth century, of Tycho Brahe on his island of Hven in the sixteenth century and of Jai Singh in India in the seventeenth century, each of which absorbed sensibly large fractions of the available resources of their nations. As for international efforts, there were the gigantic expeditions of the eighteenth century to observe the Transits of Venus708.”

To the list could be added any number of other endeavours, large in scope and impact; the “Great Survey” organised by France in the 1730s to determine the true shape of the Earth, the series of major ship-borne expeditions of the 18th and 19th centuries which combined mapping with biological and other scientific studies (The Beagle with Darwin aboard, James Ross’ quest for the South Magnetic Pole, the Challenger voyages to name only those originating in Britain).

707 Ibid p. 3. 708 Ibid p. 4. 458

By their insights, Weinberg and Price in their different ways had set a framework for debate of the nature and significance of Big Science. Several decades would pass before significant new insights emerged.

Galison and Hevly

In 1992, thirty years after the original musings of Weinberg and Price, and by which time the trends they had seen emerging were much more evident, Peter Galison and Bruce Hevly709 edited a book of essays which examined a broad range of Big Science endeavours and their common features. The contributors were drawn from a group which had attended a conference on the topic at Stanford in 1988. In his introduction to Big Science; the Growth of Large Scale Research710, Galison made no reference to the writings of Weinberg and Price (though some of the other contributors did), other than the Big Science label itself, usually without the capital letters. Perhaps he thought their insights were no longer relevant, having been overtaken by three decades of events.

Galison devoted his preface to summarising the contributions to follow, which were grouped under three headings; The Big Physics of Small Particles, Sponsored Research and External Interests and Big Science and National Security. The first section, which includes analysis of the development of large-scale research facilities at Berkeley under Lawrence, Stanford (an over-arching study done by Galison and Hevly themselves and a more targeted one dealing with the large linear

692 Galison and Hevly (1992). 710 Some reviewers commented that its title should have been perhaps “Big Physics”, since most of the projects considered were in the areas of physical sciences, astronomy and space science; no studies were presented dealing with biomedical science and pharmaceuticals, earth science or agriculture. However, this may merely reflect the fact that it is those areas that the trends and issues the book examines were first evident. From the perspective of framing Oliphant’s work in particle physics, the narrowness of the choice is no impediment.

459

accelerator there), CERN and KEK in Japan, is of most relevance to the Oliphant enterprise.

From the papers in the first section, we learned, inter alia, that

o Big Science in California began long before the increased defence-related outlays of World War Two, with an involvement from the early 1920s in finding solutions to problems in the burgeoning electric power industry. o A case can be made against the “inevitability” ascribed by Weinberg to the transition to Big Science: in the specific case of CERN, such a multinational development required a particular confluence of circumstances. In particular, it did not grow out of a long tradition of national scientific or scientific-military concerns. o European particle physicists did not have the same close relationship with engineers that was displayed in the USA. o Questions can be asked about the impact on the careers and reputations of young physicists of working in the large teams typical of Big Science, due, inter alia, to the mismatch between the long time frames of the machine building and the much shorter periods in which theses are completed or tenure is typically held.

Each of these issues finds a resonance in the Oliphant project and in our analysis of it. As set out earlier, “big physics” for Oliphant too had its origins in the early 1930s. The particular enterprise was not “inevitable” but required a particular mix of circumstances of time, place and personality. Oliphant’s relationship with his engineers was more typical of that common in Europe rather than in the USA, despite his wartime experience in the latter country. We have already noted the two-way

460

impact on the Birmingham enterprise of the mismatch of the time-frames of completing a PhD thesis and completing the machine.

Galison’s preface also yielded some broader perspectives. A key one concerns the linkage of a Big Science project to concerns within broader society: “… the change in the scale of science has required scientists to align their activities with the broader elements of the society. Whether big science derives its support from designing patents, working for industry, contracting to the federal government or building weapons, the implications for the wider society cannot be ignored711.”

We see this characteristic in the bid to build the Birmingham accelerator. In framing his requests for funding, Oliphant placed great emphasis on the potential for research done at the much higher energies to underpin new technologies for releasing nuclear energy, so making it cheaper and more abundant. As the proposal progressed to higher levels of government, that rationale became over-shadowed by a second, the use of such equipment to train the coming generation of nuclear physicists and engineers, vital if the UK was to build its own civilian and military nuclear power programs.

More clarity on the meaning of Big Science

In the wide variety of essays that make up the Galison and Hevly book, the authors explored a range of projects that appear to fall within the scope of Big Science. We can here note only a few of the insights presented.

711 Ibid p. 2. 461

In her discussion of the development of high-energy physics in Japan, Traweek712 quoted Lew Kowarski713, whom she described as a key figure in the development of CERN (which surely must count as Big Science). As summarised by Traweek, Kowarski made the following key points:

o Big Science was born of wartime emergency, and thrives on national and multinational prestige, interdisciplinary efficiency, military-like hierarchies, autocratic leaders, committees and money. o All Big Science is successful for the same reasons: sharply defined and ambitious objectives, “mixing scientific competence with administrative competence”, “participation of respected and strong personalities” and political realism. o All Big Science laboratories have the same problems: dwindling numbers of “pure scientists compared with support staff (technicians, clerks, administrators, librarians), few “sacrificially- minded people” (since this work is now well funded), caste- stratified teams with no recognised intellectual “master”, a preference for leaders who “play safe”.

All this recalls Weinberg’s troika of concerns about Big Science; too many publications, too much money, too much administration.

Traweek also reports Kowarski’s concern with the connection between Big Science and public opinion. Having noted studies of the ability of some scientists to “seduce an audience”, she wrote “the role of this awed audience for science is not to judge the value of the projects of scientists and engineers; its function is to approve, fund and provide

712 Traweek (1992) p. 100. 713 Kowarski (1977). 462

recruits, and, as Kowarski said, it is essential for Big Science714.” Weinberg, of course, said much the same, and with a similar tone of concern.

Kargon, Leslie and Schonberger were among those who added to the definition of Big Science, in their case by emphasising the importance of organisational structure and external linkages.

Big Science is not just a matter of scale, and of the money needed to support that scale of effort. Big Science involves a pattern of organisation that informs the scientific endeavour and its relation to the wider society. Big Science is marked, as is well known, by the hierarchically organised team….This team is plugged into a facility…with its own hierarchically-organised structure. This facility, in turn, connects to the wider society, nowadays in the United States through an umbilicus to the federal government715.”

On one level this is a statement about a process for accountability for the spending of large amounts of public money. On a second, it relates to the way the goals of the enterprise transcend the purely scientific and find justification through their linkages with goals beyond science and beyond the enterprise. This is a recurring theme in many commentaries. On yet another it deals with the need for a more complex managerial organisation, even a “command structure”, as is found in all large enterprises to enhance efficient operation and achievement of goals.

714 Ibid. p 102. 715 Kargon (1992), p. 335. 463

Big and Little Science: insights from astronomy

A further important insight into the troublesome division between Little Science and Big Science came from Robert Smith in his account of the development of the Hubble Space Telescope. He starts with the obvious issue of cost and quotes Price

…. as Derek Price, one of the first to investigate the nature of modern big science, pointed out over 25 years ago, “Without doubt, the most abnormal thing in this age of Big Science is money. The finances of science seem highly irregular and …they dominate most of the social and political implications”. Most obviously, if large sums of money are not forthcoming, there can be no Big Science. And without $2 billion, there could have been no Hubble Space Telescope, probably the most costly scientific instrument ever built716

But do the size of the hardware and the size of the budget define what Big Science is and what it is not? If so, astronomy has been “big” for centuries, at least since William Herschel began his “quest for aperture” in the late 18th century, or even earlier, when Huygens and others began to build refracting telescopes with very long focal lengths to try to overcome the defects of the lenses then in use. Herschel would not have been able to build his big telescopes without special financial support from the king.

As Smith pointed out, even the largest telescope could be used by only one observer at a time. He quoted George Ellery Hale, who asserted that even astronomers using the then world’s-largest 100 inch instrument at Mt Wilson “ploughed a lone furrow”, often personally

716 Smith (1992) p. 186. 464

undertaking every step in the process of plate preparation, exposure, development and interpretation, in the tradition of “little science”. Here we have an example of the kind Price recognised, with big and little science mingled in practice.

Given such examples, Smith concluded

The common equation of big science…that big bucks plus a big machine equal’s big science, is therefore flawed. We would, moreover, do well to employ the notion of little science carefully. For example, among the characteristics that have been identified in big science are politicization, bureaucratization, high risk and loss of autonomy. Often the implication is that these characteristics are absent from little science, an implication not always borne out by the historical evidence. In addition, a periodisation is often implicit in the use of little science, that big science emerged from World War II and that before then the scientific enterprise consisted entirely of little science. … In talking about both big science and little science, … it is important to realize that they are both broad descriptions, not explanations, and that they have been used somewhat ambiguously in the past, Examinations that take account of both cost and scientific practice are likely to be more historically rewarding than those based on cost alone.717

The astronomy example is worth further study, as it may lead to a further refinement of our understanding of the nature of Big Science. As with much of science, astronomy has been driven at least in part by the development of new technology of data collection and analysis which has, inter alia, greatly increased the productivity of telescopes, so

717 Ibid p. 186. 465

allowing many researchers to use the same instrument, if not simultaneously. Among the first of these, in the mid 19th century, was photography, which removed the need for observers to make time- consuming sketches at the eye-piece, so permitting many more observations in the course of an evening’s viewing.

Another influential factor would have been the movement of large instruments from private to public ownership. It is not clear that anyone other than the owner/builders (or perhaps their close friends) was able to use Herschel’s “40 foot” or the Earl of Rosse’s “Leviathan of Parsonstown”. But by early in the 20th Century the biggest telescopes were publicly owned and funded, and researchers could apply to use them. Productivity and impact rose again.

Further advances in technology, such as the use of computers and electronic sensors to guide and control telescopes and to gather and analyse data, so enhanced their productivity that they could support whole communities of researchers, some of whom would control the telescope remotely or rely on support personnel to gather the data they would interpret. The most striking examples here are the space-based telescopes, such as the Hubble or the Spitzer. Yet from Huygens to the Hubble, these instruments were always in the vanguard in terms of size, cost and technology. They were always entitled to be called “big”, at least in relative terms. Does this point to another, perhaps critical, dimension of Big Science as we understand it today, its capacity to generate a greatly enhanced flow of data? Is not productivity a key characteristic of Big Science?

The task of summing up the wealth of material in Big Science fell to Hevly. In his afterword, Reflections on Big Science and Big History, he wrote

466

What constitutes “bigness” then if not size alone? Big science touches many areas beyond the boundaries of science narrowly defined; that is its essential quality. It can be understood only by integrating those areas into a coherent picture, a requirement that forces historians to take seriously the fundamental task of our craft, to place the past in it proper context. Again the work presented here suggests that the changes creating big science are not simply changes of scale. While scientific machines surely increased in size and power, as did the groups operating them, and objects of research were perceived through ever more precise systems of measurement, data collection and analysis, new forms of institutional, political and social organisation arose.

