An International Journal of the History of Chemistry Substantia an International Journal of the History of Chemistry

Home , Georges Urbain

2532-3997 S FIRENZE UNIVERSITY PRES An International of the Journal History of Chemistry Substantia Vol. 2 - n. 2 September 2018 September

September 2018 Vol. 2 – n. 2 Substantia An International Journal of the History of Chemistry Substantia An International Journal of the History of Chemistry

Vol. 2, n. 2 - September 2018

Firenze University Press Substantia. An International Journal of the History of Chemistry

Published by Firenze University Press – University of Florence, Italy Via Cittadella, 7 - 50144 Florence - Italy http://www.fupress.com/substantia

Direttore Responsabile: Romeo Perrotta, University of Florence, Italy

Cover image: Polarized light photomicrograph of dendritic crystals of sucrose formed afer drying some drops of the Italian drink Aperol. Field of view: c. 2 mm. Copyright Bernardo Cesare (www.microckscopica.org).

Open Access. Tis issue is distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY-4.0) which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Te Creative Commons Public Domain Dedication (CC0 1.0) waiver applies to the data made available in this issue, unless otherwise stated.

Substantia is honoured to declare the patronage of:

Museo Galileo - Institute and Museum of the History of Science Piazza dei Giudici, 1 - 50122 Florence, Italy http://www.museogalileo.it

Fondazione Prof. Enzo Ferroni - Onlus Via della Lastruccia, 3 50019 Sesto Fiorentino, Italy http://www.fondazioneferroni.it

With the fnancial support of:

Fondazione Cassa di Risparmio di Firenze, Via M. Bufalini 6 - 50121 Firenze, Italy No walls. Just bridges

Substantia is a peer-reviewed, academic international journal dedicated to traditional perspectives as well as innovative and synergistic implications of history and philosophy of Chemistry. It is meant to be a crucible for discussions on science, on making science and its outcomes. Substantia hosts discussions on the connections between chemistry and other horizons of human activities, and on the historical aspects of chemistry. Substantia is published open access twice a year and ofers top quality original full papers, essays, experimental works, reviews, biographies and dissemination manuscripts. All contributions are in English. We proudly welcome our new Associate Editors: Carin Berkowitz, Neil R. Cameron, Stephen Hyde and Ernst Kenndler.

EDITOR-IN-CHIEF

Pierandrea Lo Nostro Department of Chemistry “Ugo Schif” University of Florence, Italy phone: (+39) 055 457-3010 email: [email protected] - [email protected]

ASSOCIATE EDITORS

Virginia Mazzini Neil R. Cameron Australian National University, Australia Monash University, Australia University of Warwick, UK

Stephen Hyde Ernst Kenndler Australian National University, Australia University of Vienna, Austria SCIENTIFIC BOARD as of 18 November 2016

Ferdinando Abbri Luigi Dei Pierluigi Minari University of Siena, Italy University of Florence, Italy University of Florence, Italy Tito Fortunato Arecchi Sarah Everts Juzo Nakayama University of Florence, Italy C&ENews, Berlin, Germany Saitama University, Japan Marco Beretta Juan Manuel García-Ruiz Barry W. Ninham University of Bologna, Italy University of Granada, Spain Australian National University, Au- stralia Paolo Blasi Andrea Goti University of Florence, Italy University of Florence, Italy Mary Virginia Orna ChemSource. Inc, USA Elena Bougleux Antonio Guarna University of Bergamo, Italy University of Florence, Italy Adrian V. Parsegian Univ. of Massachuset Amherst, USA Salvatore Califano Marc Henry University of Florence, Italy University of Strasbourg, Seth C. Rasmussen France North Dakota State University, USA Luigi Campanella University of Rome La Sapienza, Italy Roald Hoffmann Cornell Adrian Rennie University, USA University of Uppsala, Sweden Andrea Cantini University of Florence, Italy Ernst Homburg Piero Sarti Fantoni, University of Maastricht, The University of Florence, Italy Louis Caruana Netherlands Gregorian University of Rome, Italy Vincenzo Schettino Stephen Hyde University of Florence, Italy Elena Castellani Australian National University, University of Florence, Italy Silvia Selleri Australia University of Florence, Italy Luigi Cerruti Juergen Heinrich Maar University of Turin, Italy Brigitte van Tiggelen Univ. Federal de Santa Science History Institute, USA Martin Chaplin Catarina, Brasil London South Bank University, UK Barbara Valtancoli Roberto Livi University of Florence, Italy Marco Ciardi University of Florence, Italy Richard Weiss University of Bologna, Italy Stjepan Marcelja Georgetown University, USA Australian National University, Australia Françoise Winnik University of Helsinki, Finland Sir John Meurig Thomas University of Cambridge, UK

