Laurence%2Cg.C

Home , Montreal Laboratory

Presentation of ANS Certificate of Recognition for George C. Laurence http://www15.pair.com/buchanan/laurence/present.htm

PRESENTATION OF THE

ANS CERTIFICATE OF RECOGNITION

OF THE ACCOMPLISHMENTS OF GEORGE C. LAURENCE

TO THE LAURENCE FAMILY

REMARKS BY

STANLEY R. HATCHER

PRESIDENT

ATOMIC ENERGY OF CANADA LIMITED

RESEARCH COMPANY

CHALK RIVER

1988 AUGUST 3

Mrs. Laurence, members of the family, distinguished guests, ladies and gentlemen:

Were it not for George Laurence, we would probably not be able to assemble here today on top of a reactor that embodies many of the attributes of the nuclear system that is the best in the world.

Robert Bothwell in his book on the history of AECL accurately writes, “…year after year, Canadian- designed heavy water reactors were at or near the top of the world ranking of capacity factors. Taken together, they marked the culmination of a thirty year cycle that stretched back to George Laurence's modest experiments in the basement of the NRC building …”. It is fitting that the first Canadian mentioned in that history is Dr. Laurence.

However, long before the publication of Bothwell's book, George Laurence himself wrote a spell-binding pamphlet on the “Early Years of Nuclear Research in Canada”. I'm sure most of you have read it and were as enthralled as I was, with the excitement of those early days.

Throughout his career, Dr. Laurence was at the forefront of ideas and developments. He was a pioneer in the dosimetry of radium and x-rays at NRC after he returned to Canada from his sojourn in Cambridge with Lord Rutherford. He helped to develop radiation safety regulations for North America, co-authoring the first bulletin of the Radiological Society of North America's Standardization Committee. By 1941 he was a member of the Royal Society of Canada.

Pivotal in the Canadian nuclear power story was his pioneering work in 1940 on nuclear chain reactions using a mixture of graphite and uranium oxide shortly after the discovery of fission. Foiled by lack of funds - a phrase that is still familiar to us today - he was unable to obtain uranium or graphite of satisfactory purity to continue his work. But he had established his credibility as a nuclear scientist. His studies enabled and encouraged Canada to enter into a scientific partnership with the distinguished group of British and European

1 of 4 8/13/2008 9:57 PM Presentation of ANS Certificate of Recognition for George C. Laurence http://www15.pair.com/buchanan/laurence/present.htm

scientists to from NRC's Montreal Laboratory Division, which eventually became AECL.

George Laurence joined that Montreal group in 1942 as the senior Canadian scientist. Hans von Halban, the first Director of the Anglo-Canadian project, described him as “a likeable and brilliant man”. With others, he assembled a sub-critical mock-up of the NRX lattice and measured some of the constants needed for the design of the reactor. He also worked on the instrumentation and control system of the NRX reactor. In 1945 he came to Chalk River.

Dr. Laurence had strong opinions - and he expressed them. One of his colleagues at the time notes that he was the one person who stood up to W.B. Lewis. Around 1950, when ideas for a distinctive Canadian nuclear power system were being developed, it was George Laurence who insisted that a non-breeder power plant suitable for Canadian requirements should be designed, based on the features proposed for NRU. That persistence paid off; the line that George Laurence consistently upheld won the day. The natural uranium reactor, cooled and moderated by heavy water, became the Canadian nuclear power system.

George was to be heavily involved in its evolution.

In 1952, as Director of the Chemistry and Engineering Division in the newly formed AECL, he was responsible for the conceptual design and detailed engineering of NRU; the reactor below us. Later he was to direct the reactor physics studies needed for the design of the first Canadian power reactors, NPD and Douglas Point.

He was the epitome of the professional scientist. Ask a colleague about his characteristics and invariably “integrity” is in the reply. He insisted on scientific rigour - and had an intuitive sense of what was right and what was scientific nonsense. He was wary of the unthinking use of computers, recognizing the ease with which users could be seduced into uncritically accepting results. One of his staff, presenting the results of some neutron scattering calculations performed on the latest Chalk River computer, was devastated to be peremptorily told that they were way out - they just did not look right. Sure enough, it was discovered that incorrect data had been used.

He was a perfectionist, not only in getting the science right but in reporting in correctly. A colleague describes the despair of George's associates as draft succeeded draft of any report, long, they felt, after the best version had been attained. And that was in the days before word processing!

In a series of interviews that Dr. Laurence gave late in his life, he recalled the early Chalk River days. “It was an exciting period, and we knew it would eventually lead to extremely important applications. We were all young, and there was an air of keenness and enthusiasm that made life here interesting and pleasant.” His reflections on how success was achieved point to the strength of the laboratories then - and now. “It is the bringing together of ideas from many sources, and many little ideas, many contributions to the solution of the same general problems, that brings about advances. Years later, when you try to think back to who had that bright idea, you can't - it simply evolved out of discussions. Most of the progress in this science has been through the collective conception of new ideas.”

George's colleagues recall that meeting were memorable occasions. To his associates in those early days he was “Mr. Nuclear Physics”. One colleague notes that George would add to the excitement by pacing up and down as the discussion proceeded. As he became more and more wound up, the pacing got faster and faster. Another recalls that he had a habit of conducting meetings with his shoes off.

Dr. Laurence's approach to reactor safety evolved from 1944 when he was first presented with the question of safety in the NRX reactor. With his continuing involvement, the lessons learned from the NRX accident in 1952 lead to many improvements being incorporated into the NRU design, operation, and means of control. He was the natural person to be appointed Chairman of the Reactor Safety Advisory Committee, set up by

2 of 4 8/13/2008 9:57 PM Presentation of ANS Certificate of Recognition for George C. Laurence http://www15.pair.com/buchanan/laurence/present.htm

the Atomic Energy Control Board to advise on the health and safety of nuclear reactors and power stations.

In 1961, George left AECL to become the President of the Atomic Energy Control Board. Although a natural progression in the technical sense, it was not an easy move for George Laurence to make. He did not like Ottawa bureaucracy, and it was not until 1965 that he physically moved his office to Ottawa. A colleague recalls that, after a mandatory senior executive course some two years after he was appointed, he returned incensed by the management jargon to which he had been exposed. It completely reinforced his distaste for the bureaucracy. He remained a scientifically-oriented president.

Under his leadership, the Canadian reactor safety philosophy developed in quite a different way from that in many other countries. In the Canadian approach, the regulatory authority sets the limits for doses to the public and workers, and leaves it to the owner to convince the regulator that the plant will operate within these limits. We, as designers and operators, are convinced that his approach is the best and is a fundamental reason for the excellent safety record of Canada's nuclear power plants.

Mrs. Laurence, all of us associated with nuclear science and technology in Canada owe a great deal to George. Indeed, his influence has spread far beyond the bounds of Canada. I am very pleased to be able to tell you that the Nuclear Reactor Safety Division of the American Nuclear Society is proposing to introduce an award to be named the “George C. Laurence Pioneering Award” for achievements in the field of nuclear safety. This will be a continuing celebration of the high regard in which George Laurence is held by his peers.

He received many awards during his career. Sadly, this award must be posthumous. I was very privileged to be asked by the American Nuclear Society to present you this plaque that recognizes his pioneering achievements. The citation reads:

“The American Nuclear Society

Certificate of Recognition of the Accomplishments

of George C. Laurence”

“This award is made to confer recognition of a lifetime of achievements in the development of safety philosophy. Over a career spanning 40 years, his pioneering leadership in the area of nuclear reactor safety led to development of the safety concept of numerical safety goals based on risk. His safety philosophy and principles for nuclear power plants, first expressed in the late 1950s, are the basis for the Canadian licensing approach applied to CANDU reactors and have served as the impetus for safety developments in many other countries. Major aspects of these principles include separation of safety and operating systems, two-out-of-three logic for safety systems, and the first numerical safety goals for significant radiological releases due to an accident. During his years of leadership at the Atomic Energy Control Board, he created a climate in which innovative approaches to safety could be developed in ways that enhance both safety and operational efficiency.

This award recognizes his pioneering achievements.”

Signed by:

L. Walter Deitrich

Chair, Executive Committee

Roger W. Tilbrook

3 of 4 8/13/2008 9:57 PM Presentation of ANS Certificate of Recognition for George C. Laurence http://www15.pair.com/buchanan/laurence/present.htm

Chair, Honors and Awards Committee

June 13, 1988

Mrs. Laurence,

Please accept this citation from the American Nuclear Society, a citation that is sincerely endorsed by his friends and colleagues here in Canada.

Home

4 of 4 8/13/2008 9:57 PM CPI.Q (Canadian Periodicals) Print http://find.galegroup.com.myaccess.library.utoronto.ca/itx/printdoc.do?co...

OBITUARY Nuclear researcher was pioneer in field George Craig Laurence. (News). Globe & Mail (Toronto, Canada) (Nov 9, 1987): pD13. (180 words)

DEEP RIVER, Ont. -- Canadian Press DEEP RIVER, Ont.

George Craig Laurence, a physicist and pioneer of nuclear research in Canada, has died. He was 82.

Mr. Laurence, who joined the National Research Council in 1930 to study radium and X-rays, died on Friday in his sleep.

He played a major role in developing Canada's nuclear capabilities, but much of his career focused on the safety of nuclear energy.

He served briefly in the post-war years on the United Nations Atomic Energy Commission, which was formed after the atomic bombs were dropped on Japan. The commission's role was to consider problems arising from the development of nuclear energy and its devastating power.

Mr. Laurence then became the head of reactor research and development at Atomic Energy of Canada Ltd., which developed the Canadian Candu nuclear reactor.

From 1961 until his retirement in 1970, Mr. Laurence served as president of the Atomic Energy Control Board of Canada, which regulates and supervises nuclear development in Canada.

He leaves his wife, Freda, and a daughter, Patricia Buchanan, who lives in the United States.

Source Citation: "OBITUARY Nuclear researcher was pioneer in field George Craig Laurence. (News)." Globe & Mail (Toronto, Canada) (Nov 9, 1987): D13. CPI.Q (Canadian Periodicals) . Gale. UNIVERSITY OF TORONTO. 13 Aug. 2008 .

Gale Document Number: A165137914

1 of 1 8/13/2008 10:18 PM

Letter to editor Nuclear critics distort hazards, says former atomic official

The campaign against nuclear power began years ago with the falsehood that nuclear power stations might explode like nuclear bombs. It ascribed without supporting evidence upward fluctuations in infant mortality to leakages of radioactivity from nuclear power stations, while ignoring the downward changes. It grossly exaggerated the risks of lung cancer and genetic harm from the radon pollution in Port Hope cellars. It implied that uranium miners today are exposed to as dangerous concentrations of radon in the mines as they were a few years ago. It implied that the continuation of research to develop and improve the technology for disposal and storage of nuclear wastes was proof that the industry does not know how to do it safely.