In seeking to summarise some of the defining characteristics of Big Science, Hevly listed the following:

o Not merely an increase in resources devoted to research, but the concentration of those resources into a declining number of centres dedicated to specific goals. o Specialisation also within the centres, with specialised teams and hierarchical structures. o The attachment of social and political significance to specific projects, addressing broader objectives such as health, military power, industry or national prestige.

He distilled from within the essays four major themes.

First, within big science, the relationship between science and technology has taken new forms, which have influenced the nature of both endeavours. Second, the social context of big

467

science has included the interactions of scientists, engineers and military officers, and the latter, especially, have been largely overlooked by historians up to this point. Third, precisely because of its size, big science’s institutional context commands the attention of historians and its extensive documentation supports especially productive research along these lines. Finally, the importance of collaborative research, both for scientists and historians, is highlighted here. The first two themes especially have been traditional concerns of social sciences and historians, but these essays provide new insights and the basis of fresh departures. The latter two themes are just beginning to be explored; here this volume begins to break new ground.718

Impact of the Galison and Hevly book

Big Science was generally well received by reviewers, being perceived as filling a gap in consideration of the issues, which had been left little explored since Weinberg and Price; it was welcomed for the quality of the contributions and of the production. The conclusion from Galison and Hevly that there is “no one size fits all”, that Big Science was not one thing but many things and diverse experiences, was reinforced in many of the reviews. If a succinct definition was sought, one reviewer719 pointed to the definition first put forward by Weinberg, namely that a science becomes Big Science when its funding is non-negligible on the scale of advanced national economies.

Recurring themes are also underlined, such as the tendency for historians to overrate World War II as the cause of Big Science and to underrate prewar developments in places like Lawrence’s laboratory at

718 Galison and Hevly (1992), p. 357. 719 Pickering (1992), p. 110. 468

Berkeley. In her review Bromberg720 commented that the papers in the collection were united by a pair of claims, firstly that “current characterizations of big science – in terms of expensive equipment, large and hierarchical teams and interdisciplinary research – do not fit enough real life instances”; secondly that a new definition, drawn from case studies, was needed. The papers, she claimed, showed that that “present day definitions fail before examples that our instinct tells us are truly big science” but did not succeed in providing “a more commodious definition. Instead they suggest that the present definitions of big science and little science are more usually taken as complementary idealizations. Any concrete case is likely to show, simultaneously, elements of both”.

This underlines an issue raised by Price and others, that any clear distinction between big and little science is difficult and perhaps not productive. We have rather a continuum of scientific endeavours shading from what is clearly “little science” to what is unequivocally “big science” but with much uncertain territory between. Given the complexity of criteria, it may well be that Oliphant’s enterprise at Birmingham lay somewhere in the middle.

In his review Robert Kohler wrote

Another idea that recurs is the notion of control, which came to the fore as physicists entered into working relationships with engineers, managers, politicians and government officials. Big Science projects were shaped by contests to control scientific work. The relationship between physicists and engineers also emerges here as a critical one, ripe for competitive and thematic

720 Bromberg (1993) p. 441. 469

study. The key issue was not pure versus applied science but different visions of applied science721.

Both these issues have emerged in the current study significant in the context of Oliphant’s endeavours, particularly in Birmingham. Given that Oliphant’s requests for substantial funding for the latter project had to pass through a bureaucratic process and be ultimately approved by politicians, it is reasonable to expect that his civil service and political masters would have required some accountability regarding the spending of the money in the light of the outcomes Oliphant had led them to expect. However, our examination of the documentary record has yet to show what form, if any, that process of accountability took. It may prove that no-one took responsibility for determining whether the money had been well spent.

There is already some anecdotal evidence concerning the complex relationship between Oliphant and his team of physicists, many of them postgraduate students, on the one hand, and the few professional engineers who were engaged to support the project. Oliphant had an outstanding innate technical sense but that was combined with a “can do” or “crash through” attitude, which led him to regard engineers as overly conservative. It also clear that he had far less engineering support than the American machine-building projects which were proceeding at the same time, and that is one of the several factors determining the differing outcomes of the various projects.

Catherine Westfall and “mezzo-science”

A substantial critique of the approach to Big Science taken by Galison and Hevly was contained in a study by Catherine Westfall722, written a

721 Kohler (1993), p. 417. 470

decade later, and largely devoted to documenting the development of the Bevalac at Berkeley. This was an accelerator of heavy ions constructed by linking two existing machines, one of them, the Bevatron, being one of the first three proton synchrotrons. She regards this as an example of “mezzo-science”, a category lying somewhere between “big” and “small”. (or “modest” and “grand” in her terminology).

Westfall framed her discussion with consideration of the sorts of issues raised by Galison and Hevly. In some aspects, she aligned with the conclusions drawn by them, in others she departed markedly. An example of the first is the question of “periodizing“, that is, the notion that Big Science had its beginning in the massive wartime projects that delivered the atomic bomb and the many uses of microwave radar. She noted that several contributors to Galison and Hevly volume undercut this common point of view by demonstrating that large and expensive projects, drawing substantially on external (mostly industry) funding, were well underway in the 1930s in Berkeley, Stanford and elsewhere.

This view is supported further by Oliphant’s experience. It can be argued that “Big Physics” entered his career, as early as 1932, when he began working in tandem with Rutherford on the first of the basement accelerators inspired by the achievements of Cockcroft and Walton. Compared with the previously-used method of securing bombarding particles for nuclear physics investigations, namely natural radioactivity, these devices were large and complex, drawing on high-voltage engineering to accelerate the particles, and advances in electronics to detect outcomes. On those grounds, if not on others, they were probably entitled to the “big physics” tag.

722 Westfall (2003). 471

From that point, Oliphant’s endeavours were guided by one motive, to secure ever higher accelerating voltages or beam currents, and those endeavours inevitably required larger and more complex equipment. The milestones along the path (once he left the basement) were one (or possibly two) generations of high-tension equipment which under his supervision were installed in the new Austin Wing of the Cavendish, the Nuffield cyclotron and the proton synchrotron in Birmingham, and the unfinished accelerator at the Australian National University. It is a matter of debate at which point along this path, we entered the realm of Big Physics.

Along the way, we also encounter a second commonly-ascribed characteristic of Big Science, increasing costs and the need for external funding. Oliphant was well aware of this well before the onset of World War II. The high tension-laboratory was possible only through a bequest from Lord Austin (the funding also provided Cockcroft with the opportunity to build a cyclotron); the cyclotron at Birmingham was paid for mostly by Lord Nuffield, who also contributed to the proton synchrotron (through the Nuffield Foundation) , though the bulk of the funding came directly from the British government. The Australian government would later underwrite the building of the Canberra accelerator.

In her conclusions, Westfall was uncompromising. She had previously noted that Galison himself had acknowledged that the Big Science label does little to help us understand large scale research. In his study of high-energy particle detectors, she commented that as an analytical term, “big physics” is about as useful to the historian of science as “big building” would be to a historian of architecture.723

723 Ibid p. 36. 472

As an aside, we could say that the replacement of the term “big building” with “skyscraper” might yield a useful metric. The emergence of skyscraper (initially meaning buildings with as few as 20 or 30 stories) was not driven (at least not primarily) by a desire for grandeur but by practical and economic considerations, such as how to house more workers or residents in the same (increasingly costly) foot print. It was made possible by technological developments in various fields; new construction techniques, better steel for framing, reinforced concrete, the invention of the elevator. Furthermore, skyscrapers had much greater impact than smaller buildings on their surroundings though increased traffic, the “concrete canyon” effect, the casting of shadows and the demand for utilities. We can see parallels with the “skyscrapers” of Big Science, such as particle accelerators.

Westfall had other grounds for wanting to do away with the term altogether. Worse than merely not useful, it is counter-productive.

The Bevalac story contains other lessons. Particularly important is the demonstration of what is lost if historians relay on the rhetoric of participants. The problem has been that along with the term “Big Science” we pick up values that may not suit our purposes. For example, for high-energy physicists talk of Big Science underscores the superiority of their science, with its focus on probing fundamental nature with the grandest, more expensive equipment. For them, the Bevalac, like all smaller forms of large-scale research, does not figure in the discussion because it is puny and useless in the pursuit of superior science. Although belief in the peerless grandeur of their subfield might help high-energy physicists compete for resources, the preoccupation with bigness merely restricts and obstructs our view. If we continue to frame discussions in terms of Big

473

Science, in fact, there will continue to be many untold tales and therefore many missed lessons.

The lessons learned from the Bevalac story may have led Westfall to adopt this tone. The new machine was not well accepted by high-energy physicists with dreams of building something much larger and who therefore saw it as a compromise. At this point it is worth quoting from a high-energy physicist, though one not involved with the Bevalac. The contribution of Wolfgang Panofsky to Galison and Hevly was based primarily on his experience as director of the Stanford Linear Accelerator Centre (SLAC).

This chapter should document that big science as typified by SLAC has indeed made necessary some changes in academic practices and methods of managing scientific enterprises. Yet one should not lose sight of the fact that the motivation of big science and small science remains the same; expansion to big science is brought by technical necessity, not by delusions of grandeur of the scientific community. We simply do not know how to obtain information on the most minute structure of matter (high- energy physics) on the grandest scale of the universe (astronomy and cosmology), or on statistically elusive results (systematic genetics) without large efforts and large tools. The evolution of technical and scientific fact has driven the changes, not change in motivation or ethics of the scientific leaders724.

It can be argued that Oliphant’s motivations in promoting the Birmingham proton synchrotron seem closer to those outlined by Panofsky than to those set out slightingly by Westfall. It was apparent to him that further significant progress in elucidating the mechanisms

724 Panofsky (1992) p. 145. 474

behind the release of nuclear energy would require the imparting of much higher energies to the bombarding particles and that would necessarily require a large machine (though nowhere near as large and expensive as if existing principles such as those embodied in the cyclotron had been followed).

Nonetheless, Westfall pursued her uncompromising solution.

To remedy the problem, we need to abandon talk of Big Science and find rhetoric that fits our own agendas. As a step in this direction, this discussion of the formation of the Bevalac and relativistic heavy-ion physics from various types and sizes of science speaks of modest, mezzo and grand science. This is only one example of finding words to tell our own story. Thinking in terms of “expensive science” might allow us to focus more precisely on the important issue of how much large-scale science of various kinds costs taxpayers, not only in construction costs but in continuing yearly expenses. We might also consider “multidisciplinary large-scale science”. Argonne National Laboratory’s Advanced Photon Source, for example, serves physicists, material scientists, biologists and others. How other networks of large-scale research extended by researchers from such disparate fields, and what new features of the large-scale research arise from the disconnections, connections, and interconnections of these networks.725

Clearly Westfall thought that after nearly half a century of usage the term Big Science has ceased to be of value, and we need alternative terminology (such as her tri-fold division into “modest”, “mezzo” and “grand”). Certainly circumstances have greatly changed since the term

725 Westfall op cit p. 56. 475

was first popularised by Weinberg and Price, and that has inevitably led to an evolution in its meaning.

Oliphant’s endeavours lay relatively early in the evolution of Big Science, much nearer the first stirrings of “big nuclear physics” in the early 1930s than to the writings of observers today, who have many more, and much grander, examples to choose from. The proton synchrotron in Birmingham was completed nearly a decade before Weinberg’s first use of the expression. It may be that Oliphant’s machine displayed only some of the characteristics that observers delineated when attempting to fill the term with more meaning than Weinberg and Price had given it.