EDITORIAL BOARD Moira Ambrosi, University of Florence, Italy Antonella Capperucci, University of Florence, Italy Laura Colli, University of Florence, Italy Annalisa Guerri, University of Florence, Italy

ASSISTANT EDITOR Duccio Tatini, University of Florence, Italy

MANAGING EDITOR Alessandro Pierno, Firenze University Press, Italy Firenze University Press www.fupress.com/substantia

Editorial Goodnight and Goodluck1: Te End of a Building at the Australian National University 2

May 31 marked the end of the old Applied Maths cian, John Jaeger. Later it became the Research School Building at ANU. of Earth Sciences, one of the leading Schools in its dis- We moved into it in 1971, 47 years ago. For about ciplines in the world. For nearly twenty years they did twenty years before that it had already been home to their work here. the famous Department of Geophysics and Geochemis- Ted Ringwood’s high pressure laboratory was the try. Before the war, when both Lake Burley Grifn and Tower block in the center. ANU were dreams, there were nurses quarters close by From that work Ted invented Synroc, a new way of and the old hospital. Tis building has been a center of disposing of nuclear waste. research for 70 years: there from the very beginning of Ten there was Bill Compston who developed mass ANU. With a magnifcent view over the Lake and the spectrometry that began in Western Australia and meas- Brindabella mountains, the building occupied a prime ured the age of the Earth. Ten Merv Patterson was site on campus, where the Freeway crosses the creek. there. And Ross Taylor, known for his analysis of rocks Before that the Ngunawal peoples who owned this coun- that came back from the moon, and for work on the ori- try forever caught fsh where the Molonglo river turned gin of planets. Paleomagnetism research was so impor- and passed by Black Mountain. Tere was no asbestos. tant for Carey’s theory of continental drif. It was still Te building worked very efectively for collaboration heretical even in the early 1960s. between scientists. It had been declared a major heritage Ted’s idea was to put the nuclear waste back into site twice. No building will replace it for the foreseeable synthetic rocks which resembled those that the urani- future. Academics will just have to double and triple up. um originally came from. Te competitive technology Its demolition is a signifcant, symbolic act of barba- was to put it inside solid glass and bury it in the ocean. rism. Ted missed out because the competition was owned by As Henry Lawson suggested: “Something ought to Rockwell International which is an organization that be said. We should have a party or something”. makes hydrogen bombs and whose CEO was a friend of Tere are some lines from Hamlet that are apt. President Reagan. Reagan and the Head of Rockwell had “My words fy up, my thoughts remain below; Words adjoining ranches in California. Rockwell pushed glass without thoughts never to heaven go”. disposal. Glass won but to no good purpose because the Toughts, not words. Tat’s the thing . sea water cracked the glass. If Ted had won, the world Apart from lamenting the good old days, as one would have been a better place. does, why on earth could it matter - for the future of After Earth Sciences moved, our Department of universities? Applied Mathematics came in. It made a lot of remark- First of all remember that in those early times the able contributions in the Natural Sciences over the years. ANU was an Institute of Advanced Studies, a place of And a lot of applied experimental and development work cutting edge research, a place in which undergraduate was done over and above only mathematics. We were teaching was a diversion from the main game, and not scientists and engineers who worked in the enabling allowed. disciplines, physical, colloid and surface chemistry that And research was done in this building of a quality underlie all of biology and chemical engineering. no one imagined. For example the Department’s laboratories did the It was frst the ANU Dept. of Geophysics and Geo- first measurements on molecular forces, something chemistry founded by the eminent applied mathemati- Isaac Newton tried to do in the 1600s and failed. It did