It has grossly exaggerated both the chances and the possible consequences of nuclear reactor accidents such as bursting heat transport pipes, and pretended that the action of the Energy Control Board in reducing the permissible operating power of certain power stations until recently discovered faults were corrected showed that there was imminent threat of a horrible disaster. It made the outrageous exaggerations that we might have lost Deep River if certain equipment failed in the NRX accident 26 years ago and that hundreds risked their lives during decontamination following an accident with an irradiated fuel rod from NRU a few years later.

Many more examples can be given of the untruthfulness, misinterpretation and exaggeration in the campaign to destroy the nuclear power industry. Its spokesmen have repeatedly disparaged the judgment and maligned the sincerity of distinguished scientists and experienced officials who do not share their prejudices.

Before the public allows itself to be persuaded by the foes of nuclear power to commit its grandchildren to extremely distressing energy shortages, it should know who they are, and what background of long training, study and experience in relevant sciences and technologies qualifies them to instruct the public on the possible hazards of nuclear energy. In 36 years of participation in the Canadian nuclear program before I retired, I knew well most of the engineers and scientists that played leading roles, but I never heard of any of these anti-nuclear critics in Canada until they attracted attention by dogmatic and sensational misrepresentation. Frequent misuse of scientific words and errors reveals that many of them would probably not recognize a nuclear reactor if they saw it, and know almost nothing about radiation and its behavior. They have no conception of the practical difficulties and probable costs of propelling cars and planes and driving machines of industry by sunshine.

The public depends on the news media for reliable information on questions of national importance. The media can serve the public better if they choose their sources of fact and opinion on nuclear energy more wisely. George C. Laurence (formerly President of the Atomic Energy Control Board) Deep River

Document glob000020011126da8h088ui

Letter to Editor Nuclear safety

The Harrisburg nuclear power-plant accident killed no one. Neither did the accidental destruction of the NRX reactor in 1952 nor the Windscale (England) reactor in 1957. Tens of thousands are killed in highway traffic accidents, and hundreds of thousands die by malnutrition every year. Have the media lost all sense of proportion?

Anti-nuclear extremists excuse this inconsistency by claiming that thousands might have been killed at Harrisburg. That claim is not justified.

The suggestion that thousands might die in a nuclear power plant accident was made 22 years ago in a U.S. Atomic Energy Commission report. The report failed badly in judging the danger to the public from a nuclear power station accident because its authors had little idea of how much radioactive material could escape in a form that could be carried far by the wind.

Much has been learned since the report was written - through research and the actual experience of very destructive accidents to nuclear reactors - about the behavior of different kinds of radioactive waste products in an accident and the risks of radiation sickness, cancer, and other dangers.

The worst credible accident to a CANDU nuclear power station, assuming simultaneous failure of protective measures and containments, would be unlikely to kill as many as a single bad air crash. And 20 years of nuclear power-station operation has shown that a comparable nuclear accident is much less likely to occur.

Precautions would be needed, of course, to avoid additional harm after the contaminated air had blown away. The supply of milk and vegetables from contaminated farms might have to be controlled for a few weeks or months. Those living in specified areas might be advised to avoid eating uncovered food, and to wash clothes, furniture and dishes on which radioactive dust might have settled. Large-scale evacuation of the public would not be required, although the occupants of any exceptionally badly contaminated homes might be asked to vacate them temporarily. The greatest and most costly task would be the removal and disposal of the huge amounts of radioactive material in the damaged reactor building, but it would be accomplished without unwarranted overexposure of those involved.

As in other industries, there will be many breakages, leaks and other failures of equipment that injure no one. The Atomic Energy Control Board requires that these be reported to it and repaired promptly. They do not, however, imply great imminent danger, as some would like the public to believe. George C. Laurence (Former president of Atomic Energy Control Board) Deep River

Document glob000020011126db6406b6d

Early Years of Nuclear Energy Research in Canada

by George C. Laurence

Published by Atomic Energy of Canada Limited, May 1980 Aussi disponible en français .

About the author

Given more time, more assistance, more funds and purer materials, the first man-made nuclear chain reaction might have been achieved in Canada. George C. Laurence made the first Canadian experimental attempt in 1940-42.

Born at Charlottetown in 1905, Dr. Laurence was educated at Dalhousie and Cambridge universities. He joined the staff of the National Research Council of Canada (NRCC) in 1930 and became active in improving the measurement of radiation dosage in the treatment of cancer and in promoting safety from radiation exposure.

He has been involved in nuclear energy development in Canada since its beginning, continuing in the Montreal nuclear energy laboratory in 1943-44 and at the Chalk River Nuclear Laboratories. He directed the staff that did the preparatory research and development and the conceptual design of the NRU reactor.

In 1946-47 he served as scientific advisor to the Canadian delegation to the United Nations Atomic Energy Commission in New York. In 1956 he was appointed chairman of the Reactor Safety Advisory Committee set up by the Atomic Energy Control Board (AECB) to advise on the health and safety aspects of nuclear reactors and power stations. In 1961 he left AECL to become president of the AECB from which he retired in 1970. Since then he has lived at his home in Deep River, Ontario.

He was awarded the MBE for his scientific work during the war, the Canadian Association of Physicists medal for achievement in physics in 1966, the W.B. Lewis medal from the Canadian Nuclear Association in 1975, and a number of honourary degrees.

Note: Links to web sites about some of the people in this document are made at the first mention of the person, and in the appendix.

Introduction

Forty years ago the future role of Canada in nuclear research and development would have seemed impossible. There was little public confidence in our scientists and development engineers. Technology was something to import.

Research in the National Research Council of Canada (NRCC) laboratories has always been predominantly "mission-oriented", assisting industry and other government departments. Work of immediate and obvious

1 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

application left little time for "purely curiosity-inspired" research. By 1940, the attention of the laboratories had been shifted almost entirely to war problems.

This isolation from basic research made it more difficult for the scientists to keep in touch with progress that was relevant to their work, but the scientific periodicals told us of the discovery of nuclear fission, that it sometimes happens when uranium atoms are hit by neutrons, and that more neutrons are released by fission than are captured to produce it. Therefore, a reaction might be possible in which the supply of free neutrons was constantly replenished, or even greatly increased. A very large number of fissions produced in such a reaction would release a large amount of energy. There was evidence to show that a large increase in the rate of producing fissions would be easier to accomplish if the neutrons were moving slowly. They would move more slowly if they encountered large numbers of very light atoms such as hydrogen atoms; therefore, it might be advantageous to associate with the uranium a suitable quantity of material containing hydrogen, such as ordinary water.

One report told of the attempt by Joliot, Halban and Kowarski in Paris to prove that a large release of energy by fission might be possible in a solution of a uranium compound (such as uranyl nitrate) in water. Their attempt failed because the ordinary hydrogen atoms, though they slowed the neutrons, captured so many of them that an insufficient number remained to produce more fissions.

They decided to try heavy water with the uranium, instead of ordinary water, because heavy hydrogen would capture fewer neutrons than ordinary hydrogen. Before they could do so, however, the German Panzer divisions by-passed the Maginot Line and advanced across France. Halban and Kowarski escaped to England bringing their heavy water with them, and did the experiment in Cambridge.

By this time, the nuclear scientists in England and the United States had stopped publishing the results of their research, but they continued their work in secrecy. We assumed that the German scientists were also hard at work, and we were convinced that if Germany produced a nuclear weapon first they would win the war.

Experiments in Ottawa

Heavy water was scarce and costly to produce. The 185 kilograms, that the French scientists had obtained from a hydroelectric plant in Norway and brought to England, was most of the world's supply. Rough calculations with the inaccurate data then available suggested that it might be possible to obtain a large release of energy using some form of carbon, instead of heavy water, with the uranium. Carbon would be less suitable for the purpose but was cheaper and easier to obtain. I decided to experiment with carbon and uranium oxide. The experiment would have to be done mostly in overtime because my small section was very busy assisting Canadian industry to become proficient in the radiographic inspection of parts for military aircraft and other equipment. Months later, I learned without surprise that similar experiments with carbon and uranium had been started both in England and the United States at about the same time.

The purpose of the experiment was to determine whether a very large release of nuclear energy would be possible in a large bulk of the kinds of uranium and carbon which I had. It would be possible if at least as many neutrons were released by fission as were captured. That implied that if an independent source of neutrons is surrounded by a small quantity (i.e. a few tonnes) of the combination of uranium and carbon, more neutrons would reach the surrounding walls than if the combination of materials was not present.

2 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Cutaway drawing of the Ottawa sub-critical assembly, 1941-42

In our experiments in Ottawa to test this, the source of neutrons was beryllium mixed with a radium compound in a metal tube about 2.5 centimetres long. Alpha particles, emitted spontaneously from the radium, bombarded atoms of beryllium and released neutrons from them. The carbon was in the form of ten tonnes of calcined petroleum coke, a very fine black dust that easily spread over floors, furniture and ourselves. The uranium was 450 kilograms of black oxide, which was borrowed from Eldorado Gold Mines Limited. It was in small paper sacks distributed amongst larger paper sacks of the petroleum coke.

The sacks of uranium and coke were held in a wooden bin, so that they occupied a space that was roughly spherical, 2.7 m in diameter. The wooden bin was lined with paraffin wax about five centimetres thick to reduce the escape of neutrons. The arrangement is shown above, as a sectional view through the bin and its contents.

A thin wall metal tube supported the neutron source at the centre of the bin, and provided a passage for insertion of a neutron detector which could be placed at different distances from the source. In the first tests the detector was a silver coin, but in most of the experiments it was a layer of dysprosium oxide on an aluminum disc.

The experimental routine was to expose the detector to the neutrons for a suitable length of time, then remove it quickly from the assembly and place it in front of a Geiger counter to measure the radioactivity produced in it by the neutrons. The Geiger counter tubes and the associated electrical instruments were homemade because there was very little money to spend on equipment.

The relative rates of neutron capture and neutron release by fission were calculated from the data obtained. If the release had been greater than the capture it would have been possible to estimate the "critical quantity"

3 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

of uranium and coke, that is the minimum quantity needed to produce a self-sustained reaction that would release a large amount of nuclear energy.

Prof. B. Sargent of Queen's University joined me in these experiments during the summer university vacations of 1941 and 1942. Progress was slow because the work was interrupted by other duties and we lacked the better equipment that would be available today.

By late summer in 1942, our measurements had shown that the release of neutrons by fission in our combination of materials was a few percent less than the capture. Therefore, it would not be possible to obtain a large release of nuclear energy in that combination of materials even if large quantities were used. There was too much loss of neutrons by capture in impurities in the coke and uranium oxide and in the small quantities of paper and brass that were present. We did not then realize how a little impurity could lead to failure.