The ethics of Big Science

A number of commentators have drawn attention to the ethical dimensions of big science. The Birmingham enterprise, coming early in the evolution of the phenomenon, did not face some of these issues, such as the implications of the increasing numbers of multi-author papers for the traditional requirement for individual researchers to take responsibility for their results and conclusions. According to available information, all the papers generated by the Birmingham machine had only one or two authors.

Other ethical challenges were more central to the Birmingham saga. We have noted how the granting of public funds which made the machine possible was the result of a decision-making process in which Oliphant himself played a central role, along with a number of fellow physicists all keen to get their hands on some of the government largesse. With the exception of the occasionally-present Treasury official Alan Barlow, no- one involved in the process could be judged objective. This point is

476

underlined by Barlow’s comment, noted in Chapter Six, that the Subcommittee might perhaps have sought first to consider what facilities the nation needed, and then to allocate those facilities among the interested parties, rather than to essentially ask each other what they wanted. The Subcommittee defended its actions, but the self-interest of the parties is evident, as is the privileged position they enjoyed in pursuing that self-interest.

There are other dimensions to this issue worth exploring. The securing of the necessary resources for these facilities (substantial in their day) required the exercise of political power, as a capacity to influence the bureaucratic decision-making process. The physicists in question possessed that power by virtue of their positions and their reputations, and the perception that they had played a major role in winning the war. Over the previous half-decade, science and government had become intermeshed to an unprecedented degree. The personal links and lines of influence between physicists like Oliphant and Chadwick, politicians like Anderson and science bureaucrats like Akers and Appleton had greatly strengthened. It was through that network that the physicists exercised their power over policy.

Capshew and Rader726 assessed the situation, with particular reference to the United States, as follows.

World War Two underscored the links between science and politics and brought about a new era in the relations between the American scientific community and the federal government. Scientists, especially physicists, enjoyed the power and prestige that accompanied the perception that their efforts had done much to win the war. Eager to reap the benefits of their new position,

726 Capshew and Rader (1992). 477

scientists gained unprecedented support for basic as well as applied research. The expansion of support and the ensuing growth of scientific activity brought policy issues to the fore727.

Nonetheless, the planets had to be aligned. The British physicists got their way (and their money) only because the keepers of the public purse were willing to give them a hearing, having judged that the national interest demanded substantial public outlays for research into the military and civilian applications of nuclear power. Those being the grounds on which funds would be made available, the requests for support had to be appropriately crafted. Oliphant understood that as well as anyone, and indeed was ahead of the game. From earliest times, his applications for funding made no reference to the value of “knowledge for the sake of knowledge”, which would have sufficed to secure the much smaller resources needed for research at the Cavendish a decade before.

Now, with the stakes much higher, the emphasis had to be on practical outcomes in the national interest, on advancing the technology for the utilisation of nuclear power and on training the nuclear physicists that national interest would require. Oliphant was effectively offering his services, and those of his colleagues, to the state in return for financial support. Again, given the fact that such large-scale, publicly-funded endeavours were still new (and small compared with those which would follow) it is certainly a consideration.

As we have already noted, at least one of those rationales proved barren. Oliphant’s machine contributed nothing to advancing the understanding of the release of nuclear energy, though its contribution through training to the national effort in nuclear physics is easier to

727 Ibid p. 13. 478

substantiate. Its major impact was in fact on “knowledge for the sake of knowledge”, on the corpus of understanding of basic nuclear physics, such as the theory of nucleon interactions, the production of pions and the first sightings of resonances. The same was true of the other first- generation synchrotrons.

Beyond that, its enduring legacy was largely in stimulating the growth of a centre of skill in interpreting the visual records of nuclear reactions, so enabling the deeper physics to be plumbed. This was a most valuable development, but again, given that most of those records came from the operations at CERN, when the energies available were sufficient to initiate processes that hardly ever occur in the everyday world, the practical value was of such work was slight at best. Capshew and Rader quoted Abelson’s assessment of the practical benefits derived from the huge public investment in “particle physics” (as the field in which Oliphant worked could now be named); “Never, in the history of science, have so many fine minds been supported on such a grand scale and worked so diligently, and returned so little to society for its patronage”728. In his assessment, Big Science, at least in this field of research, had been a dry well in practical terms.

Conclusion

The purpose of this analysis has been two-fold: to examine some of the literature of Big Science over 40 years, and to begin to place Oliphant’s Birmingham enterprise in the spectrum of issues and concerns that the study of Big Science generates. This not only provides one of the several contexts into which the story should be set, but also allows us to frame a set of characteristics of Big Science and then see how well the Birmingham enterprise matches them. We here set out the

728 Ibid p.13. 479

characteristics as a series of questions, to most of which possible answers have been given during the discussion above.

o What was the confluence of circumstances that led to the initiation of the project? To what extent could it be judged “inevitable” in Weinberg’s terms, either specifically or more generally? o Was the Birmingham accelerator genuinely “big” in relation to the size and cost of other projects proceeding at much the same time in the UK and elsewhere? o How far back can the origins of the Birmingham machine be traced? Was it essentially continuous with earlier relevant endeavours, or is there an identifiable transition? o What was the nature of linkages to players beyond the university, to machine builders elsewhere, to industry, to the representatives of public and private funders? o How autonomous was Oliphant in controlling the project; in other words by what process, if any, was he held accountable by other stakeholders for progress and outcomes? o How was the project justified in terms of its likely impact on broader trends in society? o How hierarchical was the management of the project? By what mechanism was progress monitored and directed? What degrees of specialization were evident among the participants and teams? o How far did the influence of the project spread beyond Oliphant’s own group of physicists and engineers? To what extent could it be judged “interdisciplinary”? o To what extent did the project display concurrently the characteristics of “little science” alongside the “big”? o How did involvement in the project impact on the later careers of participants?

480

Only by addressing these questions can an answer by given to the question which forms the title of this chapter. What follows is an attempt to frame an answer.

In terms of physical size, cost and complexity of operation, the Birmingham machine can legitimately be judged as “big” in the context of time and place. Nothing remotely comparable existed in Europe at the time, and certainly not in the UK, and only the 184 inch cyclotron at Berkeley was in the same league at the time construction commenced. Indeed in terms of tonnage of steel the Berkeley machine was larger, but its complexity of operation was noticeably less. Oliphant's accelerator was also big in terms of what had gone before, though it was surpassed on that scale within a decade, and rendered obsolete within two. As has already been noted, the cost of the machine, and of the rest of the small group of new facilities authorised in the UK in the immediate post-war years, was noticeably high when compared with earlier machines, and indeed with the expenditure from the public purse on any other area of science at the time. Whether or not this constituted "bigness", it certainly meant the machine loomed large in government expenditure on science.

We should at the same time remind ourselves that the Birmingham machine was certainly less expensive than its rivals in the United States, and that a lower cost reflected to some extent the much smaller numbers of research and technical staff involved. As we have noted, the roster at Birmingham barely numbered a dozen at any one time, and probably less than 30 over all, at least until the time of its completion. These smaller numbers, combined with the particular collegial atmosphere evident within a university setting, meant that the hierarchical structure of specialised teams, which some have taken to

481

be a characteristic of big science, was not evident. Certainly there was specialisation, with small groups of one or two taking command of particular aspects of the project, but the coordination of these endeavours was informal rather than hierarchical.

Oliphant himself has commented that the Birmingham machine, for all its size and complexity, represented an example of the old Cavendish "sealing wax and string” tradition. This is of course merely a metaphor, but it does point to the fact that so much of the machine, and of the systems that supported it, was built in-house by the physicists who would later use them, as had been the case in the Cavendish, and indeed in many laboratories up until this time. To that extent, it displayed one of the traditional characteristics of "little science".

At the same time, the machine did display that increased productivity which we earlier suggested may be a relevant metric of Big Science. The machine was able to be used by a number of individual researchers and research teams; indeed demand was such that at times it was operated around the clock. This trend had been evident for at least a decade. Whereas Oliphant's basement accelerator was simply for his use with a few colleagues and under the guidance of Rutherford, the Cavendish high-tension laboratory and the Nuffield cyclotron had both been designed to give observing time to a range of investigators.

It is not yet possible to assess the degree to which Oliphant (and those who came after him in command of the machine), possessed the “autonomy" in their actions which some have suggested has to be largely given up in the Big Science era. Given the magnitude of the public expenditure, it would have been reasonable to expect that Oliphant be required to report on a regular basis, perhaps annually, to the government on progress. However no documentary evidence of

482

such accountability has yet been uncovered. Certainly the University reported through its annual publications such as those of the Joint Committee on Scientific Research, but that was to a general audience.

On the issue of the inevitability of the Birmingham machine, given Weinberg's insistence that Big Science is “inevitable”, we can give both a general and a specific answer. The Birmingham machine itself was by no means inevitable. Had Oliphant not been there with his vision, his drive and his stock of capital it is unlikely it would have ever been created. But more generally, as Panofsky and others have noted, and as Oliphant himself argued, advances in nuclear physics did inevitably require the building of larger machines to generate beams of higher energy. This trends has been in evidence right through to the present day.

On the basis of these diverse considerations, can we reach any conclusion? On some metrics, the Birmingham accelerator was "big", in others it was not. To some extent this reflects a lack of precision in the definition of Big Science, and that over time, the goalposts for Big Science have moved. Oliphant's endeavour, coming relatively early in the evolution of this phenomenon, is not as indisputably big as some accelerators that came later, such as or CERN. What cannot be disputed is that, for its time and place, it was a major achievement.

483

10. CONCLUSION

The purpose of this investigation has been to examine the role of Mark Oliphant in the early development of the proton synchrotron as a new class of high-energy particle accelerator. To be more specific, we proposed to test the hypothesis that Oliphant is entitled to pre-eminence among the pioneers of the proton synchrotron, in terms of both his own contribution and his influence on others. We argued that if this hypothesis can be sustained the role commonly ascribed to Oliphant in histories of this area of research significantly understates his importance.

We also advanced a secondary hypothesis: that the Birmingham accelerator is to a great extent an embodiment of Oliphant’s own style and personality, in both its positive and negative aspects and that it reflects the “capital” that Oliphant, as a consequence of his native skills, training, experience and contacts, was able to bring to the project. This would imply that the project, though collaborative, could not have been set in motion without him. His influence endured. The project was able to sustain a course to its successful completion, if more slowly, even after his direct personal involvement was withdrawn.

Setting the context

To undertake such an examination we have undertaken several supporting tasks which serve to set the conception, design and construction of the Birmingham machine in a number of contexts. These tasks were:

484

(a) To undertake an attributional survey to examine the role commonly ascribed to Oliphant in these developments. This is reported in Chapter Two. In brief, we found that Oliphant is usually named as one of the three independent co-inventors of the “synchrotron principle”, along with the Russian Vladimir Veksler and the American Edwin McMillan. It is also commonly acknowledged that Oliphant was first to seek to apply that principle to the acceleration of protons rather than electrons. However the resulting machine commonly receives less space in overviews than the two almost contemporaneous American machines, the Cosmotron at the Brookhaven National Laboratory and the Bevatron at the Radiation Laboratory at the University of California Berkeley, and certainly less favourable mention.