Substantia. An International Journal of the History of Chemistry 2(2): 5-6, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-55 6 Barry W. Ninham outstanding pioneering work in fbre optics, on porous Te life of the Building spanned a gentler time for media which got the equivalent of the Nobel prize in scholarship and learning in Universities. chemical engineering. Tese things were commercialised With thanks to Jan Morris of Farewell the Trum- too. It brought strange new non Euclidean geometries pets, a poem of E.W. Horning afer the Great War catch- into science. Tey turn out to be common geometries es a whif of it, and the ghosts of those who were here. of nature; from inorganic chemistry to biology. It made major contributions to membrane biology and changed ‘Who are the ones that we cannot see, Tough we feel the face of physical chemistry. Mark Oliphant, one of them as near as near? the fve eminent founding fathers of ANU, was our frst In Chapel one felt them bend the knee, At the match research visitor for two years afer being moved on from one felt them cheer. his position for being too old, at age 65! In the deep still shade of the Colonnade, In the ring- Te Applied Maths Building was optimal for per- ing quad’s full light, sonal engagement and collaboration. Everybody includ- Tey are laughing here, they are chafng there, Yet ing myriad overseas visitors, loved it. It had all the right never in sound or sight’ stuf, call it the wisteria, open corridors or a psycho- ceramic environment, as well as the old Staf Centre. Te lights are going out in Universities across Aus- Tis was a pub without peer, famous world wide, great tralia, with the triumph of a grim political correctness for collaboration. and the death of history. Te old Australian larrikin Whatever, it worked. It attracted visitors from many dipped his “lid” to no man. He is gone. Te Enlighten- countries around the world. It produced over 100 full ment has gone and with it Science itself. professors so far in all kinds of felds, in this country So remember Ozymandias, and what we used to and overseas. Tere were many PhDs. Te Department think Universities stood for, as John Molony has so gained all kinds of awards and distinctions. Four of its elequently expressed further in a following piece. members or colleagues mentored were Chairs of the Nobel prize Committee in chemistry. It is all forgotten now, but what we all did will stand. Ozymandias And it will be remembered elsewhere. By Percy Bysshe Shelley At the end of every speech that the elder Cato gave to the Roman Senate, he always fnished with: Cartha- I met a traveller from an antique land, go delenda est. CARTHAGE MUST BE UTTERLY Who said – ‘Two vast and trunkless legs of stone DESTROYED. Stand in the desert. . . Near them, on the sand, Half sunk Something like that is happening here. We are a shattered visage lies, whose frown, And wrinkled lip, bemused. Why erase the past? and sneer of cold command, Tell that its sculptor well Why bulldose a perfect working building when those passions read funds are being cut? Which yet survive, stamped on these lifeless things, Tere are plans for a new building in three or fve Te hand that mocked them, and the heart that fed; And years time. But if you believe that you believe in fairies. on the pedestal, these words appear: Meantime academics will have to double and triple up in My name is Ozymandias, King of Kings; Look on my cramped quarters. Not so bad really? Works, ye Mighty, and despair! Nothing beside remains. Tere is no reason advanced here for the abolition Round the decay Of that colossal Wreck, boundless and of a twice declared heritage site with so many triumphs. bare Te lone and level sands stretch far away.’ And memories. Imagine if some administrator decided to abolish a College at Cambridge, including say New- Vale Applied Maths Building ton’s old rooms. It could not happen. And it ought not to have happened here. From the old Applied Maths Build- ing a wide range of immeasurably new technologies Canberra, 2 July 2018 came forth based on fundamental research. Barry W. Ninham Firenze University Press www.fupress.com/substantia

Feature Article Te Lorenz-Lorentz Formula: Origin and Early History Citation: H. Kragh (2018) The Lorenz- Lorentz Formula: Origin and Early History. Substantia 2(2): 7-18. doi: 10.13128/substantia-56 Helge Kragh Copyright: © 2018 H. Kragh. This is Niels Bohr Institute, University of Copenhagen, Blegdamsvej 17, Copenhagen, Denmark an open access, peer-reviewed article E-mail: [email protected] published by Firenze University Press (http://www.fupress.com/substantia) and distribuited under the terms of the Abstract. Among the many eponymous formulae and laws met in textbooks in physics Creative Commons Attribution License, and chemistry, the Lorenz-Lorentz formula merits attention from a historical point of which permits unrestricted use, distri- view. Te somewhat curious name of this formula, which relates the refractive index bution, and reproduction in any medi- of a substance to its density, refects its dual origin in two areas of nineteenth-century um, provided the original author and physics, namely optics and electromagnetism. Although usually dated to 1880, the for- source are credited. mula was frst established in 1869 by L. V. Lorenz (optics) and subsequently in 1878 Data Availability Statement: All rel- by H. A. Lorentz (electromagnetism). Apart from discussing the origin and priority of evant data are within the paper and its the Lorenz-Lorentz formula the paper outlines its early use in molecular physics and Supporting Information fles. physical chemistry. During the late nineteenth century studies of molecular refractivity based on the formula proved important in a number of ways. For example, they led Competing Interests: The Author(s) to estimates of the size of molecules and provided information about the structure of declare(s) no confict of interest. chemical compounds.