Meanwhile in the United States, E. Fermi, H.L. Anderson, B. Field, G. Weil and W. Zinn, after a first attempt that was also unsuccessful, did succeed in showing that a large release of energy would be possible using purer uranium and very pure carbon in the form of graphite. Using the necessarily larger quantities, the Americans then built the first nuclear reactor and operated it on December 2, 1942. They called it an "atomic pile".

In the summer of 1940, R.H. Fowler visited Ottawa, followed soon by J.D. Cockroft . They had been to the United States to stimulate greater American interest in research of military importance. They told me about the nuclear energy research in England and that in the United States which they had just seen.

With Prof. Fowler's introduction, I visited L.J. Briggs, who was chairman of the committee that coordinated the American nuclear energy research at that time, and also J.B. Conant, E. Fermi, H.C. Urey and P.H. Abelson and learned of their work. After my visit, we received in Ottawa copies of reports on the American nuclear energy research for the next two years. One of them that was particularly helpful was "A Study Concerning Uranium as a Source of Power" by J.B. Fisk and W. Shockly, dated September 17, 1940, a remarkable theoretical discussion of the feasibility of a nuclear reactor to have been written so early.

In response to Cockroft's suggestion when he returned to England we received a gift of $5,000 from Imperial Chemical Industries, which was involved in the nuclear research in England, in support of our experiment. It was an important addition to our budget, but I valued it most as an expression to Dr. Mackenzie of British confidence in our work.

Joint British-Canadian Laboratory Established in Montréal

At first there was greater progress in Britain than in America. It was predicted that if the uranium-235 was separated from the natural uranium in which it occurs it could be used as a very powerful explosive for military purposes. The British estimated the quantity that would be needed to make a bomb, and they proposed methods for the separation of uranium-235. It was shown by theory and experiment that the release of nuclear energy from natural uranium would be accompanied by the production of plutonium, and that plutonium like uranium-235, was fissionable. They also estimated the quantity of plutonium that would be needed for a bomb.

Halban and Kowarski , then in England, carried out their experiments that they had been prevented from doing in France, with a uranium compound dissolved in heavy water, and concluded that a great release of nuclear energy might be possible with a much larger quantity of such a solution. At first, the interest in a "boiler", as they called it, using uranium with heavy water or carbon was based on the hope of using it as a source of energy for industrial purposes. With the discovery that plutonium was fissionable the boiler had

4 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

added importance; it could be used to produce plutonium for the construction of a bomb.

It was suggested that the nuclear research in Britain should be moved to the United States. At first the idea was more acceptable to the American scientists that to some of those in Britain. By the spring of 1942 the advantages to Britain of moving the scientists from there to this continent were more obvious because the industrial resources and effort that would be needed to produce nuclear weapons were already committed heavily to other war purposes. In the United States, however, the work had become well advanced, and the help that might come from England no longer seemed so valuable. The Americans felt that it was too great a security risk because the senior members of the group, that would be sent from England, included refugees from countries that were occupied by the Germans and it was thought that they might be influenced by pressure on their relatives in Europe.

The British then suggested that a joint British-Canadian laboratory be established in Canada for nuclear research leading to the construction of a pilot plant for the production of plutonium. It would be staffed by some of the scientists from Britain directed by H.H. Halban, with Canadian scientists recruited into the project. Vannevar Bush, Chairman of the U.S. National Defence Research Committee, informed Sir John Anderson, Home Secretary of the U.K., that the Americans would accept this arrangement and agree to the exchange of information on research that was relevant to the design of the pilot plant.

On February 19, 1942 Malcolm Macdonald , the British High Commissioner to Canada, with Sir George Thomson and W.A. Akers called on Dr. Mackenzie to discuss this proposal. Later that day, Dr. Mackenzie introduced them to the Hon. C.D. Howe and acquainted him with the proposal. Sir George Thomson, Malcolm Macdonald and Professor R.E. Peierls discussed it again on June 15th, with Mackenzie and Howe, and also with Canadian Prime Minister William Lyon Mackenzie King. Dr. Mackenzie visited Vannevar Bush in Washington and they discussed the possibilities of American cooperation with the proposed British- Canadian laboratory. The British raised the question again on August 17th and on September 2nd.

The objective was speculative; it was doubtful if it could be completed before the end of the war. It would divert scientists, equipment and material from other war work, and would commit Canada to the expenditure of many millions of dollars, and there would be difficulties in the procurement of materials, particularly the many tonnes of heavy water.

Dr. Mackenzie said later that the deciding consideration was that when peace returned atomic energy would be bound to have applications of social and economic significance far beyond the possibilities of imagination and prediction, and the proposed Canadian-United Kingdom research effort would provide an opportunity for the training of Canadian scientists in this field. So, Mr. Howe agreed on September 2, 1942 that Canada would receive scientists from England, provide the laboratory facilities and supplies and administer the project in Montréal as a division of the NRCC.

The first of the staff from England arrived about the end of the year 1942. They were P. Auger and B. Goldschmidt of France, G. Placzek of Czechoslovakia, S.G. Bauer of Switzerland, H. Paneth and H.H. Halban of Austria and R.E. Newell and F.R. Jackson of Great Britain. We temporarily occupied an old residence at 3470 Simpson Street belonging to McGill University. Three months later, we moved into a 200 square metre area in the large, new building of the University of Montréal, and more scientists and technicians arrived from England. It was part of my job to recruit Canadian staff, and Professor David Keys (C) , representing the Wartime Bureau of Technical Personnel, helped greatly in finding them, and in obtaining approval for their transfer from other employment. The staff grew quickly to over three hundred, of whom about one half were Canadian. The names of the Canadians as they appear in the text will be followed by (C).

The project was started in a mood of enthusiasm and expectation of great scientific adventure. Never before had such a talented group of scientists been brought together in Canada with a single purpose.

5 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Some of the initial research staff of the Montréal Laboratory, 1943.

Standing: A.M. Munn (C), B.L. Goldschmidt , J.W. Ozeroff (C), B.W. Sargent (C), G.A. Graham (C), J. Guéron , H.F. Freundlich, H.H. Halban, R.E. Newell, F.R. Jackson, J.D. Cockroft (visiting the laboratory), P. Auger, S.G. Bauer, N.Q. Laurence, A. Nunn May . Seated: W.J. Knowles (C), P. Demers (C), J.R. Leicester, H. Seligman, E.D. Courant, E.P. Hincks (C), F.W. Fenning, G.C. Laurence (C), B. Pontecorvo , G.M. Volkoff (C), A. Weinberg (U.S. Liason officer), G. Placzek .

Soon, however, we became quite impatient as we waited for the expected close collaboration with the American scientists to develop. American anxiety about the security of information increased on account of the mixed national background of the Montréal team. They proposed that the exchange of information with the Montréal Laboratory become much more restricted, and they stopped sending us copies of their scientific reports, and deeply offended the British. The administration of their security measures became very tight after the U.S. Army assumed control of the whole American nuclear program in June 1942, but it was the senior American scientists, Vannevar Bush and J.B. Conant, who resisted most strongly close collaboration with the scientists in Montréal.

The choice of Halban as Director of the Laboratory had seemed logical, but turned out to be unfortunate. He was involved in the unhappy circumstances that brought bitterness and distrust into the relations regarding nuclear energy between France and the United States. With Joliot, Kowarski and Perin, he held patents which claimed control of the use of nuclear reactors for some of their most important applications. The interest of Imperial Chemical Industries in these patents provoked the American distrust of International cartels. An ill-timed visit by Halban to Joliot in France to discuss the patents aggravated American worries about security because Joliot was a member of the French Communist Party.

His associations with the NRCC also did not always run smoothly. He was impetuous and vacillating in decisions and unreasonable in his demands of the administrative staff in Ottawa and unfair in critisising them. He failed to inform Dr. Mackenzie or any other Canadian about important decisions regarding the research program of the laboratory. Eventually, E.W.R. Steacie (C) , then the Director of the Chemistry division of the NRCC laboratories, acted as a part-time Assistant Director under him, but Dr. Steacie's

6 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

influence came too late and was too remote from the administrative problems in Montréal.

At the Québec Conference on August 17, 1943, Britain's Prime Minister Churchill discussed with American President Roosevelt the very unsatisfactory state of British-American collaboration in nuclear energy research, including the work of the Montréal Laboratory. They agreed that arrangements should be made "to ensure full and effective collaboration between the two countries in bringing the project to fruition". A "Combined Policy Committee", under the chairmanship of H.L. Stimson, the U.S. Secretary of War, was named to work out the basis for implementing this agreement, but it rarely met. There was little progress towards effective collaboration, excepting a few scientific discussions on strictly limited topics. Morale in the Montréal Laboratory became very low. The scientists felt they would be better employed in other work.

Good Research in Canada, 1943 - 1946, Notwithstanding Difficulties

In spite of the discouragement and the uncertainties of purpose which lasted until the end of 1944, a remarkable amount of good research was done in the Montréal Laboratory during that period as well as in the happier months that followed. It provided scientific data that were needed for the design of a fission reactor and of chemical plants for the extraction of plutonium and uranium-233 produced in the reactor, and it contributed to knowledge and understanding of the structure and behaviour of atomic nuclei.

Part of the research was investigation of the penetration of neutrons through the kinds of materials that might comprise the core of the reactor, such as heavy water, ordinary water and graphite, and combinations of these with uranium and other materials. This information was essential for the design of a reactor because a large release of nuclear energy by fission is possible only if sufficient numbers of the neutrons that are liberated by fission retain their freedom, avoiding capture in the wrong kinds of atoms, until they are captured again in fissionable atoms to produce more fissions.

The nuclear fission process in heavy water.

There was particular interest in the possibility of a reactor that would use natural uranium and heavy water. The Americans doubted the conclusion from the experiment by Halban and Kowarski in Cambridge

7 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

University that a reactor would be feasible using a solution of uranyl nitrate in heavy water because it seemed improbable theoretically and the experiment was too difficult to do accurately. The heavy water had been brought to Montréal, and F. Fenning repeated the experiment there and approximately corroborated its conclusions. For further confirmation, he investigated in particular the proportion of slow ("thermal") neutrons captured in the solution that produced fission, and B. Pontecorvo measured the proportion of fast neutrons that were captured before their spped was reduced sufficiently to produce most of the fissions. Halban suggested that the theoretical objections to the experimental results might be too pessimistic because they overlooked the possibility of an "n-2n effect" that would increase the supply of neutrons, i.e. a reaction by which the capture of a neutron in a deuterium atom of the heavy water would result in the release of two neutrons. A. Nunn May , L.G. Elliot (C) and E.P. Hincks (C), however, showed that any n-2n effect would be too infrequent to explain the discrepancy. It remained a mystery.