(b) To examine the development of the technology of experimental nuclear physics since the early experiments by Ernest Rutherford and his co-workers; in other words to establish the “ecology” of the field. This set the background against which the development of the proton synchrotron took place. It is reported in Chapter Three. We found that the essentially-unchanging method of investigation into the structure of the atomic nucleus and the nature of the forces operating in that realm was to direct a beam of high energy subatomic particles at nuclei embedded in fixed targets in order to stimulate nuclear reactions and then to intercept the products of such reactions so that they could be characterised. The bombarding particles were initially natural emissions from radioactive nuclei. These were replaced from the early 1930s by particles (protons and later deuterons) artificially accelerated by various designs of linear accelerator and by the cyclotron invented by Ernest Lawrence and his colleagues. This chronology is dominated by

485

an ongoing quest for higher energies in the bombarding particles in order to generate new phenomena for study. The proton synchrotron continued this strategy of investigation.

(c) To chronicle the earlier stages of Oliphant’s own career, so as to examine how he gained the “capital” that he was later able to bring to bear on the Birmingham enterprise. The term “capital” includes technical skill, experience in the generation and management of large projects (including the raising of funds), leadership, reputation and networks. The outcomes of this investigation are reported in Chapter Four (covering the period until the outbreak of World War II) and continued in Chapter Five. We found that his high-order technical skills had been evident from a young age. His career through the 1930s at the Cavendish Laboratories and then at the University of Birmingham saw him begin to build (and in some cases complete) a number of particle accelerators of increasing energy, size, complexity and cost, culminating in the Nuffield cyclotron at Birmingham, which required a significant element of fund-raising. His period as Deputy Director of Research at the Cavendish under Rutherford was followed by appointment to the Poynting Chair of Physics in Birmingham; his reputation was enhanced by his being elected FRS at only 36. He was able to build strong and enduring relationships with a number of figures in his field, including James Chadwick, John Cockcroft and Ernest Lawrence. All of these constitute elements of “capital”. Through the war, his standing was enhanced by the leading role played by his laboratory in the development of microwave radar and later by his prominence in the British contribution to the Manhattan Project.

486

(d) To seek to understand how the immediate context of the Second World War provided a stimulus to the initiation of the proton synchrotron. In the context of our investigation, the issue of nuclear energy is obviously of central importance, from the discovery of nuclear fission less than a year before the outbreak of war to the first use of atomic weapons, against Japan in August 1945. An account of these issues is given in Chapter Five. We found that, though much occupied with the development of radar even before the war began, Oliphant was an insider on nuclear energy matters as well. The famous and influential Frisch-Peierls Memorandum was written in his laboratory; his contacts in the defence science establishment (such as Tizard) enabled him to bring that document to the attention of the right people, stimulating the formation of the MAUD Committee of which he was an early member. In 1941 he played a leading role in generating interest in the USA in the military potential of nuclear energy, so stimulating the establishment of the Manhattan Project. In late 1943 he led a team of British scientists to participate in the project, many (including Oliphant himself) joining Lawrence at Berkeley to work on electro-magnetic methods to separate the isotopes of uranium for use in the weapons, technology that was put into practice at the Oak Ridge plant code-named Y-12. Oliphant was essentially deputy both to Lawrence (in terms of technical development of the electro- magnetic separation process) and to Chadwick (in terms of the overall contribution of the British team to the Project). It was in that context that Oliphant first conceived the proton synchrotron and determined to build such a machine once the war was over. Given the obvious continuing importance of nuclear technology, both civilian and military, governments (including the British) were willing to invest in the expensive new technology needed to

487

continue progress in nuclear science, a willingness from which Oliphant and others benefited.

Oliphant’s place in history

The considerations of context summarized above brings us to the central element of this study, an examination of Oliphant’s place in the events immediately following the war that resulted in the initiation of the first proton synchrotrons. The evidence is presented in Chapter Six. Our investigation has identified two groups of people with substantial involvement in the early story of the proton synchrotron. The members of the first group, McMillan, Veksler and Oliphant, discovered (or invented) what we have called the “synchrotron principle”, namely the strategy of constraining particles in an orbit of constant size and of accelerating them by applying an alternating voltage which increased in frequency as the particles picked up speed. As far as can be determined, the three men reached this conclusion independently, a fact generally acknowledged in the literature, though as we have found, and will summarise below, the path by which Oliphant came to the concept of the synchrotron was markedly different to that taken by the other two.

A second group, comprising Oliphant, William Brobeck and Stanley Livingstone, took the concept of the synchrotron and applied it to the acceleration of protons, leading the teams that built the first such machines, Oliphant in Birmingham, Brobeck in Berkeley (the Bevatron) and Livingstone at Brookhaven (the Cosmotron ). Only Oliphant was a member of both groups. Indeed his first ideas on the machine were clearly directed toward the acceleration of protons as well as electrons, and addressed many of the resulting technical challenges. This was not true (except in the most general terms and not in the immediate postwar years) for McMillan and Veksler. In terms of practical application of the

488

principle, McMillan concentrated his attention on building an electron synchrotron at Berkeley. While Veksler was a key player in the building of a large machine in the Soviet Union (known as a syncrophasatron in his terminology), this was not commenced until well after the Birmingham accelerator was under construction.

Furthermore, Brobeck drew his inspiration largely from the work being done at Berkeley to apply the synchrotron principle as enunciated by McMillan to improving the performance of the cyclotron, though as we have seen, Oliphant was also influential. As we will be reminded below, Livingstone’s inspiration for the Cosmotron was news of the work at Berkeley, conveyed to him by Rabi. In Oliphant’s case the transition from conception to construction was more seamless. As a member of both groups, Oliphant was his own inspiration.

These elementary considerations alone are enough to give Oliphant a special status among the pioneers of the proton synchrotron. But we have been able to go much further, so finding more support for our hypothesis, largely through the analysis of surviving documentation, little examined until now. We shall summarise some of our findings under three headings.

I. The origin of the concept of the proton synchrotron

We have advanced the case that Oliphant was the author of the first written accounts of the concept of a synchrotron as applied to the acceleration of protons. We have identified four of these, spanning a period of six months from late 1944 to mid 1945, and showing the development of Oliphant’s ideas over this period. A one-page account from November 1944 was followed by a letter to Akers in January 1945, then the undated “Oliphant Memorandum” as it survives today, mostly

489

likely composed in the late spring of 1945, and lastly a second letter to Akers in June 1945, essentially a covering note to a modified version of the Memorandum. These documents demonstrate that Oliphant began to set down thoughts on the synchrotron principle (though he did not use the term) at least six months before anyone else and as much as two years before any substantial written proposal to use that principle to accelerate protons (rather than electrons).

It appears that these documents drew on ideas which had come to Oliphant while stationed at Oak Ridge (Site X) in the USA (that is, during the “owl watch”). To that extent, Oliphant’s own account of the genesis of the concept is credible. The same cannot be said of other details of his account. As we have shown, the most likely time of creation of the Memorandum was not either September 1943 or some time in 1944, as Oliphant variously claimed, but towards the middle of 1945. We cannot substantiate Oliphant’s claim that Akers replied to an early letter raising the possibility by telling Oliphant to “get on with the war”, as no such reply survives in the record. Indeed the provenance of two of these documents (including the Memorandum) can be challenged, but collectively the documents present a consistent and intelligible picture.

We have made an important distinction between Oliphant on the one hand and McMillan and Veskler on the other regarding the thought process whereby the three men conceived the way the synchrotron would work and the benefits of the technology. McMillan and Veksler were concerned mostly to overcome the energy limit imposed on the cyclotron by the relativistic mass increase of the orbiting particles. Their solutions were similar in principle, if variously named. They paid comparatively little attention to the use of such a concept in the acceleration of protons. In the course of their investigations, McMillan, in particular, came upon the notion of stable orbits which led him to

490

conceive (and name) the synchrotron. Oliphant however was little concerned with the mass increase issue, though he acknowledged it. As a result he missed the synchro-cyclotron concept, as he has admitted. His problem with the cyclotron was its size and cost, and the resulting limitations on beam energy, which, as he realised, could be greatly reduced by confining the particles to an orbit of constant radius.

II. The first proton synchrotron

It is also clear from the surviving documentary record that Oliphant was first to seek funding for the building of a proton synchrotron (in July 1945), the first to secure it (in March 1946), the first to undertake detailed design (from September 1946), and the first to commence construction (in mid 1947). In all of these dates the Birmingham project was well in advance of either the Cosmotron or the Bevatron. The decision to build the two American machines was not taken until April 1948, more than two years after Oliphant had secured his funding, though design studies for the two accelerators had already been undertaken.

Despite its earlier start, the Birmingham accelerator was not the first to achieve a proton beam in the BeV range. It first went on-line in July 1953, a little more than a year after a beam was achieved at the Cosmotron, though it can be argued that Birmingham was first to reach its design energy. The reasons for the delay were several: two substantial design/construction faults in the magnet, each of which generated significant breakdowns requiring week or months of repairs, the departure of Oliphant from the project in mid 1950, requiring management to be taken over by Moon who was neither well-prepared nor initially enthusiastic for the task, and the almost simultaneous death from illness of Gooden, the key operative in the project.

491

More generally, it can be argued that the project suffered from a lack of thorough planning and an over-reliance on speed and innovation in the solving of difficulties. Such an assessment is widely supported by participants in the enterprise. Construction began before many of the technical issues had been solved. In general there was far less reliance on engineering expertise than in the United States. In a positive light, the Birmingham approach was arguably the only way such a project could be set going in immediate-postwar Britain, with industry still recovering from the ravages of war, with many materials in short supply and great demands being made on universities faced with a flood of new students. The Birmingham project also appears to have delivered a beam at much lower cost per MeV than the American machines, a reduction of at least 50%. The close involvement of the physicists in design and construction tasks, which might have been undertaken by engineers elsewhere, allowed them to be more responsive to the possibilities for upgrades and modifications that delivered benefits to the experimentalists.

It has been suggested, including by Oliphant himself, that the Birmingham machine suffered from “first-mover disadvantage” (my terminology), that its design as a simple circle, without straight sections as in the other two early machines, made for difficulties, such as with beam extraction, so limiting its usefulness. If there was such an initial disadvantage, it appears to have been overcome by typical ingenuity. Certainly in time a beam was extracted and put to good use. Indeed, later in its life, the accelerator delivered two beams, so increasing productivity. The mooted disadvantage may have been reflected in other ways. For example an early decision was made to tune the RF voltage with a mechanical device using mercury, rather than by using the new ferrite materials just coming on the market (and which were used in the

492

US accelerators). In the long run, this seems to have made little difference; the mercury inductance worked well enough and was in time replaced by one using ferrites.

It has also been suggested by Moon and others that the decision to locate the machine in an existing space, rather than to wait for a new building to be constructed as was done for the American machines, had severe adverse impacts, limiting the energy of the beam the machine could produce and the capability to add on the auxiliary equipment needed for experimental work. The evidence does not support either of these statements. We have seen that in Oliphant’s proposals the size of the machine was determined plausibly by the desire to reach a beam energy above the rest mass of the nucleon, and so to reveal new phenomena. He made no reference to the limitation of space, though this may have been an unspoken secondary consideration. The space around the machine may have been limited compared with that at Brookhaven and Berkeley but again ingenuity seems to have found room both for the auxiliary equipment and the shielding needed to protect workers. The ongoing challenges were more to do with errors in magnet design and construction, resulting in the “pole pieces disaster”, the two substantial breakdowns from shorting-out of the magnet coils and the frustrating influence of the fringing field.