Keywords. L. Lorenz, H. A. Lorentz, optical refraction, Clausius-Mossotti formula, molecular refractivity.

1. INTRODUCTION

In 1902 the famous Dutch physicist Hendrik Antoon Lorentz (1853-1928) received the Nobel Prize in physics sharing it with his compatriot Pieter Zee- man. In his Nobel lecture delivered in Stockholm on “Te Teory of Elec- trons and the Propagation of Light” he referred to the refraction of light and the recent insight that the phenomenon was due to vibrating electrical charges (electrons) in the refracting substance. Many years earlier he had succeeded in explaining on the basis of electromagnetic theory “the approximate change in the refractive index with the increasing or decreasing density of the body.” Lorentz continued: “When I drew up these formulae I did not know that Lorenz at Copenhagen had arrived at exactly the same result, even though he started from diferent viewpoints, independent of the electromagnetic theory of light. Te equation has therefore ofen been referred to as the formula of Lorenz and Lorentz.”1 It is the early history of this formula, variously called the Lorentz-Lorenz and the Lorenz-Lorentz formula or law, which is the subject of the present paper. In brief, the formula dates from 1869, when it was frst proposed by

Substantia. An International Journal of the History of Chemistry 2(2): 7-18, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-56 8 Helge Kragh the relatively obscure Danish physicist Ludvig Valentin Lorenz (1829-1891) on the basis of experiments and optical theory. Nine years later it was independently derived on a very diferent basis by 25-year-old Lorentz in the Netherlands, his frst major scientifc work. Te Lor- enz-Lorentz formula, as I shall call it (and justify later), soon became accepted as an important law not only in optics and electromagnetic theory but also as an eminently useful tool in the new feld of physical chemistry. Indeed, chemists embraced the formula at an early date, applying it in various ways to determine the molecular refractivity of chemical compounds and thereby to gain information on their constitution. Ever since the 1880s the Lorenz-Lorentz formula has played a signifcant role in the physical sciences and it continues to do so. Still today, about 150 years afer it was frst proposed, it is an active research area in branches of physical chemistry, crystal chemistry and materials science. Te paper focuses on the period ca. 1870-1890 and in particular on the contributions of the little known Lorenz. A specialist in the mathematical theory of optics, contrary to Lorentz he never accepted Maxwell’s electromagnetic theory and preferred to represent optical phenomena in terms of abstract wave equations with no particular physical interpretation. Although Lorenz, independently of Maxwell, suggested an innovative electrodynamic theory of light in 1867, he did not apply it Figure 1. Newton’s measurements of the “refractive power” (col- to either the refraction or the dispersion of light (but see umn 5) relating the refractive index (column 2) to the density relative to water (column 4). the end of Section 5).2

Te Newton-Laplace rule was tested experimentally 2. REFRACTIVITY AND DENSITY by J.-B. Biot and F. Arago in a work of 1806; the next year their investigations were continued by E. L. Malus. The general idea that the refractivity index n of Although the formula agreed well with the experiments a transparent body is related to its density d was far of the French scientists for gases, it failed miserably for from new at the time when Lorenz took up the subject. liquid and solid bodies. Nonetheless it remained in use As early as in his revised edition of Opticks from 1718, for many years, even afer the corpuscular theory had Newton reported experiments on the refraction of light been replaced by the wave theory of light. in a variety of substances ranging from air to olive oil A simpler and much better expression involving (n – and diamond (Figure 1).3 On the basis of these experi- 1) instead of (n2 – 1) was proposed by an extensive series ments he discussed the possibility of a “refractive pow- of experiments performed during the period 1858-1865 er” of the form (n2 – 1) that varied proportionally to the by the leading British chemist John Hall Gladstone (Fig- body’s density. About a century later Pierre Simon de ure 2) and his collaborator Tomas Dale.4 Te two scien- Laplace, in his famous Mécanique Céleste, derived on tists established that for liquids, the basis of the corpuscular theory of light what became known as the “Newton-Laplace rule.” According to this (n −1) (n −1)v = = constant , rule d n2 −1 ≅ constant d where the quantity v = 1/d is known as the body’s specifc volume. Gladstone and Dale referred to the quantity 5 RGD = (n – 1)/d as the “specifc refraction energy.” Te Lorenz-Lorentz Formula: Origin and Early History 9