In any case, it was clear that the feasibility of a sustained nuclear reaction in a "homogeneous" mixture of uranium in heavy water would be greater if the uranium concentration was considerably increased. Unfortunately no suitable compound of uranium was known that could be dissolved in heavy water in sufficient concentration for the purpose. Hence, interest turned to a more concentrated mixture: a "slurry" consisting of a suspension of a finely divided compound of uranium in heavy water which was called "mayonnaise" on account of its appearance. It was intended that the mayonnaise would circulate through the reactor and through a heat exchanger where the heat produced by the nuclear fission would be extracted. It could be pumped directly from the reactor to a chemical plant where the used uranium could be replaced, fission products could be removed and plutonium recovered. H. Paneth, B.L. Goldschmidt , H.G. Heal and F. Morgan investigated the preparation of such a suspension, its stability at high temperatures and physical behaviour and the effects on it of the intense radiation to which it would be exposed in a reactor. D.W. Ginns considered the engineering design of the reactor. S.G. Bauer and W.J. Knowles (C) investigated problems of removing heat from the slurry. G.S. Anderson and J.H.L. Matheson (C) worked on the design of suitable heat exchangers. The possibility of adding extracted uranium-235 to increase reactivity was considered.

Interest gradually shifted from "homogeneous reactors" using a slurry to "heterogeneous systems" in which the uranium was in the form of metal bars about 2.5 cm in diameter and several centimetres apart in heavy water. There were several reasons for the change. It was realized that much less heavy water would be needed. Experiments by A.G. Maddock, N. Miller, Miss G. Gorey (C) and J. Hebert (C) gave warning that radiation from uranium and fission products suspended in the water would decompose the heavy water into oxygen and deuterium so rapidly that the reactor could only be operated at a very low power. Other practical difficulties were expected.

Meanwhile, the possibility that the reactor to be built in Canada might use graphite instead of heavy water was not rejected as long as American collaboration remained in doubt. Early success with heavy water would depend on its supply and the Americans controlled the only important production on the continent, that at Trail, B.C., by a supply contract. Sargent and I, with Hans Paneth and H.G. Hereward, and later A.M. Munn (C), measured the migration of neutrons in a ten-tonne pile of graphite rods as they slowed from their original high velocities and continued to move at slow ("thermal") velocities through the graphite. The neutrons for the experiment were released from a block of beryllium by exposing it to the radiation from a two-million volt therapy X-ray machine. We used the same techniques to measure the diffusion of slow neutrons (the "thermal diffusion length") in heavy water containing lithium carbonate.

As I became involved in other research, Sargent directed further experiments on neutron migration in heavy water containing uranium bars in arrangements that might be used in a practical reactor. Associated with him in these other experiments on the motion of neutrons in heavy water were D.B. Booker, P.E. Cavanaugh, H.G. Hereward and N.J. Niemi (C). These experimental studies of the motion of fast and slow neutrons in graphite and in heavy water confirmed theories of their motion in these materials developed by theoretical physicists, thereby making it possible to use their theories with greater confidence in designing a reactor.

8 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Other kinds of reactors were also considered. For example, S.G. Bauer was interested in "breeder reactors" which would replenish their fuel supply by producing plutonium from uranium-238 or uranium-233 from thorium. M.H.L. Pryce , E. Courant and B. Pontecorvo studied the possibility theoretically. Pontecorvo, B.H. Flowers, G.A. Graham (C) and H. Seligman also investigated the motion of neutrons in ordinary water containing bars of uranium, lead or iron to obtain information that would be relevant to the design of reactors that use enriched uranium and ordinary water, such as those now operating in the United States.

The theoretical physicists studied such questions as the minimum quantities of uranium and heavy water or graphite that would be required in a reactor, the saving in these quantities that could be made by surrounding the reactor with a thick layer of graphite to impede the escape of neutrons from it, the effectiveness of moveable bars containing substances that capture neutrons, such as boron or cadmium, as a means of stopping or controlling the fission reaction, and other problems of design.

For the purpose of calculating such aspects of reactor design, G. Placzek, R.E. Marshak, B. Davison, E. Courant, J.C. Mark (C), F.T. Adler and others produced formulae which described the migration and distribution of neutrons by "diffusion" equations similar to those which are used to describe the flow of heat, sound, matter in solution or radiation in stars. They used "transport theory" (describing the motion of neutrons in vector notation) to show how the quantities in these formulae were related to information that could be obtained by measurements of the average distance between collisions with atoms, the changes in direction and speed caused by the collisions, and the risk of their capture by the atoms which they encountered. They also used transport theory to calculate the net flow of neutrons from one material (e.g. heavy water) into another (e.g. uranium or other metal) where the surface between the two materials was not flat, a frequent problem in designing reactors. G. Volkoff (C), Jeanne LeCaine (C), P.R. Wallace (C), H.H. Clayton (C), S.A. Kushneriuk (C), M.H.L. Pryce , E.A. Guggenheim and others investigated how the release and distribution of neutrons in a reactor would be affected by the size, shape and relative position of the parts of the reactor. B. Carlson organized a computing section.

The role of the small group of engineers in the Montréal Laboratory was to conceive in outline the design of the reactor, how its uranium fuel would be cooled and how the heat would be dissipated, how it would be shielded to protect the operators from radiation, how the used fuel containing fission products would be removed and stored safely, how the operating power of the reactor would be regulated, what automatic protective devices would help to ensure safe operation, and many other features of design. The group comprised R.E. Newell, D.W. Ginns, J.H.L. Matheson (C), G.S. Anderson, S.G. Bauer and others.

It was also their responsibility to advise those who would design the reactor and chemical plants. Restrictions that were very unusual in engineering experience had to be imposed on the choice of materials and other features of design in order to satisfy somewhat conflicting requirements of working temperatures, rates of heat transfer, permissible leakages of fluids, tendency to capture neutrons, speed of response of instruments and other details.

Since the original purpose of the project was to demonstrate the feasibility of producing plutonium in a heavy water-moderated reactor, the chemists and chemical engineers developed the process by which the plutonium would be separated from the uranium and the radioactive fission products after the fuel was removed from the reactor. They also developed the process by which uranium-233 (another fissionable material) could be separated from thorium and fission products.

The development of these processes was made possible by American contributions of small amounts of plutonium and uranium that had been exposed to neutrons and therefore contained the nuclear fuel and fission products produced in a reactor. A group in Montréal including B.L. Goldschmidt , L.G. Cook (C), T.J. Hardwick and others tested a very large number of solvents to find the most suitable for separating the plutonium and uranium from the very radioactive fission products by preferential dissolving. Another group in the National Research Council laboratories in Ottawa, including S.E. Cambron (C), R.H. Betts (C),

9 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

R. Mungen (C), E.E. Winter (C), Louis Siminovitch (C) , M.B. Wilk (C) and R.P. Cahn (C), further developed the process so that it was possible to obtain plutonium quite free from both uranium and fission products. I.R. Mills (C) and C.H. Simpkinson (C) designed the plant which was later operated in Chalk River, until it was dismantled when it was clear that its plutonium production was not economically profitable at that time.

The comparable process for separating uranium-233 from the thorium, in which it is produced, and the accompanying fission products, was developed by B.L. Goldschmidt , L.G. Cook (C), J. Guéron , F. Morgan, J.W.T. Spinks (C) and L. Yaffe (C) . A processing plant was also built and used later at Chalk River. P.E. Gishler (C), R. Wilkinson, T. Boyer (C), P.J. Sereda (C) and C.R.G. Holmes (C) investigated the corrosion of alloys, that might be used for protective covering of uranium fuel rods in the reactor, by Ottawa River water, and the possible need for water treatment.

The chemical properties of uranium, neptunium and some of the fission products were investigated by J. Guéron , W.E. Grummitt (C), A.C. English, G. Wilkinson, C.E. Mackintosh and others. H.G. Heal experimented with the electrochemistry of uranium compounds and H. Greenwood tested the corrosion resistance of uranium-silicon alloys.

It is impossible in a short history to mention all the important contributions to nuclear science by the Montréal Laboratory. A list of its scientists and engineers, compiled in August 1945, is appended. Many of them are better remembered today for their work in later years at Chalk River, or in universities in the United Kingdom and France after leaving Montréal, particularly those who were in Montréal for only a few months.

A notable achievement in basic science was the delineation of the "4n+1" series of nuclear isotopes - identifying them and observing their properties - by T.E. Cranshaw, P. Demers (C), A.C. English, J.A. Harvey (C), E.P. Hincks (C), J.B. Jelley and A. Nunn May . The "4n+1" series is a group of heavy isotopes, not found in natural minerals, that is produced by successive disintegrations from uranium-233. It is analogous to the three series of naturally radioactive isotopes that have long been known to result from the disintegration of uranium, thorium and actinium. Knowledge of the properties of the "4n+1" substances is important in developing methods of recovering pure uranium-233, which is a most promising nuclear fuel that can be derived from thorium.

Nuclear fission provides two means by which the number of known kinds of atoms has been greatly increased: the production of fission products, and the release of neutrons which by impact on existing kinds of atoms transform them into new kinds. The increase in the variety of atoms available for study permitted research that has greatly extended our knowledge of the structure and internal behaviour of atomic nuclei. Some of the experiments involved bombarding the new substances with neutrons or other atomic particles or radiation, and observing the radiation or particles that emerged to discover what changes were caused in the atoms when they were hit. Some of the research in the Montréal Laboratory was of this kind using small quantities of fission product material received from the United States. It provided preliminary experience that led some of the young scientists involved to very productive careers in nuclear research later at Chalk River and elsewhere, when much greater quantities of these materials could be produced in the nuclear reactors.

A useful experimental technique developed by P. Demers (C) involved the use of very fine-grained photographic emulsion in which the "tracks", roughly one hundredth of a millimetre in length left by individual alpha particles emitted from specks of radioactive matter resting on the emulsion, could be clearly seen with a microscope. The lengths of the "tracks" helped to distinguish the kinds of isotopes from which the alpha rays came. It also revealed the "tracks" produced by the fission fragments of single uranium atoms.

Other experimental techniques were developed and special instruments were devised. They included beta-ray spectrometers for measuring the velocity of electrons, by L.G. Elliot (C) and B. Kinsey; ion chambers of various kinds to detect neutrons and other atomic particles, by H. Carmichael and J.F. Seljes; an

10 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

electronic device called a "kick sorter", by C.H. Westcott and G.C. Hanna; electronic amplifiers for measuring the very small direct currents of electricity through ion chambers, by R.J. Cox and F.J.M. Farley; ion chambers for measuring the energy of motion of atomic particles, by T.E. Cranshaw, J.A. Harvey (C) and others; and an accelerator to produce fast moving heavy hydrogen ions, by A.G. Ward (C) and J. Warren; and other special pieces of equipment that could not be obtained commercially. H.M. Cave (C), E.B. Paul (C), J.B. Jelley and P. Lamb (C) converted a 600,000 volt X-ray generator that I had built a few years before in Ottawa into an accelerator of positively charged atomic particles for nuclear research.