Despite all the difficulties, the synchrotron operated successfully, and almost continuously, for 14 years, with an ongoing program of improvements that maximised the value to be gained from it. Its users enjoyed significant interaction with other centres of research. Contrary to the adverse comment we have quoted from Alvarez, it appears that much useful physics was done with it, given its limitations of energy and beam intensity, and a generation of physicists gained experience which, because of the small size of the machine, tended to be more

493

encompassing than with the larger American machines where involvement by students was more specialised. We can state with confidence that the Birmingham machine achieved the two broad objectives that had been set for it, both by Oliphant and those who supported his vision with funding: to train physicists for the national effort in nuclear energy and to uncover new phenomena to extend understanding of nuclear physics. That that new knowledge did not transform the release of nuclear energy by nuclear fission, as Oliphant had argued it might, does not undermine those achievements; such a transformation has not occurred in the nearly six decades since even though vastly greater energies are available from today’s accelerators.

Even after the machine had been closed down and dismantled, its influence remained, in the form of a core of expertise in the interpretation of the visual records of nuclear reactions obtained with bubble chambers and similar equipment. The reputation of the Department’s “visual techniques” group endured for many years.

III. Oliphant’s influence in the USA

It has been argued, most strongly by McMillan and Livingstone, that plans and progress in Birmingham were either little known in the USA or had little effect on developments there. This is questionable based on the evidence available and presented here. We have here argued that Oliphant influenced the genesis of the Bevatron from at least mid 1946. Certainly the Bevatron pioneer William Brobeck acknowledged that influence, which may have begun with a lecture Oliphant gave at Berkeley in May 1946; at that time Oliphant’s preparations for his own machine were well advanced and those at Berkeley still embryonic. Earlier influence exerted through the presence of Otto Frisch in Berkeley in March, 1946, is harder to prove, even though he had been asked (by

494

Chadwick) to gather information on accelerator developments in Berkeley. It remains plausible and worthy of further investigation.

Furthermore, given that several accounts assert that the concept of the proton synchrotron was taken from Berkeley to Brookhaven by Rabi, it can be argued that Oliphant’s influence extended, secondhand as it were, to Brookhaven and to the Cosmotron, though Livingstone denied any such influence. The claim by some of Oliphant’s co-workers that the decision to build a proton synchrotron at Brookhaven, instead of the earlier proposed synchrocyclotron, came as a result of Livingstone’s visit to Birmingham late in 1947, cannot be sustained at present, nor even the lesser claim that this visit resulted in significant design changes to the Cosmotron design, such as the addition of straight sections.

Looking further back, the balance of evidence suggests that Oliphant’s plans were known to the Berkeley community well in advance of their first appearance in the scientific literature in March 1947. More specifically, it appears that Oliphant sought Lawrence’s advice on the feasibility of the project during the latter years of the war, and that Lawrence supported the concept in general terms. It also appears that McMillan was aware of the broad dimensions of the proposal before the end of 1946, despite his later denials.

The reason for this failure to acknowledge Oliphant’s early influence, a failure which extended through the following decades, is perhaps best explained by the reference to the concept of phase stability. McMillan, who with Veksler had first set the concept on paper in detail, regarded it as central to the feasibility of the machine, as indeed it was, and he doubted that Oliphant understood it. A case has been made here that Oliphant did in fact have an adequate, if elementary, understanding of the notion, without (as Oliphant himself later admitted) the necessary

495

detailed mathematical treatment. This is, however, not relevant to McMillan’s judgment. The first detailed mathematical presentation of the issue to emerge from the Birmingham team did not appear until March 1947729. Oliphant’s earlier presentations, such as the January 1945 letter to Akers which contained some consideration of the issue, were unknown to McMillan. So as late as the close of 1946, McMillan would have been justified in thinking that Oliphant did not understand phase stability and therefore that, in his terms, the proposed machine was not a synchrotron. This may have influenced the way McMillan interpreted the information he received.

Oliphant and his machine

We come now to our secondary hypothesis, with less documentary evidence to draw on. We have argued here that the Birmingham accelerator, as it was designed and constructed, is in large measure an expression of Oliphant’s personality and style. This assessment applies to both the strengths and weaknesses of the enterprise. In Chapter Eight we recounted some of the assessments of Oliphant’s character as recalled by people who knew him well and had worked closely with him on this project and elsewhere. Those recollections amplify impressions which can be gained from earlier chapters of this study and other sources730. Some of these characteristics can be summarised as follows.

Oliphant as technically skilled. From childhood and throughout his career, Oliphant possessed an outstanding technical competence, dubbed by some “technical green fingers”. He had a remarkable ability to “dominate apparatus”, a quality that arguably

729 See Gooden, Jensen and Symonds (1947) 730 See Cockburn and Ellyard (1981), Carver et al (2002), Beaney (2001). . 496

first drew him to Rutherford’s notice. As a result, he did not fear the challenges involved with large-scale apparatus, but rather welcomed them.

Oliphant as visionary. Oliphant was commonly a forward thinking man, ready to propose and fight for a new project likely to advance his objective. He had argued strongly, against the initial reluctance of Rutherford, for the HT laboratory at the Cavendish, and for greatly expanded facilities, including a cyclotron when appointed to Birmingham. In the closing months of the war, he wrote a number of documents setting out his thoughts on a number of issues of importance in the post-war world, including his earliest ideas about the synchrotron

Oliphant as a risk taker. Oliphant was in a general sense an impatient man, anxious to push ahead with whatever project was at hand, and with a great faith in “the fire in the belly” and in innovation to tackle issues as they arose. He commonly (but perhaps deliberately) underestimated the time needed for completion. He did not see the necessity for detailed planning, and regarded engineers as unnecessarily conservative, more likely to delay a project than to expedite it. At the time he sought funding for the Birmingham venture there was no certainty that it would work. This same characteristic, of being ready to “do what it took”, made him impatient with what he saw as unnecessary constraints, such as the “compartmentalization” designed to ensure security during the Manhattan Project, and caused some to label him indiscreet.

Oliphant as a fount of ideas. To those who knew him, Oliphant displayed a mind of great imagination and inventiveness. Many

497

commented on his ability to generate not merely one but many possible solutions to any technical challenge. However, the proposals were commonly very variable in quality and feasibility. Oliphant’s ability to “winnow the wheat from the chaff” was less developed and he relied on those around him to undertake that task. This sometimes required substantial courage in his collaborators; once Oliphant settled on an idea he was often hard to shift.

Oliphant as a charismatic leader. There can be little doubt about Oliphant’s leadership qualities. He was well able to inspire and motivate the members of his team, notably the younger ones, though some of his senior colleagues, such as Moon, were often more cautious about his enthusiasms. He was persuasive in both the spoken and written word and could be tenacious in pursuit of what he regarded as the right course, taking others with him.

We argue that we can see many of these qualities reflected in the Birmingham adventure, to the extent that it is hard to envisage the project proceeding at all without him. The evidence supports the following statements.

The project “needed” Oliphant. There can be little doubt that Oliphant was the heart and soul of this project, as well as of the Department, in a way not comparable with the role of Brobeck in Berkeley or Livingstone at Brookhaven, central as their contributions were. Both Brobeck and Livingstone were essentially project managers, without the overall responsibility for the institution that housed their creation. Furthermore, Oliphant’s accumulated “capital” made the project possible. He had

498

conceived it, argued for it through the bureaucratic process, used his contacts to secure the needed materials when those were in short supply. Though Gooden managed the project, Oliphant motivated it. Nonetheless, the project survived Oliphant’s departure in mid 1950. Enough of the team remained, notably Symonds and Riddiford, to carry the vital innovative spirit through to completion.

The project was innovative. Nothing of its scale or complexity had even been attempted before. Of the several nuclear physics machine-building projects that were set underway in post-war Britain, drawing on the same source of funds as had supported Oliphant, only his involved a new, indeed untested, technology. Other endeavours such as Chadwick’s at Liverpool and Dee’s in Glasgow proposed to use established technology such as betatrons and cyclotrons, though it is interesting to note that Dee’s betatron became an electron synchrotron and Chadwick’s cyclotron a synchro-cyclotron once those technologies had been proven feasible at Berkeley.

The project was mostly “on the edge”. Undertaken with a minimum of detailed planning and engineering support and commenced with many of the key technical questions unanswered, the project was driven forward by the desire to be the first to get a beam in the BeV range, even if the beam intensity was low and the beam could initially not be extracted. Technical challenges, at least in the early days, were met by brainstorming by the group, with one or two people then given the task of bringing the proposed solution to reality. The outcome of this approach was a machine with some inherent faults that continued to impact on its performance for a decade.

499

Oliphant was willing to leave the project before completion. His departure for Australia in 1950 was a severe blow for the project, especially when combined with the almost simultaneous death of John Gooden. Oliphant’s decision to leave was most likely a combination of factors; his participation in the ANU had been sought by the Australian government since 1946, and he felt the call of duty to his homeland. The unfortunate timing of his departure can be seen as the intermeshing of his over-optimistic estimates of the pace of completion and his impatience that progress was in reality so slow. Always the visionary, Oliphant was mentally engaged in the challenges and opportunities presented by his new position, and the radical new form of accelerator he planned to build there. He may well have been increasingly disengaged from the day-to-day issues of the Birmingham enterprise, especially when a meeting of the whole team had become a rare event.

Was the Birmingham Proton Synchrotron “Big Science”?

In Chapter Nine we broadened our perspective by reviewing the literature on the phenomenon commonly known as Big Science. Our purpose was to assess whether, and in what respects, the Birmingham enterprise could be seen as an example of this development. Such considerations are hard to avoid, and indeed have value. Large-scale nuclear physics facilities are commonly cited, by Weinberg and others, as archetypes of Big Science. When the Birmingham venture was set against a set of criteria that we may draw from the literature and against the context from which it emerged, we were able to draw some conclusions.

500

Some commentators see Big Science (using the capitalization favoured by Weinberg) as a phenomenon emerging from the Second World War, in particular from the Manhattan Project and the development of industrial-scale machines housed in very large laboratories. Certainly the early proton synchrotrons appeared at that time. However, as we have explored in Chapter Three, a tradition of large-scale apparatus for nuclear physics research dates back to 1930 or even earlier, driven by the demand for higher energy in the beams of bombarding particles. Thus in a series of transitions, researchers moved from natural emanations from radioactive elements, to particles accelerated by high static voltages and on to resonance machines such as betatrons and cyclotrons.