Figure 2. J. H. Gladstone (1827-1902). Source: https://en.wikipedia. org/wiki/John_Hall_Gladstone.

Figure 3. Lorenz’s 1869 memoir on “Experimental and Teoretical Te relation was widely used for analyses of solu- Investigations on the Refractivity of Substances” published by the Royal Danish Academy of Sciences and Letters. tions, glasses and crystals, and determinations of the “Gladstone-Dale constant” are still part of modern mineralogy, geochemistry and materials science. How- Ludvig V. Lorenz, a physics teacher at the Military ever, the Gladstone-Dale constant is not a characteristic High School in Copenhagen, was trained as a chemical parameter of the refractive substance as it varies consid- engineer at the city’s Polytechnic College. In the early erably with its physical state. Moreover, the Gladstone- 1860s he established a general, phenomenological theo- Dale rule and other rules proposed in the mid-nine- ry of light from which he claimed that all optical phe- teenth century were basically empirical relations lacking nomena could be deduced.8 Te basis of the theory was a proper theoretical foundation. Te rule was later pro- three partial diferential equations for a so-called light vided with a theoretical justifcation, albeit this proved vector propagating with a velocity equal to the velocity possible only by means of ad hoc hypotheses concerning of light and satisfying the condition that the waves were the structure of the ether.6 It remained an empirical rule, only transversal, not longitudinal. Lorenz had originally practically useful but of limited scientifc importance. suggested that something similar to the Newton-Laplace During the latter half of the nineteenth century sev- rule would follow from his equations, but in 1869 he eral other refractivity-density relations were proposed, arrived at a diferent result.9 In a memoir of that year but these had very restricted applicability and were lit- published by the Royal Danish Academy of Sciences and tle more than extrapolations from a limited number of Letters, of which Lorenz had become a member three experiments. To mention but one example, in 1883, afer years earlier, he reported for the frst time the Lorenz- the Lorenz-Lorentz law had been generally accepted, the Lorentz formula (Figure 3). German chemist W. Johst proposed that n −1 = constant d 3. LORENZ’S OPTICAL ROUTE

From a series of elaborate experiments Lorenz estab- Te formula was discussed for a brief period of time lished in his 1869 memoir a number of empirical formu- afer which it was forgotten.7 10 Helge Kragh lae, for example by measuring the refractive index for the yellow sodium light passing water at diferent temperatures t. In the interval between 0 ˚C and 30 ˚C he found that