There were miscellaneous research and development activities which, after August 1944, were grouped together in a division called Technical Physics under my direction. They included research on heat transfer from simulated reactor fuel rods to cooling water by S.G. Bauer, J.W. Knowles (C) and E.W. Guptill (C) in Montréal, and by P.E. Gishler (C) in the National Research Council laboratories in Ottawa, in which particular attention was given to the effects of surface boiling, steam film formation and thin water films; the development of reactor control rods and aluminum cladding of uranium fuel rods to prevent corrosion by G.S. Farnham (C) and others in the Department of Mines and Resources; the development of control rod mechanisms and instruments by N.Q. Lawrence; and the construction of Geiger counters and of "boron chambers" for the measurement of neutron flow by N. Veall. H.F. Freundlich, J.W. Ozeroff (C), R. Callow and J. Elsey designed and supervised the construction of much of the electronic research instruments for the laboratory. D.C. Douglas (C) investigated the possible contamination, by radioactive substances, of air and water circulated through a reactor for cooling it. M.W. Lister (C) advised on the possible spread of any radioactive contamination in the atmosphere, and G.W.C. Tait (C) used smoke tests to investigate the movements of air in the river valley where the reactor was to be built. Dr. C.B. Pierce supervised medical tests to reveal harmful effects of radiation on the health of employees.

Scientists in other laboratories contributed to the program. At McMaster University, Prof. H.G. Thode and others used mass spectrometer methods in testing heavy water and in separating isotopes. Help was received from Prof. F.E. Beamish at the University of Toronto with chemical analysis, from Prof. L.M. Pidgeon at the University of Toronto in uranium chemistry, and from G.S. Farnham, R.L. Cunningham and others in the Department of Mines and Resources in metallurgical subjects. McGill University provided library services.

Secrecy about the work of the laboratory was extreme. Younger scientists were forbidden by Halban to discuss their work with others in a different scientific field. When the distinguished Danish physicist, Niels Bohr, visited the laboratory he was addressed as "Mr. Baker". Code words were also used in referring to uranium compounds, heavy water, and other important materials.

Decision at Last

There was little benefit for Canada in continuing the research in the Montréal Laboratory without close cooperation with the British and American scientists, and this view was expressed to the authorities in both countries. The situation was worsened in both by poor communication among the heads of states, officials and scientists. (Aside: In this article, very little has been said about the misunderstandings regarding the supply of uranium and heavy water, and of political differences. A comparison of American, British, French and Canadian accounts of wartime nuclear cooperation reveals them clearly. See the references cited at the end of this paper.)

Better understanding and good will were achieved, eventually, through the efforts of: James Chadwick, the most renowned of living British atomic scientists; General Groves, head of the whole American nuclear effort; and Dr. C.J. Mackenzie. Groves and Chadwick were of very different backgrounds, disciplines and personalities but they trusted and respected each other. Mackenzie understood very well the points of view on both sides. Their good sense prevailed; the Montréal Laboratory was saved.

11 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

The long-awaited decision to proceed with the design and construction of a heavy water moderated nuclear reactor in Canada was made at a meeting of the Combined Policy Committee in Washington on April 13, 1944, which was attended by both the Hon. C.D. Howe and Dr. Mackenzie. General Groves was present and we are told that he assisted greatly in reaching an agreement. It was agreed that there would be a full exchange of information relevant to the design of the reactor and the extraction of the plutonium it produced. Halban and the other scientists who were not British subjects would leave, and the English Physicist, John Cockroft, would be Director of the laboratory.

A group photo taken in 1945 shows a number of those involved in the construction of the Chalk River Nuclear Laboratories.

From left-to right: Major J.H. Brace, president, Fraser Brace Ltd; Arthur N. Budden, Department of Munitions and Supply; unidentified; Dr. C.J. Mackenzie, president, National Research Council; F.J. Palmer, general superintendent, Fraser Brace Ltd; H. Grenville Smith, vice-president, Canadian Industries Ltd. (Defence Industries Ltd., or D.I.L.); D.S. Kirkbride, general superintendent, Services, Petawawa Works, D.I.L.; unidentified; Major A.B. McEwan, manager, Special Projects division, D.I.L.; H.J. Desbarats, manager, Petawawa Works, D.I.L.; Dr. J.D. Cockroft, director of the Canadian Project, 1944-46; C.J. Jackson, chief engineer, Special Projects division, D.I.L.; Gordon R. Stephens, general manager, Fraser Brace Ltd.; H.S. Milne, resident engineer, Petawawa Works, D.I.L.; M. Green, superintendent, Fraser Brace Ltd.

The atmosphere in Montréal changed quickly. The Americans fully supported the project. Scientific information, supplies of materials and help in many ways came from the United States, and scientists from Montréal visited American nuclear research centres where relevant work was being done.

Defence Industries Ltd. was engaged to do the engineering design of the reactor, the laboratories, the services, and the town where the employees would live, and to operate them for the National Research Council of Canada which would direct the research. Fraser Brace Ltd. would do the construction under contract to Defence Industries Ltd. After a score of possible sites was considered, one near Chalk River was chosen and called the "Petawawa Works". The location now known as Deep River was selected for the town. Design began almost at once, and construction proceeded rapidly in spite of difficulties in procuring

12 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

materials and labour. The story of this construction is omitted here because it is more suitably a part of the history of NRX.

Dr. J.D. Cockroft arrived at the CRNL site on April 26, 1944 and the direction of the Laboratory was reorganized. E.W.R. Steacie (C) was its Assistant Director and acted also as Head of the Chemistry division after Paneth left. Halban became Head of the Nuclear Physics division as mentioned earlier. Sargent (C) became Head of the Nuclear Physics division when Halban left at the end of March, 1945, and Volkoff (C) succeeded Placzek a little later as Head of the Theoretical Physics division.

Under Cockroft, the administration of the laboratories became less difficult, morale improved and a sense of clear purpose was restored. While the laboratories were being built at Chalk River, the staff moved there gradually from the Montréal Laboratory which was closed down in July, 1946.

ZEEP

Success in operating the proposed new reactor, which was named NRX (National Research eXperimental), would depend critically on the design of its "lattice". That is, on the size, shape and composition of the uranium fuel rods and their distance apart in the heavy water. The dimensions had been calculated by the mathematical physicists in the Montréal Laboratory from theory. The general validity of the theory had been proved by the success of the first American reactors, including one in Chicago which used heavy water (CP-3, for Chicago Pile 3) and was designed by a group under Canadian-born Walter Zinn. However, before the theory could be used as a final guide for the design of NRX, a confirmation was needed by experiment of the behaviour of neutrons in the kind of reactor lattice that would be used. It was done in Montréal by Sargent (C), H.G. Hereward, A. Munn (C), P. Cavanaugh, D. Booker, M. Burrow, N. Niemi (C) and L. Nirenberg (C) with small quantities of the materials.

This background of theory and experiment made it possible to decide the outline design of the NRX reactor, its shielding, the location of auxiliary equipment and the building that would house them, so that the construction could begin. Before NRX was completed, however, it was desirable to have some experience in operating a comparable reactor that could be easily altered. After two months of study it was decided, on August 24, 1944, to build a very simple reactor that could be completed quickly, and was designed so that its uranium rods and other parts could easily be changed or rearranged.

13 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Dr. Lew Kowarski , who had declined to come to Canada to work under Halban, had joined the staff under Cockroft. He, with New Zealander Charles Watson-Munro and Canadians George Klein and Don Nazzer, were given the task of designing this small reactor and putting it into operation. They were helped by A.H. Allan, F.W. Fenning, G. Fergusson, C.W. Gilbert, and E.P. Hincks (C). Kowarski named it ZEEP, which he said stood for "zero energy experimental pile".

14 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Simplicity and flexibility were achieved in this reactor by omitting any provision for removing heat produced by the nuclear reaction, very heavy shielding to absorb radiation, and the more elaborate control and safety equipment that is needed in high power reactors.

ZEEP's reactor vessel is an aluminum cylinder about two and one half metres high and two metres in diameter, surrounded by blocks of graphite, as shown above. The uranium rods are hung vertically from a frame across the top. The heavy water is stored in a tank beneath, from which it can be pumped into the vessel. Small cylinders of cadmium were arranged so that if the power became dangerously high, they would drop automatically into the reactor vessel where they would stop the reaction by absorbing neutrons.

15 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

Dr. G.C. Laurence, Dr. C.J. Mackenzie , Hon. C.D. Howe and Dr. J.D. Cockroft in August 1945, a few weeks before ZEEP started up.

On September 4, 1945, at Chalk River, the construction of ZEEP was completed. The uranium metal rods, clad with aluminum, were in place; it remained to add the heavy water. The small team worked carefully, controlling the flow of heavy water into the reactor vessel, and watching the instrument that gave the first indication that nuclear fission had occurred in ZEEP.

They changed to less sensitive instruments, designed by H. Carmichael, that would be suitable when operating ZEEP at the intended power. Ion chambers, sensitive to neutrons, were placed close to the reactor vessel. One of them was connected to a galvonometer on the control desk that projected a spot of light on a millimetre scale marked on a strip of glass for the operator to see. The power at which ZEEP operated would be shown by the displacement of the spot of light across the scale.

The next day, September 5, the heavy water was pumped into the reactor vessel again, a little at a time. When the vessel was partly full, the spot of light was seen to move very slightly. After that, each small addition of heavy water moved the spot faster and farther. At length, the spot showed that ZEEP was operating at the nuclear power for which it was designed.

Thus, for the first time a nuclear reactor had been operated outside the United States. They found that the amount of heavy water required was almost exactly what had been predicted from theory by J. Stewart (C) and G.M. Volkoff (C).

In later years, ZEEP has been used for important research on the behaviour of neutrons in reactors and other purposes by B.W. Sargent and others. It was used by Andrew Pressesky (C), David Walker (C), D.W. Hone (C) and others to provide data for the design of other reactors.

In 1945 and 1946, most of the British scientists returned to England and began the research that was to lead the United Kingdom to become the third nuclear military power, and the first to produce electricity in large nuclear power stations economically. Some joined the Canadian staff and remained here.