In each transition the way was open for apparatus larger and more complex than used before. High-voltage accelerators began with Cockcroft-Walton type machines providing a few hundred kilovolts, and moved onto Van der Graaf devices generating millions of volts. In the hands of Lawrence and his colleagues, the cyclotron went though half a dozen models, of ever-increasing size, cost and beam energy, culminating in the monster 184 inch. Each type of machine ultimately reached its beam energy limit, and had to be superseded if higher energies were to be reached. The concept of Big Science is therefore more relative than absolute, at least in this field, and carries with it acknowledgment of the potential for further growth. The proton synchrotron continued this tradition, even though in terms of size (say of tonnage of steel), the early machines were smaller than the largest cyclotrons. That of course was a major part of their appeal. Oliphant and others were quick to realise that the new technology, whose size and cost grew linearly with beam energy, rather than as the square or cube, could be essentially unlimited in size, apart from considerations of cost. The development of strong focusing, and later of super-conducting

501

magnets, loosened the constraints of cost. As a result we have seen over the last 60 years the building of a series of proton synchrotrons of ever-increasing scale. With the Large Hadron Collider at CERN, we have reached dimensions that even the visionary Oliphant could not have imagined. Yet each of those machines would rightly be judged Big Science when compared with its predecessors. On this basis, the Birmingham machine can also be so judged.

Rather than concentrating on the scale of the hardware, other commentators have looked at the human support needed, the teams of physicists, engineers and technicians and how they are organised and managed. Some see the inevitability of organisational structures akin to those in a large industrial concern, with specialist units and committees arranged hierarchically, linked by lines of reporting and command. Certainly such structures have become more common as the scale of the facilities has increased. CERN has thousands of participants, and could not be organised any other way.

The Birmingham venture did not fit that pattern, for two reasons. As a relatively early exercise, it was not of the scale that needed it. At any one time only a dozen or so people were closely involved. Secondly it was being undertaken in a university setting, where collegiality was the norm, and where the most junior researchers could tell Oliphant face to face if they thought he was wrong. The team was grouped in small units of one or two, perhaps three, with responsibility for a particular system, but formal contact between these teams, such as by meetings of the whole group, was relatively rare (other than in the first year or so when design considerations were dominant). It appears that coherence was maintained more by the leaders (Oliphant, Gooden and later Moon) speaking to each of the units informally. We can take this line a little

502

further. With his “hands-on” style of operation, Oliphant may well not have been suited to more formal, hierarchical ways of working.

We have suggested in Chapter Nine that a key aspect of Big Science, and one that increasingly separates it from “little science”, is productivity. The Cavendish provides examples of this. The older type of nuclear physics experiment, such as those used by Rutherford to demonstrate transmutation, or by Chadwick to find the neutron, were small in scale, usually sat on a bench top, were intended for a specific purpose and could be used by only the one or two researchers engaged in that experiment. The Cockcroft-Walton apparatus barely fitted into a room, let alone onto a bench, but the other characteristics were still evident. Beginning with the Cavendish high-tension laboratory, with inspiration from the Mond Laboratory, we saw the emergence of large- scale apparatus that could be used for a variety of investigations and by a number of independent researchers or teams. These facilities were therefore more productive in terms of the research results obtained, to some extent in line with the higher costs. We can see this with the Birmingham machine, which was used by a number of researchers including those from other centres.

The high cost of Big Science equipment, and the need therefore to raise substantial funds, brings to the fore another key aspect of the phenomenon, noted by several commentators. In this matter, we can see a transition affecting apparatus for nuclear physics coinciding with World War II. During the 1930s, large-scale apparatus in the UK was funded by philanthropists like Mond, Austin and Nuffield, who largely expected nothing tangible in return. In the United States, for example at Berkeley, a similar role was filled by industrial firms, though they were usually more self-interested, seeking help with their problems and access to patents. In the late 1940s, the scale of apparatus began to

503

exceed the funding capacity of philanthropists. There was a growing expectation that the public purse was now the appropriate source, provided the research could be presented as being in the public interest. The quest for scientific and technological innovation had been stimulated and nurtured through Britain’s experiences of the recent war.

For obvious reasons, nuclear physics was well placed to take advantage of this opportunity. The early proton synchrotrons, including Oliphant’s, were funded by agencies closely aligned with nuclear power issues, and on the basis that the research and training they would provide would support the national effort in that field. The public purse continues to fund all the large-scale machines in this field to this day, even though the practical implications of such enterprises (at least in terms of the science) are minimal. The support can be seen now as mostly culturally- based, funding curiosity-driven research with the capacity to excite the human imagination.

Based on the four considerations examined above, we can say that the Birmingham enterprise displayed some, but not all, of the characteristics of Big Science. Some features were more akin to the “little science” seen in Oliphant’s early Cavendish days. Perhaps Westfall’s proposed category of “mezzo-science”, lying between the ends of the spectrum, fits the Birmingham enterprise better.

Closing comments

As the summary set out in this chapter substantiates, the hypotheses advanced at the start of this study can be taken as supported by the presented evidence. Oliphant was at the very least “first among equals” in the development of the proton synchrotron, and in most respects considerably more than that. He had priority in time in the key steps in

504

the creation of this new generation of accelerator, and his influence extended beyond his own machine and his own team. Yet, for reasons examined here, his role has been underestimated in the existing histories of the field. Our purpose has been not merely to boost Oliphant’s claims, but to set the record straight, largely through study of previously unexamined documentation, which has enabled us to develop a chronology of the growth of the concept in Oliphant’s mind.

We have also argued that the machine, and the manner of its construction, reflected Oliphant’s personality, for better and for worse, as well as reflecting some of the characteristics of a new phenomenon, Big Science. The project was made possible because of the “capital” that Oliphant had accumulated and could bring to bear in the climate of post war Britain. Public funding was by then an accepted means of driving innovation in areas linked to national goals.

For all its shortcomings, the enterprise was significant and valuable, meeting its key objectives, delivering good science, training a large number of physicists and providing an enduring legacy for the University of Birmingham. Most importantly, it demonstrated the practicability of a radical new approach to the acceleration of particles to very high energies, one whose importance we can most clearly see from 60 years on, now that we know where it has led.

505

APPENDIX A

PhD degrees awarded for work carried out using the Birmingham proton synchrotron from 1954/5 to 1967/8.

1954/5 Cooper PN Nuclear investigations at high and low energies in Wilson and diffusion cloud chambers. 1954/5 Freemantle An investigation of proton and deuteron induced scattering RG reactions using nuclear emulsion as detector. 1955/6 Ledley RG Scintillation experiments with the proton synchrotron. 1955/6 March PV Studies of the nuclear interactions of high energy protons. 1956/7 Batson AP A cloud chamber study of proton-proton interactions at 650 MeV. 1956/7 White DH The total cross section for the interaction of high energy protons with matter. 1956/7 Young JD Nuclear reactions induced by high energy protons on light weight elements. 1957/8 Barnard ACL Proportional counters for spectroscopy and the interaction of 980 MeV protons with copper nuclei. 1957/8 Booth NE Nuclear cross sections for high energy nucleons. 1957/8 Colwick BB Proton—deuteron interactions at 970 MeV. 1957/8 Duke PJ A study of the interactions of high energy protons using nuclear research emulsions. 1957/8 Gore JE Some studies of the health hazards of high energy particle accelerators. 1957/8 Warren JE Nuclear reactions induced in medium and light weight elements by high energy protons. 1958/9 Batty CJ The scattering of high energy protons by complex nuclei. 1958/9 Colley DC The construction and performance of a 9 inch diameter liquid hydrogen bubble chamber operating in a pulsed magnetic field of 15 kilogauss. 1958/9 Dowdell JD Interactions of 930 MeV protons in a propane bubble chamber. 1958/9 Huq M Double scattering experiments with 950 and 700 MeV protons using a highly selective Cerenkov detector. 1958/9 Law ME The interactions of high energy protons with nucleons and light nuclei 1958/9 Munir MBA Some nuclear interactions with high energy protons 1958/9 Musgrave B Experiments with the liquid propane bubble chamber. 1958/9 Williams AW Proton-helium interactions at 970 MeV. 1959/60 Derrick M Use of a hydrogen bubble chamber to study internal pair production following negative pion capture by hydrogen. 1959/60 Finlay EAS The extraction of protons from the Birmingham 1000 MeV synchrotron . 1959/60 Hill JG A cloud and bubble chamber study of proton-proton interactions at970 MeV. 1959/60 Hoare D The elastic scattering of 925 MeV protons by helium. 1960/1 Frisken WR Elastic proton-proton scattering near 1 BeV. 1960/1 Kinson JB The use of a hydrogen bubble chamber for the study of nucleon-nucleon interactions at 965 MeV. 1961/2 Rubenstein High energy nucleon-nucleon and deuteron-nucleus R interactions 1961/2 Sacharidis Angular dependence of polarization and positive pion

506

EJ production in proton-proton collisions at 1 GeV. 1962/3 Williams PG An investigation into the effect of final state interactions on the deuteron spectrum from proton-deuteron collisions at 991 MeV. 1963/4 Dodd PW The application of bubble chamber techniques to the study of pion production in neutron-deuteron collisions at less than 1 GeV. 1963/4 Homer RJ Investigation of two pion production in proton-nucleon collisions at 1 GeV. 1963/4 O’Dell AW Single and double pion production in proton-nucleon collisions at 1 GeV. 1963/4 Reading DH Some deuteron interactions at 650 MeV. 1964/5 Jobes M A study of neutron-proton in the energy range of 290 MeV to 970 MeV using a 9 inch liquid hydrogen bubble chamber. 1964/5 Khan QH Proton-proton interactions at 990 MeV. A Q differential iso- synchronous self-collimating Cerenkov counter. 1964/5 McMahon TJ Pion production in deuteron-deuteron and deuteron-nucleon interactions and the asymmetry in double proton- carbon scattering at high energies. 1964/5 Ruddick K The interaction of 650 MeV deuterons with deuterons and with complex nuclei and calibration of carbon as an analyser for 1 GeV polarised protons. 1964/5 Ryan DG Some interactions of 650 MeV deuterons.. 1965/6 Jafer JD Some elastic nuclear interactions at high energies. 1965/6 Tandon RK Deuteron stripping at 650 MeV 1966/7 Jones TW A Hall effect magnetic field stabiliser and a study of some nucleon-nucleon interactions at 1 GeV. 1966/7 Stiegelmair Deuteron stripping and pion production in nucleon-nucleon JA collision. 1967/8 Murray TA Proton-proton and proton-deuteron scattering at 991 MeV.

507

SOURCES

This thesis has drawn on a wide range of sources, both published and unpublished. Published sources are listed in the bibliography below. As regards unpublished sources, nearly all of these have come from a number of archives, listed below, and each identified in footnotes by an acronym. For the small number of documents not so drawn, location details are given in footnotes.

The archives identified by acronyms are as follows (in alphabetical order):

BUA (Birmingham University Archive): a small number of documents not included in the SCUB (see below).

CPCC (Chadwick Papers, Churchill College): these are collected papers of James Chadwick, held in the Archives Centre at Churchill College, Cambridge. They contain for example, documents relating to the Manhattan Project and correspondence with Oliphant.

EBA (Ellyard Biographical Archive): this collection, in the personal possession of the author, holds a diverse range of materials, most of which were provided to the author by Oliphant around 1980 to support the preparation of the Cockburn/Ellyard biography, and to provide the basis for further academic research on his career. The author has continued to hold these materials pending their transfer to the Oliphant Papers at the Barr-Smith Library at the University of Adelaide. EBA holds some original Oliphant correspondence, including with Rutherford, Cockcroft and Chadwick, drafts and typescripts of reports, memoranda and lectures and newspaper clippings, as well as transcripts of

508

interviews. Most of this material was previously held in Oliphant’s files at the ANU.