2 3 −6 n(t)= n(0)+ ⎣⎡0 .0 7 6 t − 2 .8 0 3 t + 0 .0 0 2 1 3 4 t ⎦⎤1 0

Thus, at a change in temperature of 10 ˚C the observed change in refractivity was found to be only of the order 0.01 per cent. Measurements of this kind had earlier been reported by the French physicist Jules Jamin in 1856, but Lorenz’s data were more precise and in better agreement with later results.10 Te refractive index depends on the wavelength and according to A.-L. Cauchy’s semi-empirical dispersion formula of 1836 the dependency can be represented as Figure 4. Lorenz’s apparatus of 1869 for the determination of the a a a n(λ) = m+ 1 + 2 + 3 …, refractivity-density relationship for liquids. In the tank C a thin λ 2 λ 4 λ 6 tube with the liquid is enclosed between two mirror glasses l and l’. Te two parts of the tank F and F’ and the two small containers h and h’ are flled with distilled water. Te tank is mounted between where the symbols in the nominators are constants to be two Jamin mirrors B and A formed as cubes. One of the light rays determined experimentally. Te quantity m thus denotes passes the tube while the other ray passes the water in the tank with the refractive index reduced to an infnite wavelength or the result that the interference lines are displaced. By measuring the zero frequency, n(λ) → m for λ → ∞. If only the two frst number of displaced lines and the weight of the liquid Lorenz could terms on the right hand are used, we have relate the refractivity of the liquid to its density. a n(λ) = m+ 1 λ 2 investigations made elsewhere was its connection to theory, which he covered in the second part of his treatise. Proceeding from his fundamental wave equation Ten m can be calculated from measurements of Lorenz deduced in 1869 that the quantity (m2 – 1)v/ (m2 two values of n corresponding to two wavelengths λ and 1 + 1) was given by a certain function that only depended λ with the result that 2 on the distribution in space of the refractive substance. Since it was known from the Gladstone-Dale rule that 2 2 λ1 n1 − λ2n2 (m – 1)v was approximately constant, Lorenz concluded m = 2 2 λ1 − λ2 that the correct law of refractivity was given by what he called the “refraction constant,” namely Having discussed his own data and those report- m2 −1 v = constant (= RLL ) ed by other scientists, Lorenz concluded that m only m2 + 2 depends on the density and that the temperature merely enters indirectly, namely by changing the volume Tis result was independent of the form of the mol- and hence the density. He ended up with the following ecule, he argued. However, for reasons of simplicity he expression for water: assumed the refractive medium to be composed of optically homogeneous spherical molecules with mi being 2 3 −6 m(t)=1 .3 2 1 9 + ⎣⎡2 1 .0 5 t − 2 .7 5 9 t + 0 .0 2 1 3 4 t ⎦⎤1 0 their internal refractive index. With vi being the specifc proper volume of the molecules Lorenz could then write the law as Although Lorenz’s experimental work was of unsur- passed precision (Figure 4), it did not difer essential- 2 2 ly from similar measurements made in German and m −1 mi −1 2 v = 2 vi French laboratories. What distinguished his work from m + 2 mi + 2 Te Lorenz-Lorentz Formula: Origin and Early History 11

He further argued that the reduced refractive index a detailed summary of his two communications on opti- was approximately constant and for a mixture consisting cal refraction originally published in two sequels in the of k non-interacting components could be expressed as proceedings of the Royal Danish Academy.13 Using a new and simpler approach he derived the same expres- 2 k 2 sion for the relation between refractivity and density as m −1 mj −1 v = v j in his earlier theory, namely a constant value of the ratio m2 + 2 ∑m2 + 2 j=1 j (n2 – 1)/d(n2 + 1). It was only on this occasion that the international community of physicists became aware of Te observation turned out to have signifcant con- his extensive work on the refractivity-density law. Since sequences for chemical investigations. For an isotropic his memoirs of 1869 and 1875 were written in Danish, substance consisting of only one kind of molecule he they were known only by scientists in Scandinavia. deduced the approximate relation

2 2 4. OPTICAL REFRACTION AND MOLECULAR m −1 ⎛ k ⎞ v = P 1− PHYSICS m2 + 2 ⎝⎜ v 2 ⎠⎟ Lorenz was convinced that optical research provided Here P and k are two constants that depend on the a method to obtain information about the size of mol- molecular structure of the substance but not on its vol- ecules and their number in a volume or mass unit of a ume or temperature. For a gas, where v is large and m substance (Figure 5). In his 1875 paper he derived that only slightly larger than 1, for a substance composed of spherical and optical homogeneous molecules,

m2 −1≅ 2(m−1) and m2 + 2 ≅ 3 n 2 −1 n 2 −1 v = i v (1+δ ), n 2 + 2 n 2 + 2 i Lorenz noted that the expression above approxi- i mates to With β being a measure of the molecular radius, he 3 (n−1)v = P stated the δ quantity as 2

2 2 16 2 n i −1 β in agreement with the Gladstone-Dale formula. Moreo- δ = π 2 2 5 n + 2 λ ver, the Lorenz expression also accommodates the New- i ton-Laplace rule since