The first phase of nuclear research and development in Canada, when scientists and engineers from Great Britain, Commonwealth countries, France and other parts of Europe had contributed so much, and when so much help was received from the United States, was coming to an end. The centre of activity in Canada had moved to Chalk River and there was reorganization, new direction and new purpose. From that time, Canadian ideas regarding design and operation of nuclear reactors, nuclear power stations and nuclear safety began to diverge from those of other countries. Gradually the CANDU conception of nuclear power stations was to emerge.

16 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

During a visit to CRNL in 1971, Dr. Lew Kowarski reminisces about ZEEP with Dr. C.H. Millar (C) and Dr. D.H. Walker (C). In August 1944, Dr. Kowarski was placed in charge of the team that was to design ZEEP, which went into service on September 5, 1945, little more than one year later.

Although most of the British scientists returned to England in 1945 and 1946 to continue their research there, close cooperation with United Kingdom researchers continued. From the left: J.L. Gray, third president of AECL, retired December 1974; Sir John Cockroft, who directed the scientific program in Canada and later led the U.K. program; Dr. W.B. Lewis , who succeeded Cockroft and became senior vice-president, science, AECL, a post he held until his retirement; and Sir John Cook, United Kingdom Atomic Energy Authority.

17 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

The Chalk River Nuclear Laboratories is today Canada's largest establishment devoted entirely to research into the peaceful uses of atomic energy.

I am indebted to Dr. Mackenzie and many of my former colleagues, particularly G.A. Bartholomew, W.E. Grummitt, G.C. Hanna, W.J. Knowles, S.A. Kushneriuk, R.R. MacLanders, B.W. Sargent, W.H. Walker and E.E. Winter, for their help in recalling events.

APPENDIX Professional Personnel of the Montréal Laboratory Before August 1945

Canadians W.J. Allan, Jeanne L. Agnew, C.A. Barnes, Mrs. D. Bate, A.H. Booth, T.W. Boyer, G.C. Butler, A. Cambron, H.M. Cave, H.H. Clayton, M. Cohen, L.G. Cook, D.S. Craig, A.J. Cruickshank, P. Demers, D.C. Douglas, D.M. Eisen, L.G. Elliot , S. Epstein, J.M.G. Fell, F.T. Fitch, C.M. Fraser, S.C. Fultz, P.E. Gishler, G.A.R. Graham, L.M. Grassie, W.E. Grummitt, E.W. Guptill, T.J. Hardwick, J.A. Harvey, E.P. Hincks, C.R.G. Holmes, D.G. Hurst , Miss M.E. Kennedy, Miss P. Kerr, D. Kirkwood, W.J. Knowles, S.A. Kushneriuk, Mrs. J. Laird, G. C. Laurence , J. LeCaine, W.R. Legge. M.W. Lister, J.G. Machutchin, S.N. Maldrett, J.C. Mark, J.H.L. Matheson, K.J. McCallum, L.A. McLeod, J.W. McKay, N. Miller, J.R. Mills, W.A. Mohun, N. Morrow, G.B. Moses, A.M. Munn, N.J. Neimi, L. Nirenberg, J.W. Ozeroff, E.B. Paul, W.S. Peterson, E. Prevost, D.S. Russell, B.W. Sargent, P.J. Sereda, C.H. Simpkinson, J.W.T. Spinks (C) , E.W.R. Steacie , J.D. Stewart, B.M. Thall, A.L. Thompson, Miss A. Underhill, D. Van Patter, A. Vroom, G.M. Volkoff , Muriel Wales, P.R. Wallace, W.H. Walker, A.G. Ward, L. Yaffe .

The following Canadian scientists worked elsewhere on atomic energy problems in association with the Montréal Laboratory:

In the Chemistry department at the University of Toronto, under the direction of Professor F.E. Beamish;

18 of 19 8/1/2008 12:09 PM Early Years of Nuclear Energy Research in Canada http://www.cns-snc.ca/history/early_years/earlyyears.html

H.E. Bewick, J.E. Currah, and D.E. Ryan.

In the Chemistry department of McMaster University, under the direction of Professor Harry.G. Thode ; G. Dean, H.E. Duckworth, R.L. Graham, A.L. Harkness, R.C. Hawkings, D.T. Roberts and S.R. Smith.

In the Metallurgy department at the University of Toronto, under the direction of Professor L.M. Pidgeon; W.A. Alexander and A.C. Tropp.

In the Department of Mines and Resources - Fuel and Ore Laboratory, under the direction of Dr. G.S. Farnham; R.L. Cunningham, H.J. Nichols, G. Ensell and Miss A. McDowell.

From Other Countries F.T. Adler, A.H. Allen, C.B. Amphlett, G.S. Anderson, W.J. Arrol, H.S. Arms, P. Auger, A.F. Barr, S.G. Bauer, D.B. Booker, W.E. Burcham, R. Callow, B. Carlson, H. Carmichael, P.E. Cavanagh, K.E. Chackett, J.D. Cockroft , S.G. Cohen, G.B. Cook, E.D. Courant, R.J. Cox, T.E. Cranshaw, B. Davison, J. Diamond, J.V. Dunworth, J. Elsey, A.C. English, F.W. Fenning, G.J. Fergusson, B.H. Flowers, H.F. Freundlich, K.D. George, C.H. Gilbert, A.H.C.P. Gillison, D.W. Ginns, B.L. Goldschmidt , M. Goldstein, H. Greenwood, J.W.G. Gregory, J. Guéron , E.A. Guggenheim, H.H. Halban, G.C. Hanna, B.G. Harvey, H.G. Heal, H.G. Hereward, R.P. Hudson, J.F. Jackson, J.B. Jelley, K.D.B. Johnson, N. Kemmer, B. Kinsey, L. Kowarski , P. Lamb, N.Q. Lawrence, C.E. Mackintosh, A.G. Maddock, R.E. Marshak, G.R. Martin, A. Nunn May , P.M. Milner, J.S. Mitchell, F. Morgan , W.K.R. Musgrave, R.E. Newell, H.R. Paneth, C.O. Peabody, G. Placzek , B. Pontecorvo , H. Preston-Thomas, M.H.L. Pryce , C. Reid, D.T. Roberts, H. Seligman, K. Smith, B.S. Smith, R. Spence, J.F. Steljes, F. Sterry, J. Sutton, J. Thewlis, H. Tongue, N.J. Veall, C.H. Westcott , J.B. Warren, C.N. Watson-Munro, D. West, W.J. Whitehouse, G. Wilkinson, R. Wilkinson, W.W. Young.

REFERENCES

1. Canada's Nuclear Story by W. Eggleston; Clarke, Irwin and Company Limited, 1965. 2. Arms, Men and Governments; The War Policies of Canada 1935-1945 by C.P. Stacey, The Queen's Printer. 3. Britain and Atomic Energy 1939-1945 by Margaret Gowing, The MacMillan Company of Canada Limited, Toronto, 1964. 4. L'Aventure Atomique by B. Goldschmidt . Libraire Arthème Fayard, 1962. 5. The New World 1939-1946 by R.G. Hewlett and Oscar E. Anderson Jr., The Pennsylvania University Press, 1962. 6. A General Account of the Development of Methods of Using Atomic Energy for Military Purposes: The Official Report on the Development of the Atomic Bomb Under the Auspices of the United States Government 1940-1945 by Henry D. Smyth, Washington, 1945.

[Canadian Nuclear Society home page]

19 of 19 8/1/2008 12:09 PM SESSION 6. DISCUSSION ON "ANCILLARY PROBLEMS OF REACTORS" Chaim": H. S. ARMS,D.Phil., F.1nst.P.

Shielding for reactors By G. C. LACRESCE,R.I.B.E., Ph.D., F.R.S.C., Atomic Energy of Canada Lid., Chalk River, Ontario, Canada

A brief introductory discussion is given of the design of shielding for nuclear reactors. li includes consideration of design requirements, choice of materials and dimensions, nuclear reactions that affect the penetration, heating of the shielding and production of radioactivity in the coolant. Because the basic data available are insufficient for high accuracy in design calculation, the use of very complex theory and very laborious calculation has little advantage. Very approximate methods are used, and allowance made for possible error in calculating the thickness of shielding required. The problem is simplified by neglecting radiation processes that have relatively little effect on the overall penetration, but the importance of these effects depends on the design of the shielding.

The most widely recognized authority on radiation hazards laborious. The use of accurate detailed theory is at present is the International Commission on Radiological Protection. scarcely warranted because it does not yield the accuracy Its recommendations. as revised December 1, 1954, were desired owing to the inaccuracy of some of the data that is published by the British Institute of Radiology,(') and in required. In practice much simpler and more approximate other radiological journals. The Commission recommends methods are commonly used, and the shielding thicknesses for those who are exposed to penetrating radiation in their are increased to allow for possible error. normal employment that the dose of gamma radiation In this brief introduction to reactor shielding it is not absorbed in any part of the body should not exceed 30 ergs/cm3 possible to discuss the theory of penetration of radiation in any week. They recommend that the energy absorbed except to describe in simple non-mathenatic terms the more from fast protons produced by collisions of fast neutrom important effects. Comprehensive theory is given in a number should not exceed 3 ergs/m3per week. of well known texts, and its application to shielding design is In the design of reactor shielding to conform to these contained in the Reactor Shielding Manual.(') A helpful recommendations, the intensities in the shielding near its source of information for those having access to classified Outer bouzdary next io areas occupied by personnel during documents is a repor: by Henderson and Whittier.c3) normal working hours should not exceed the following limits : The primary source of radiation in the fuel may be described approximately as follows: The release of 1 W of power by 7.5 mrjh of gamma rays nuclear fission is accompanied by 2000 thermal neutrons cmP2 s-' flux 30 fast neutrons cm-* s-l flux 7.8 x 1OIo n/s 5.2 x 1O'O penetrating phoronsk The dosage rate produced by one photon cm-* s-l is given in Fig. I as a function of its quantum energy. The number N(E) of neutrons emitted per watt with energy The passage of gamma rays and neutrons through the between E and E - dE is given approximately by shielding depends in a complex manner on the energies of N(E) = 7.8 x 1O'O x dzsinh\/(E/O.5 MeV)e-E/l"eVdE re (1) I os for energies up to 17 MeV. The gamma rays have widely .-i: v distributed energies but may be grouped to simplify cal-

yl culations as ldb 2 5 x 1O'O photonsis of average energy 1 '35 MeV

wIo 2 x lo9 photonsis of average energy 5 MeV 0 0 To this should be added the gamma rays produced by 16' radio-active neutron capture in the fuel. The important IO-2 I 6' 10 O 10; penetrating components emitted with 1 W of fission power ENERGY (MeV) may be grouped for natura! uranium as follows Fig. 1. Gamma ray dosage rate produced by 1 photon 2.5 x 1O"J photons/s of average energy 0.8 MeV from fission cm-2 s-l products lo7 photons/s of average energy 2.8 MeV from fission the rays and particles, the kinds of the nuclides they encounter products and the shape and arrangement of the shield. Precise cal- 3 x 1Olo photons/s of average energy 5 MeV from fissile dation of the penetration is mathematically difficult and and fertile nuclides. s 54 BRITISH JOURNAL OF APPLIED PHYSICS Shielding for reactors Session 6 The important penetrating gamma rays produced by a distance r in a single free path is e-rij.. Taking the inverse neutron capture in other materials commonly used in the square law into account the number of gamma-rays reaching reactor core are given in Table 1. a unit area at a distance r from a point source q without scattering is Table 1. Penetrating gamma rays from neutron capture in reactor materials