LPBL (Lawrence Papers, Bancroft Library): the papers of Lawrence are held in the Bancroft Library at the University of California, Berkeley. I have drawn on these for correspondence from Lawrence and others to Rutherford, Cockcroft and Oliphant from the 1930s onward and for documents relevant to Oliphant’s war time activities.

SCUB (Special Collections, University of Birmingham): these papers are held in the University of Birmingham Library, and have been drawn on to illuminate the 1937 to 1950 period during Oliphant which was the Poynting Professor and for some time after. Their coverage of the activities of the Physics Department during this period is extensive, apparently reflecting Phillip Moon’s interest in preserving archives, but they are relatively poorly organised. It appears that previous readers have left the materials confused and out of order. The majority are held in large cardboard boxes with little guidance as to what each box contains, though a catalogue exists. (There are some exceptions. Moon’s own papers are held separately within SCUB, and they are in good order.) Substantial re-organisation is needed. To aid location, box numbers are included in footnote references.

UCL (University of Cambridge Library): the majority of papers accessed from this source carry the marker CAV and deal with developments at the Cavendish Laboratory during the 1930s, such as the High-Tension Laboratory, the cyclotron and the Austin Bequest. Many of these papers are poorly organised and out of order.

UKNA (United Kingdom National Archives): we have drawn on this major repository at Kew in London mostly for materials dealing with

509

Oliphant’s wartime activities in radar and atomic energy, and in connection with the funding of the Birmingham proton synchrotron. The most-used series of files carries the identification code AB, which deal with MAUD and TA issues, though some use had also been made of files in the DSIR and PREM series.

Published Sources

Books, book chapters, monographs, conference proceedings

Allibone, TE 1984, ‘Metropolitan Vickers and the Cavendish’, in Hendry, J (ed.) Cambridge Physics in the Thirties, Adam Hilger, Bristol.

Andrews, PWS and Brunner, E 1955, The Life of Lord Nuffield: A Study in Enterprise and Benevolence, Blackwell, Oxford.

Boag, JW, Rubinin, PE and Schrenberg, D (eds) 1990, Kapitza in Cambridge and Moscow: Life and Letters of a Russian Physicist, North- Holland, Amsterdam.

Bowen, EG 1987, Radar Days, Adam Hilger, Bristol.

Bourdieau, P 2004, Science of Science and Reflexivity, University of Chicago Press, Chicago.

Brown, AC and MacDonald, CB 1977, The Secret History of the Atomic Bomb, Dial Press, New York.

Brown, L 1999, A Radar History of World War II: Technical and Military Imperatives, Institute of Physics Publishing, Bristol.

Brown, L, Pais, A and Pippard, P 1995, Twentieth Century Physics, Institute of Physics Publishing, Bristol.

Burcham, WE and Shearman, EDR 1990, Fifty Years of the Cavity Magnetron, School of Physics and Space Research at the University of Birmingham, Birmingham.

Chadwick, J 1984, ‘Some Personal Notes on the Search for the Neutron’ in Hendry, J (ed.) Cambridge Physics in the Thirties, Adam Hilger, Bristol.

510

Childs, H 1968, American Genius: the life of Ernest Orlando Lawrence, Dutton Press, Hialeah.

Clark, RW 1965, Tizard, Methuen, London.

Clark, RW 1972, A Biography of the Nuffield Foundation, Longman, London.

Cockburn, S and Ellyard, D 1981, Oliphant: the life and times of Sir Mark Oliphant, Axiom Press, Adelaide.

Cockcroft, JC 1984, ‘Recollections of Low Energy Nuclear Physics’ in Hendry, J (ed.) Cambridge Physics in the Thirties, Adam Hilger, Bristol.

Courant, ED 1985, The Early History of the Cosmotron and AGS at Brookhaven, Brookhaven National Laboratory (BNL-36659).

Courant, ED 1995, ‘The evolution of high-energy accelerators’, AIP Conference Proceedings, vol. 326, pp. 1-11

Crease, RP 1999, Making Physics: A Biography of Brookhaven National Laboratory, University of Chicago Press, Chicago.

Crowther, JG 1974, The Cavendish Laboratory 1874 -1974, Macmillan Press, London.

Crowther, JG and Whiddington, R 1947, Science at War, DSIR, London.

Dahl, P 2002, From to Nuclear Fission 1932- 1938, Institute of Physics Publishing, Bristol.

De Sola Price, D 1963, Little Science, Big Science, Columbia University Press, New York.

Edgerton, D 2006, Warfare State: Britain 1920-1970. Cambridge University Press, Cambridge

Eve, AS 1939, Rutherford: being the Life and Letters of Rt. Hon. Lord Rutherford OM, Cambridge University Press, Cambridge.

Foster, S and Varghese, M 1996, The Making of the Australian National University, 1946-1996, ANU Press, Canberra.

Frisch, O 1979, What Little I Remember, Cambridge University Press, Cambridge.

511

Galison, P 1997, Image and Logic: a Material Culture of Microphysics, University of Chicago Press, Chicago.

Galison, P and Hevly, B 1992, Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Gowing, M 1964, Britain and Atomic Energy, St Martins Press, New York.

Groves, LG 1962, Now it Can be Told, Harper and Row, New York.

Halpern, Paul 2010, Collider: Search for the World’s Smallest Particles, John Wiley and Sons, Hoboken.

Hartcup, G and Allibone, TE 1984, Cockcroft and the Atom, Adam Hilger, Bristol.

Heilbron, JL, Seidel, RW and Wheaton, BR 1981, Lawrence and His Laboratory: Nuclear Science at Berkeley, University of California Press, Berkeley.

Heilbron, JL and Seidel, RW 1989, Lawrence and his Laboratory: a History of the Lawrence Berkeley Laboratory, University of California Press, Berkeley.

Hendry, J (ed.) 2004, Cambridge Physics in the Thirties, Adam Hilger, Bristol.

Huddeson, L, Henricksen, P, Meade, R and Westfall, C 2004, Critical Assembly; a technical history of Los Alamos in the Oppenheimer Years, Cambridge University Press, Cambridge.

Hughes, J 1998, ‘Plasticine and Valves; Industry, Instrumentation and Emergence of Nuclear Physics’ in Gaudlliere, JP and Lowy, I (eds) The Invisible Industrialist: Manufactures and the Production of Scientific Knowledge. (Macmillian Press, London

Hull, P 2008, Lord Nuffield, Osprey Publishing, Oxford.

Jones, RV 1978, Most Secret War, Hamilton, London.

Kargon, R, Leslie, S and Schonberger, E 1992, ‘Far Beyond Big Science: Science Regions and the Organisation of Research and Development’ in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

512

Kollath, R 1967, Particle accelerators, Isaac Pitman and Sons, London.

Kowarski, L 1977, ’New Forms of Organisation in Physical Research after 1945’ in Charles Weiner (ed.) History of Twentieth Century Physics, Academic Press, New York.

Kragh, H 1999, Quantum Generations; a History of Physics in the Twentieth Century, Princeton University Press, Princeton.

Lawson, JD 1997, Early British Synchrotrons: an Informal History, National Technical Information Service, US Department of Commerce.

Livingstone, MS 1954, High-Energy Accelerators, Interscience Publishers, New York.

Livingstone, MS 1969, Particle Accelerators: A Brief History, Harvard University Press, Cambridge, Mass.

Miller, DP 2004, Discovering Water: Henry Cavendish, James Watt and the Nineteenth Century ‘Water Controversy’, Ashgate Publishing, Aldershot.

Moon, PB and Ibbs, TL 1981, Physics at Birmingham 1880 –1980, Department of Physics, University of Birmingham.

Nye, MJ 1996, Before Big Science: The Pursuit of Modern Chemistry and Physics 1800-1940, Twayne Publishers, New York.

Oliphant, ML 1967, The Genesis of the Nuffield Cyclotron and the Proton Synchrotron, Department of Physics, University of Birmingham.

Oliphant, ML 1972, Rutherford: Recollections of the Cambridge Days, Elsevier, London.

Panovsky, W 1992, “SLAC and Big Science: Stanford University” in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Peierls, R 1985, Bird of Passage: Recollections of a Physicist, Princeton University Press, Princeton.

Pestre, D and Krige, J 1992, ‘Some Thoughts on the Early History of CERN’ in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Price, A 1979, Instruments of Darkness, Granada Publishing, London.

513

Rhodes, R 1986, The Making of the Atomic Bomb, Simon and Schuster, New York.

Rolph, PM 1995, Birmingham Synchrotron 40th Anniversary Reunion: Proceedings of a one day symposium 16 September 1993, School of Physics and Space Research, University of Birmingham.

Rowe, AP 1948, One Story of Radar, Cambridge University Press, Cambridge.

Rutherford, Lord 1937, The Newer Alchemy, Cambridge University Press, Cambridge.

Seidel, R 1992, ‘The Origins of the Lawrence Berkeley Laboratory’ in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Sessler, A and Wilson, E 2007, Engines of Discovery: A Century of Particle Accelerators, World Scientific, New Jersey.

Smith, RW 1992, ‘The Space Telescope’ in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Steuwer, RH 1985, ‘Artificial Disintegration and the Cambridge-Vienna Controversy’ in Achinstein, P and Hannaway, O (eds) Observation, Experiment and Hypothesis in Modern Physical Science, MIT Press, Cambridge.

Traweek, S 1992, ‘Big Science and Colonial Discourse: Building High- Energy Physics in Japan’ in Galison, P and Hevly, B (eds) Big Science: The Growth of Large Scale Research, Stanford University Press, Palo Alto.

Walton, ETS 1984, ‘Personal Recollections of the Discovery of Fast Particles’ in Hendry, J (ed.) Cambridge Physics in the Thirties, Adam Hilger, Bristol.

Watson Watt, R 1957, Three Steps to Victory, Odhams Press, London.

Watson Watt, R 1959, The Pulse of Radar, Dial Press, New York.

Weiner, C (ed.) 1972, Exploring the History of Nuclear Physics, American Institute of Physics, New York.

514

Wilson, EJN 1997, Fifty Years of Synchrotrons, CERN network, viewed 10 December 2011, .

Wilson, D 1983, Rutherford: Simple Genius, Hodder and Stoughton, London. Wood, A 1946, The Cavendish Laboratory, Cambridge University Press, Cambridge.

Journal articles, conference proceedings.

Bleaney, B 2001, ‘Sir Mark (Marcus Laurence Elwin) Oliphant, 1901- 2000’, Biographical Memoirs of Fellows of the Royal Society, vol. 47, pp. 283-293.

Blewett, JP 1956, ‘The Proton Synchrotron’, Reviews of Progress in Physics, vol. 19, p. 37.

Boersma, K 2003, ‘Tensions within an Industrial Research Laboratory: The Philips Laboratory between the Wars’, Enterprise and Society, vol. 4, pp. 65-98

Bromberg, J 1993, Annals of Science, vol. 51, Issue 4, p. 441.

Burcham, WE 1947, ‘The Million-Volt Accelerating Equipment of the Cavendish Laboratory, Cambridge’, Nature, vol. 160, p. 316.

Burcham WE 1999, ‘The Cavendish High Tension Laboratory 1935- 1939’, Notes and Records of the Royal Society, vol. 53, no. 1, pp.121- 134.