Number of photons Approxinure Capiure per neutron evergy Capturing element cross-secrion captured of photon (MeV) Hydrogen 0.336 i 2.2 7.7 Aluminium 0.21 4 1.8 f 0.5 8 Iron 2.43 i 0.2 6 i0.2 4 (0.93 7.4 Lead 0.17 i 0.07 6.7 r 1.1 4 Zirconium 0.18 6 7.5 Beryllium 9.0 10.75 6.8 ENERGY (MeV) (0.5 4 2.8 Fig. 3. Variation of mean free path of gamma rays with Sodium 0.47 1.4 energy 5 Nickel 4.8 0.7 8.9 Including the contribution to the penetration by the 0.25 1.8 photons that are scattered once, twice and more often Manganese 12.6 2.1 increases the penetration by a factor B which is called the 9.2 build-up factor. Hence the total number of gamma-rays i0.2 4.0 reaching a distance r from the source is Chromium 2.9 9.7 5.0 Carbon 0,0045 I 4.95 (3)

The gamma rays are scattered frequently in passing through The build-up factor B depends on the thickness and kind a large bulk of material. However, most of the effect after of material and the initial energy of the gamma-ray quanta. penetrating a very thick layer is due to photons that have Fairly comprehensive data on build-up factors may be found traversed most of the distance from the source in a single in papers on radiation shielding.(4) long free path, as illustrated in Fig. 2, or with comparatively The build-up factor B usually pertains to the dosage rate (rih). Hence expression (3) does not represent the actual flux. It gives that flux of photons of the original energy before sczttering which would produce the same dosage rate as actually occurs. It is therefore what is requiied for the simple treatment given here in which the gamma-ray effects are derived from the penetration of the non-scattering primary rays. In general the source is extended, and it is usual in approximate calculations to treat it as an infinite plane and estimate the intensity at a distance x measured perpendicularly from that plane. The intensity at a distance x from a plane source Fig. 2. Migration of gamma rays through a very thick shield of strength 4 per unit area is few small ang!e deflexions. In spite of the fact that a very long free path is relatively improbable, the photons that have only short free paths between scattering collisions, as illustrated by the dotted line, contribute very little energy to the In practical cases xih is usually large and we can make the total penetration, owing to the greater length of their tortuous approximation paths, and the loss of energy in the many collisions. Mean free path between collisions depends on the energy of the gamma-rays as illustrated in Fig. 3. A single long free path traversing most of the distance from source to detector, as in Fig. 1, most probably occurs at that energy for which the mean free path X is greatest. The probability of traversing Where the radiation penetrates a number of layers of SUPPLEMENTKO. 5 s 55 Session 6 G. C. Laurence different shielding materials of thickness fn and mean free Most of the thermal neutrons in the outer shielding reached path An the term x/h should be replaced by the outer layers while they were fast neutrons. When t is greater than h, most of the thermal neutrons penetrating the layer of material are ones that entered it as where the mean free paths An are those characteristic of the thermal neutrons. This is true in the reflector that surrounds gamma-ray energy for which Znfn/h,,has the smallest value, the core of some reactors. In such a case the approximation i.e. the energy which gives the greatest penetration. (8) is not applicable. Diffusion theory should be used. Fast neutrons are scattered many times before reaching Expression (8) for the thermal neutron flux is not valid thermal energies and the average angle of scattering is large. near the boundary between two different materials. Three Their migration, therefore, can be regarded as a diffusion group diffusion theory or more elaborate methods are process in regions close to the source. At greater distances required for accurate description of these boundary effects. from the source simple diffusion theory does not give a Some of the effects that occur near the boundary between satisfactory approximation in representing their motion. iron and concrete shielding are illustrated by some measure- Better approximations are obtained by treating them in the ments of Bell, Miller and Rob~on(~)which are plotted in same way as the gamma-rays. Thus the intensity of the Fig. 5. The thermal neutron flux in the iron is approximately neutrons of most penetrating energy at a great distance x proportional to the fast neutron flux between A and B as from the infinite source is approximately predicted by the formula above. The thermal flux in the concrete is also approximately proportional to the fast neutron flux beyond the point D at some distance from the boundary. Transition effects are evident between B and D. The thermal neutron flux rises between B and C because The source strength q should, of course, include only those the neutrons are much more strongly captured in the iron neutrons whose energies are in the band of high transmission than in the concrete. Thermal neutrons diffuse back from and above it. The dependence on neutron energy of the the concrete into the iron contributing to the neutron flux mean free path h in water and in iron is shown in Fig. 4. in that material. The thermal flux between C and D is relatively high because iron is a much poorer moderating material than concrete. Neutrons of intermediate energies migrate from the iron to I I ! I I IO I ! I I 1 In, the concrete before they lose much of their kinetic energy. They are rapidly decelerated in the concrete providing an additional source of thermal neutrons that is important between the boundary and the point D. It may be seen from a comparison of the curves for gamma- ray intensities and thermal neutron flux that most of the gamma-radiation in the outer parts of the shielding is produced by neutron capture in the region B and C. The measurements which are plotted in Fig. 5 were made with blocks of iron and concrete plac2d in one of the experimental holes of the NRX reactor shielding. They were

lOeV IOOheV lkeV IO& IOOeV lMeV IOMeV ENERGY Fig. 4. Variation of mean free path of neutrons with energy

Lower energy neutrons would be rapidly attenuated and would give relatively little contribution to the neutron intensities in the outer part of the shielding. In the case of iron, very fast neutrons have shorter mean free paths than those of energies of a few mega-electron volts. However, these are also available for penetration of the iron at the energies of maximum mean free path when their energies have been reduced by scattering nearer the source. The fast neutrons, as they lose kinetic energy in collision, produce a thermal neutron flux which is given approximatelyby

I I I lfFOOTL 'I where is the mean free path of the most penetrating fast I neutrons, & is the capture mean free path of the thermal neutrons, L2 is the thermal diffusion area and Lz is the Fig. 5.. Neutron flux near boundary between iron and migration area for slowing to thermal velocity. concrete as measured in the NRX reactor shielding Expression (8) for the thermal flux applies only when the thermal diffusion length L is less than the mean free path h greatly affected by leakage of neutrons and gamma-rays of the fast neutrons, which is true of most shielding materials. through the cracks around the blocks. The curves of Fig. 5 S 56 BRITISH JOURNAL OF APPLIED PHYSICS Shielding for reactors Session - 6 should not be used therefore for accurate prediction of of very high energy, for example energies greater than 5 MV, effects 111 other shielding. is longer than in a number of other common materials, A convenient approximation in shielding calculations is to including most common metals. For example, see Fig. 4. regard the reactor as an infinite plane source located at the A combination of water and steel or lead or some other metal Outer boundary of the core. Thus if s(t) is the source strength provides more rapid attenuation for the whole spectrum of (number of photons or neutrons emitted per cm3 per second) fission neutrons. The most suitable thicknesses for these at a depth t below the boundary and k is the absorption layers may depend on a number of factors. It may be desir- coefficient in the core materials, the approximately equivalent able to provide more metal on the outer regions of the thermal Source strength at the outer boundary is shield than in the inner, because the metal is a much more effective absorber for the gamma-radiation produced by neutron capture in the shielding itself. 4 = a'o,.-kt. (9) Compounds of lithium or boron may be added to the water in the thermal shield to absorb thermal neutrons The contribution to q by radiation originating at depth t without producing high energy gamma-radiation. Practical decreases rapidly as f increases, and therefore simplifying considerations such as increased corrosion, however, may approximations are easily made in evaluating the integral. outweigh the advantage. The first step in designing the shielding is to estimate the The part of the shield outside the thermal shield is called heating effects produced by the absorption of radiation in the biological shield, because it reduces the intensity of the structural parts including the shell of the reactor vessel. In radiation to a safe level for personnel working near the water-cooled power reactors, the vessel shell may be of steel reactor. It is usually much thicker than the thermal shield. several inches in thickness. Absorption of the radiant energy To avoid excessive heating or radiation damage, the rate of in the steel may result in a much higher temperature in the energy absorption should not exceed interior of the wall than at its outer surfaces, causing excessive 104 ergs cm-3 s-l in concrete thermal stresses unless it is prevented by radiation shielding lo3ergs cmw3s-! in wood, masonite and phenolic pro- between the core and the shell. A convenient form of shield- ducts ing is plates of steel immersed in water to cool them. Heavy 102 ergs cm-3 s-l in most plastics. water may provide the cooling in a heavy-water moderated reactor. Most of the gamma-radiation entering the biological shield Outside the shell the radiation is still intense enough to results from neutron capture in the thermal shield and the produce excessive heating or radiation damage in such reactor vessel wall. Usually gamma-radiation produced by common shielding materials as concrete or masonite. The fission in the core is relatively unimportant in the biological inner layers of the shielding, therefore, are constructed of shielding. In calculating the intensities of these gamma-rays, materials that are resistant to radiation damage and are it may be assumed for simplicity that the capture of a thermal easily cooled. This part of the shield, which is called the neutron in water produces a gamma-ray quantum of 2.3 MeV, thermal shield, may consist of water, or of steel cooled with in graphite of 4.5 MeV, and in iron and most other materials water passing through suitable channels, or a combination of 7.5 MeV. of both. In the biological shield, since heating effects and neutron In calculating the heating effects it may be assumed for damage are unimportant, a wider choice of materials can be simplicity that each primary gamma-ray quantum and each considered. The simplest and, in most cases, the lightest thermal neutron absorbed contributes an energy of about biological shield is a large volume of water in which the 8 MeV, or 1.4 x lo-!' ergs to the shield material. This reactor and thermal shield are immersed. The neutrons are assumption errs on the side of caution. The effects of the attenuated more rapidly than the gamma-rays in the water kinetic energy of the fast neutrons are comparatively negligible and the required thickness is determined by the penetration in the shield because most of their kinetic energy is already of gamma-rays from the thermal shield. In contrast, the lost, and only the energy released by their capture as thermal required thickness of a shield that contains no hydrogen, neutrons need be considered. The energy absorbed per unit such as one composed entirely of metal, is determined by the volume is therefore penetration of the neutrons. Saving in weight and thickness is obtained by using a materia1 containing both hydrogen and &Nu (1.4 x lo-" ergs) for thermal neutrons heavier elements in suitable proportions. Ordinary concrete and I/h (1.4 x ergs) for hard gamma-rays. is very suitable for this purpose and is convenient and inexpensive. Using c.g.s. units throughout, these expressions give the heat The hydrogen content of most kinds of Portland Cement production in ergs cm-3 s-'. One erg ~m-~s-l = 0.009 65 concrete is greater than 5 % of the number of atoms, and this B.t.u. h-l ft-3. is sufficient to ensure that the biological effectiveness of the Since the absorption of neutrons results in the production neutrons emerging from the outside of the shield is unim- of very penetrating gamma-rays it is desirable to reduce the portant in comparison with that of the gamma-radiation. neutron flux as rapidly as possible in the inner layers of the Special kinds of aggregate material may be used in the thermal shield. The choice of a material for this purpose concrete to increase its effectiveness in absorbing gamma-rays. depends on cost, mechanical simplicity, and whether it is The rate of attenuation of the gamma-radiation in the more important to save weight or thickness of shielding concrete is approximately proportional to its specific gravity. material. Water (on which useful data has been published('? Crushed rock of high density or steel punchings or other high iS particularly satisfactory in saving weight owing to the very density material may be used to increase the density of the high scattering cross-section of hydrogen for neutrons of . concrete, but since the quantities required are large the cost energy below 1 MeV. Where small thickness is desired, and prbxhity to the source will be major considerations in water alone does not provide the most effective shield for deciding to use a special aggregate. neutrons because the mean free path in water for neutrons For special purposes, such as mobile reactors, where small SUPPLEMENTNo. 5 s 57 Session 6 G. C. Laurence bulk and weight are reqilired, the whole shield may be made T, is duration of exposure of layers of metal and hydrogenous materials. For example, Td is time to flow from reactor’to equipment layers of lead or iror, might be used with layers of water, T, is time to circulate once. masonite or other organic materials. The best relative proportions would have to be determined in each case. For a single pass cooling system the concentration of radio. Great care is necessary in the design of reactor shielding active nuclides is given by to avoid the escape of radiation through cracks, pipe chases n - N+(1 - e-Te/-)e-Ta/: and other openings and through large blocks of material (11) embedded in the shielding that are poor shielding material. The number R of radioactive atoms discharged per Second For example, a one-inch steel bar embedded in the concrete in the effluent of a single pass cooling system is biological shield with its axis in the direction of propagation of the radiation is a channel through which fast neutrons can penetrate much more easily than through the surrounding where V is the volume exposed to the neutron flux. concrete. If the bar is long the radiation intensity at its The presence of impurities can contribute greatly to the outer end may be excessive, Methods for estimating the radioactivity of cooling liquids. In water or heavy-water leakage of radiation through narrow passages are given in a closed-circuit cooling systems, the concentration of these number of published reports, for example reference (2). impurities can be kept to a very low level by the use of ion The liquid or gas that is used for cooling the reactor exchange resins and filters. The most important radioactive becomes radioactive by the capture of neutrons producing impurities that may occur in a closed water-cooling system radioactive nuclides. The cooling fluid passes through are 24Na from aluminium and 59Fe and 55Mn from stainless piping, heat exchangers, pumps and other equipment outside and other kinds of steel. Where water is used from lakes the reactor. These need shielding also. and rivers for single pass cooling the dissolved minerals in Pumps and other equipment which require maintenance the water can produce a very high radioactivity in the efFuent. may be placed in rooms with walls designed to give the In some reactors there are spaces in regions of high neutron necessary shielding, if these rooms are kept locked while the flux that contain air. Argon-41 is produced and may be an plant is operating. However, the radiation may continue to important health hazard if it is allowed to leak into the be hazardous during periods of shutdown owing to residual operating room. This can be prevented by drawing the air activity in the coolant or the deposition of radioactive through suitable shielded ducts and discharging it from a high impurities on inner surfaces of the equipment. It may be stack. Under some meteorological conditions, such as the desirable, therefore, to provide local shielding directly on the existence of temperature inversions, the radioactive air dis-