Burcham, WE 1989, ‘Nuclear Physics in the United Kingdom 1911 - 1986’, Reports on Progress in Physics, vol. 52, no. 7, p. 823.

Burcham, WE and Isaak, GR 1996, ‘Phllip Burton Moon 1907-1994’, Biographical Memoirs of Fellows of the Royal Society, vol. 42, p .248.

Capshew, JH and Rader, K 2004, ‘Big Science: Price to the Present’, Osiris, 2nd series, vol. 7, pp. 2-25.

Carver JH, Crompton, RW, Ellyard, DG, Hibbard, LU and Inall, EK 2004, ‘Marcus Laurence Elwin Oliphant, 1901-2000’, Historical Records of Australian Science, vol. 14, no. 3, p. 337.

515

CERN Communication Group 2008, ‘What is the LHC?’, CERN FAQ, January 2008.

Cockcroft, JC 1964, ‘Science in the 1930s’, The Australian Physicist, vol. 1, pp. 99-104.

Courant, ED 1995, ‘The evolution of high-energy accelerators’, AIP Conference Proceedings, vol. 326, pp. 1-11.

Courant, ED 2003, ‘Accelerators, Colliders and Snakes’, Annual Review of Nuclear and Particle Science, vol. 53, pp. 1-27.

Crowley-Milling, MC 1983, ‘High energy particle accelerators’, Reports on Progress in Physics, vol. 46, pp. 51-95.

Dean, K 2003, ‘Inscribing Settler Science: Ernest Rutherford, Thomas Laby and the Making of Careers in Physics’, History of Science vol. 41, pp. 217-240.

Dowdell, JD, Griffith, AR and Viven WF 2011, ‘William Ernest Burcham CBE 1913-2008’, Biographical Memoirs of Fellows of the Royal Society, vol. 57, p.63.

Eve, AS and Chadwick, J 1936-38, ‘Lord Rutherford 1871-1937’ Obituary Notices of Fellows of the Royal Society, vol. 2, pp. 395-423

Farren, WS and Jones, RV 1961, ‘Henry Thomas Tizard 1885-1959’, Biographical Memoirs of the Royal Society, vol. 7, pp. 313-348.

Fuller EW and Hibbard LU 1954, ‘An accurate voltage integrator for magnetic field measurement’, Journal of Scientific Instruments, vol. 31, p. 36.

Gamow, G 1928, ‘Zur Quantentheone des Atomkernes’, Zeitschrift fur Physik, vol. 51, issue 3-4, p. 204.

Gamow G 1928a, ‘The quantum theory of nuclear disintegration’. Nature, vol. 122, no. 3082, p 205.

Glaser, DA 1952, ‘Some Effects of Ionizing Radiation on the Formation of Bubbles in Liquids’, Physical Review, vol. 87, no. 4, p. 665.

Gooden, JS, Jensen, HH and Symonds, JL 1947, ‘Theory of the Proton Synchrotron’, Proceedings of the Royal Society, vol. 59, p. 677.

516

Greinacher, H 1921, ‘Ueber eine neue Methode, Wechselstrom mittels elektrischer Ventiie und Kondensatoren inhochgespannten Gleichstrom zu verwandeln’, Zeitschrift fur. Physik, vol. 4, p. 195.

Hanbury Brown, R, Minnett, HV and White, FWG 1992, ‘ 1911-1991’, Biographical Memoirs of Fellows of the Royal Society, vol. 38, pp. 41-65.

Hibbard, LU 1950, ‘The Birmingham Proton Synchrotron’, Nucleonics, vol. 7, no. 4, p. 30.

Hibbard, LU 1954a, ‘The radio-frequency system of the Birmingham proton synchrotron’, Journal of Scientific Instruments, vol. 31, p. 363.

Hibbard, LU 1954b, ‘Electronic control of a synchronous motor’, British Journal of Applied Physics, vol. 4, p.147.

Hibbard, LU and Caro, DE 1953, ‘A stable source of high voltage’, Journal of Scientific Instruments, vol. 30, p. 378

Hibbard, LU, Caro, DE and Freeman, JY 1954, ‘A servo system for accurate speed control’, Journal of Scientific Instruments, vol. 31, p. 139.

Hibbard, LU, Randorf, W and Riddiford, L 1953, ‘A wide band, high power radio frequency amplifier’, Journal of Scientific Instruments, vol. 30, p. 245.

Hirst, EL 1951, ‘Walter Norman Haworth 1883-1950’, Obituary Notices of Fellows of the Royal Society, vol. 7, p. 372.

Hughes, J 2004, ‘Deconstructing the Bomb: recent perspectives on nuclear history’, British Journal for the History of Science, vol. 37, no. 4, pp 455-464.

Hughes, J 2010a, ‘Divine Right or Democracy? The Royal Society ‘revolt’ of 1935’, Notes and Records of the Royal Society, vol. 64, pp. 101-117.

Hughes, J 2010b ‘Introductory comments to the final discussion session’, Notes and Records of the Royal Society, vol. 64, no. Supplement 1, p. S173.

Kurti, N 1958, ‘Franz Eugen Simon 1893-1956’, Biographical Memoirs of Fellows of the Royal Society, vol. 4, pp. 225-256.

517

Lee, S 2007, ‘Rudolf Ernest Peierls 1907-1995’, Biographical Memoirs of Fellows of the Royal Society, vol. 52, pp. 265-284.

Livingstone, MS, Blewett, JP, Green, GK and Hayworth, LJ 1949, ‘Design Study for a Three-Bev Proton Accelerator’, Review of Scientific Instruments, vol. 21, no. 1, p. 7.

Livingstone, MS 1952, ‘High-Energy Accelerators’, Annual Reviews of Nuclear Science, vol. 1, pp. 169-174.

Lofgren, EJ 1950, ‘The Proton Synchrotron’, Science, vol. 111, p. 295.

Massey, H and Feather, N 1976, ‘James Chadwick 1881-1974’, Biographical Memoirs of Fellows of the Royal Society, vol. 22, pp. 11- 70.

McMillan, EM 1945, ‘The synchrotron – a proposed high energy particle accelerator’, Physics Review, vol. 68, pp.143-144.

Moon, PB, Riddiford, L and Symonds, JL 1955, ‘Experimental Characteristics of the Proton Synchrotron’, Proceedings of the Royal Society A, vol. 230, no. 1181, pp. 204-215.

Oliphant, ML 1928, ‘Selective Adsorption from Gaseous Mixtures by a Mercury Surface formed in the Mixture’, Philosophical Magazine, vol. 4, p. 423.

Oliphant, ML 1930, ‘The liberation of electrons from metal surfaces by positive ions’, Proceedings of the Royal Society A, vol. 127, p. 373.

Oliphant, ML, Shire, ES and Crowther, BM 1934a, ‘Disintegration of the Separated Isotopes of Lithium by Protons and by Heavy Hydrogen’, Nature, vol. 133, p. 377.

Oliphant, ML, Shire, ES and Crowther, BM 1934b, ‘Separation of the Isotopes of Lithium and some Nuclear Transformations Observed with Them’, Proceedings of the Royal Society, vol. 46, p. 922.

Oliphant, ML, Harteck, P and Rutherford, ER 1934, ‘Transformation effects observed with heavy hydrogen’, Proceedings of the Royal Society A, vol. 141, p. 692.

Oliphant, ML 1938, ‘The New High-Voltage Laboratory at Cambridge’, Nuovo Cimento, vol. 15, no. 3, pp. 160-166.

518

Oliphant, ML 1941, ‘The 184 inch Cyclotron at Berkeley, California’, Nature, vol. 148, no. 3763, p. 717.

Oliphant, ML, Gooden, JS and Hide, GS 1947, ‘The Acceleration of Particles to Very High Energies’, Proceedings of the Physical Society, vol. 59, pp. 666-677.

Oliphant, ML 1966, ‘The Two Ernests’, Physics Today, September 1966, p.35 (Part 1), October1966, p.43 (Part 2).

Oliphant, ML and Lord Penny 1968, ‘John Douglas Cockcroft 1897- 1967’, Biographical Memoirs of Fellows of the Royal Society, vol.14, p.139

Peierls, R 1981, ‘ 1904 -1979’, Biographical Memoirs of Fellows of the Royal Society, vol. 27, pp. 283-306.

Rabinovich MS 1976, ‘In memory of VI Veksler’, Soviet Physics Uspekhi, vol. 10, p. 12.

Ramm, CA 1955, ‘The principles and apparatus of the injection system of the Birmingham proton synchrotron’, Journal of Scientific Instruments, vol. 33, pp. 52-58.

Riddiford, L 1951, ‘The vacuum system of the Birmingham proton synchrotron”, Journal of Scientific Instruments’, vol. 28, pp. 47-57.

Riddiford, L 1953, ‘The importance of high vacua to high energy physics’, Vacuum, vol. 3, no. 3, pp. 232-244.

Rutherford, ER 1928, ‘Presidential Address, 30 November 1927’, Proceedings of the Royal Society A, vol.117, no. 777, pp. 300-316.

Rutherford, ER and Oliphant, ML 1933, ‘Experiments on the transmutation of elements by protons’, Proceedings of the Royal Society A, vol. 141, p. 259.

Seidel, R 1983, ‘Accelerating science: The post war transformation of the Lawrence Radiation Laboratory’, Historical Studies in the Physical Sciences, vol. 13, no. 2, pp. 375-400.

Sherrat, T 1993, ‘”A physicist would be well out of it”: George Briggs at the United Nations Atomic Energy Commission’, Voices (Autumn 1993).

Smith, WIB 1950,’Obituary of Dr JS Gooden’, Nature, vol. 166, no. 4212, p. 13.

519

Symonds, JL 1955, ‘Methods of measuring strong magnetic fields’, Reports on Progress in Physics, vol. 18, p 83.

Veksler, V 1944, ‘A new method of accelerating relativistic particles’,(in Russian) Soviet Physics Doklady vol. 43, p 329; (in English in 1945). Journal of Physics USSR, vol. 44, no. 9, p. 393.

Weinberg, A 1961, ‘The Impact of Large-Scale Research on the United States’, Science, vol. 143, no. 3473, p. 161.

Westfall, C 2003, ‘Rethinking Big Science: Modest, Mezzo, Grand Science and the Development of the Bevalac 1971-1993’, Isis vol. 94, no.1, pp. 30-56.

PhD Theses

Hughes, J 1993, ‘The Radioactivists: Community, Controversy and the Rise of Nuclear Physics’ A Dissertation submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy at the University of Cambridge. University reference PhD 18347.

Dean, K 2004 ‘Settler Physics in Australian and Cambridge 1850-1950’, University of Cambridge.

Oliphant ML 1929-30, ‘The Neutralisation of Positive Ions at Metal Surfaces and the Emission of Secondary Electrons’. Thesis submitted in 1929/30 for the degree of Doctor of Philosophy at the University of Cambridge. Abstract reproduced in Abstracts of Dissertations Approved for the Ph.D., M.Sc. and M.Litt Degrees in the University of Cambridge for the Academical Year 1929-1930, Cambridge University Press, 1931.

Yruma, JS 2009, How experiments are remembered: the discovery of nuclear fission 1934-1938. Thesis submitted for PhD at Princeton University. Reference DAI-A 69/12. June 2009

520