Table 2. Radioactive isotopes produced by coolants Quanta per Radiation Coolant Nuclear reac:ion Cioss-section HaiT-life Threshold capture emirred 1.25 MeV P Air 40A(ny)41A 0.626 1.8 h P 11 1.3 MeV y 0.027 7.5 s 11 MeV* 1 7.1 MeV; 110 MeV Water including D,O 0.020 4.14 s 11.2 MeV* 1 MeV n 0.00022 29.4 s 0’7 4.5 MeV ,6 1.6MeVv ( 2H(ny)3H 0.0003 12.5 yr Sodium 23Na(ny)24Na 14.8 h P Water 27Al(na)24Na 14.8 h no y-rays Water 27Al(ny)28A1 0.21 2.4 min Water s8Fe(ny)59Fe 0.36 46 days (0.5 1.1 MeVy 10.5 1.3MeVy Water 54Mn(ny)5jMn 13 * The fraction of the fission neutron spectrum above these thresholds is 2.2 x 10-4. equipment also. An altemative measure is to limit the time charged from the stack may remain sragnant over the plant in which personnel are permitted to remain in the room. site. The time of exposure T, is usually short compared with Table 2 lists the important radioactive isotopes that are the decay period (1.8 h/0.69 = 2.6 h) of 41A and therefore produced in some of the common coolants. In a closed circuit cooling system the concentration of RNVN$u (13) radioactive nuclides is given approximately by the following Hence the volume V of the air spaces in the reactor should be expression kept small. R is almost independent of the rate of flow Y/T, of air through the spaces. n Ng+-(1 - e-Tek) e-Td/-/(l - e-Tc/.) - (lo) The design of the shielding for nuclear reactors is a complex, where N is concentration of parent atoms laborious and detailed undertaking. There are manv un- n is concenrration of product atoms certainties in the data. Approximation must be made In the $I is average neutron flux calculations if they are to be completed in reasonable time. U is activation cross-section It is easy to overlook important effects. To allow for error, T is decay period of product atoms it is prudent when the design is completed to increase the s 58 BRITISH JOURNAL OF APPLIED PHYSICS Shielding for reactors Session 6 thicknesses of reactor materials. The surest and most satis- have been of great assistance to the author in preparmg factory approach is to base the design as far as possible on this paper. experience of the shielding of other reactors already in REFERENCES operationor On made with a mock-u? Of the proposed (1) Supplement No. 6 to the British Journal of Radiology reactor shield. Even if this is done the designer is faced with (1 955). many problems in estimating the possible leakage of radiation (2) Report ~1D-7004(Washington: Ato;nic through pipe-chases, apertures around embedded equipment Commission, 1956). and experimental facilities, control devices and the channels (3) W. J., and wwrmR, A. c. Report 578 though which the reactor fuel is replaced. (Atomic Energy of Canada Ltd., Chalk River, Ontario). (4) GOLDSTEIN,H., and WILKINS,JR., J. W. Report NYO-3075 (Washington: National Bureau of Standards). ACKNOWLEDGEMENT (5) BELL, MLLER and ROBSON.Report on the shielding of Reports of shielding calculations by W. J. Henderson, the NRX reactor. .4. C. Whittier and I. L. Wilson, which have not yet been (6) Report T1D-5275, Chapter LI (U.S. Atomic Energy declassified because they also contain secret information, Commission, 1955).

The control of reactors By A. J. SALMON,BSc., Ph.D., A.M.I.E.E., A.Inst.P., Associated Electrical Industries Ltd., Aldermaston, Berkshire The concepts of reactivity, neutron lifetime and delayed neutrons are described and their relationships with the rate of change of neutron density and reactor power are discussed. The amounts of reactivity to be controlled in a reactor may be large and varying with time. The various components of this reactivity are noted and discussed. The physical methods of control are surveyed and the safety of reactors is briefly considered.

IKTRODUCTION where p is the prompt power equivalent to one fission per second. It is not possible to give simply an exact expression A nuclear reactor may be part of a power system, an experi- relating this thermal power to the neutron density. This is menta! prototype of a power reactor or a research reactor, because the macroscopic fission cross-section is dependent on perhaps of very low power output, acting simply as a radia- the reactor design: for example, it will depend 02 the enrich- tion souice. each case the reactor hsls to be controlled In ment of the fuel and the ratio of fuel to moderator. Further, so that, at least, it can be set at some required power level, the neutron density will have a continuous velocity spectrum maintained at that power and then shut down. Also it is which again depends on the reactor design, the mean neutron usually convenient to be able to bring the reactor up to power velocity increasing in magnitude as we consider the range of any time after it has been shut down. The possibility of the reactors from the thermal type through the intermediate to reactor getting out of control must be minimized so far as is the fast reactor. Even within the thermal group of reactors reasonable and, further, the possibility of damage to the the mean of neutron velocity will depend on the operating reactor must be reduced to a minimum if it does get out of temperature and it will in fact vary within one reactor as it control. This paper briefly discusses the physics of reactors starts operation from the cold state and rises to its normal SO far as it is pertinent to this general problem of the control designed temperature. of reactors. Let us assume that the neutron density maintains its space Throughout the paper reference is made to reactors of distribution constant as a function of time in the reactor so several types which are denoted by generally accepted that it can be specified by a Fean neutron density and a abbreviations. References are normally given for data conform factor; also that the reactor is operating under such cerning these reactors. However, in the case of the gas- conditions that the neutron density, n(c) can be characterized cooled graphite-modulated reactors which will be the initial by a velocity I;. Then we have: type used for power production in the United Kingdom, and denoted in rhe paper by the name PIPPA, any data Pp = nw .p. (w) jva(V)dV (2) quoted is only an estimate based on information generally f available. This data will, nevertheless, give an indication of the behaviour of this reactor type. nw, is, of course, a representative neutron flux which can perhaps best be regarded as the total neutron track length per cubic centimetre per second. (I;) is the probability FUSDAMENTAL CONCEPTS f per centimetre that a neutron of velocity will produce a A list symbols will be,found the appendix w of in fission. Thus from expression (2) we see that the reactor The thermal power produced in a reactor as a prompt power immediately due to the fission process is directly result of the fission process is related to the neutron density proportional to the neutron density. by the expression: The energy released by one fission is distributed(') as shown in Table 1. The neutrinos do not react with matter to any P, = jn(?:,V) 7 (c)pdcdV (1) appreciable extent and so essentially escape from the reactor 1, without releasing within it their energy. Thus, to the II- v SUPPLEMENTNo. 5 s 59