2 According to Lorenz, it followed from experiments n −1 2 that for λ = 589.3 nm (sodium light) the value of δ was = RLL (n + 2) ≅ 3RLL d approximately 0.22. Lorenz used this result for two purposes. First, he Only afer a period of six years did Lorenz return to pointed out that since δ = δ(λ-2) the expression explained his studies of refraction, this time in a predominantly dispersion, if only qualitatively, without relying on spe- experimental paper where he reported measurements on cial assumptions about molecular forces or the structure oxygen, hydrogen, water vapour, ethanol, ether and oth- of the ether. Tis contrasted with Cauchy’s earlier theory er volatile liquids.11 Lorenz’s law of refractivity, derived of dispersion which relied on such assumptions and also as a theoretical consequence of his theory of light, was unable to explain why dispersion does not take place received solid confrmation in 1880, when the Danish in void space. In Lorenz’s very diferent theory, disper- physicist Peter K. Prytz published extensive measure- sion was a property of the heterogeneity of a substance ments on the refractive constants of a variety of liquids and thus excluded dispersion in a vacuum. Importantly and vapours. Te measurements showed convincingly and contrary to other optical theories at the time, Lor- that Lorenz’s law was superior to the Gladstone-Dale enz’s theory did not assume the existence of an ethereal rule.12 medium. Prytz’s 1880 paper in Annalen der Physik und Che- Te second use he made of his result was to estimate mie was preceded by a paper in which Lorenz presented a lower limit to the size of molecules. In Lorenz’s theo- 12 Helge Kragh

Using a diferent optical method based on the scattering of light on a small sphere, in an important memoir of 1890 Lorenz refned his value of β. He also esti- mated a value for the number of molecules in one mil- lilitre of a gas, a quantity known as Loschmidt’s number (NL) and named afer the Austrian physicist and chemist Josef Loschmidt.16 Te better known Avogadro number NA is given by

−1 3 NA = 6 .0 2 2 mole ≅ 2 2 .4 1 0 × NL

19 Lorenz reported NL = 1.63×10 while the modern 19 value is NL = 2.688×10 .

5. LORENTZ’S ELECTROMAGNETIC ROUTE

Lorenz’s law of refractivity is today referred to as the Lorenz-Lorentz law, or more commonly the Lorentz- Lorenz law, because H. A. Lorentz (Figure 6) derived the same result in 1878.17 Just the year before, he had been appointed professor of theoretical physics at the Uni- versity of Leiden, at the tender age of 24. In his doctoral dissertation of 1875 Lorentz referred to the refractive index of various substances as given by their dielectric constants.18 He briefy discussed the Newton-Laplace Figure 5. Ludvig V. Lorenz. Royal Library, Copenhagen, Picture formula relating the refractive index to the density but at Collection. the time without suggesting an improved law based on the electromagnetic theory. Contrary to the Danish physicist, in his memoir of ry the quantity β was not literally the molecular radius 1878 Lorentz obtained the improved law by combin- but what he cautiously called “the radius of the molecu- ing the Clausius-Mossotti formula (see below) with the lar sphere of action, meaning the sphere surrounding a electromagnetic theory of light. However, he did not rely molecule within which there is an appreciable efect of primarily on Maxwell’s theory but rather on an alterna- the molecule’s infuence on the velocity of light propaga- tive action-at-a-distance theory proposed by Hermann tion.” Tis quantity is greater than the actual or material von Helmholtz.19 At the time Maxwell’s field theory radius of the molecule. Since was generally considered to be very difficult, almost impenetrable. Although Lorentz appreciated the theory, 2 n i −1 he thought that it depended too much on unconfrmed 2 <1 n i + 2 hypotheses. What is known as the Clausius-Mossotti formula was frst proposed, if only implicitly, by the Italian phys- and with δ known, Lorenz was able to conclude that icist Ottaviano Fabrizio Mossotti in 1847. Much later the formula was stated by Rudolf Clausius in 1879 in an β >1 .5 ×1 0 −8 m attempt to explain the dielectric properties of insulators on an atomistic basis. From a historical point of view the order of names is perhaps unfortunate, but “Mossotti- He was pleased to note that the German physicist Clausius” is rarely used. With εr the material’s dielectric Georg Hermann Quincke from recent measurements constant (or relative permittivity ε /ε0 and α denoting of viscosity and capillarity had found molecular radii the polarizability of the molecule, the Clausius-Mossotti agreeing with the limit inferred from the optical meth- formula for a unit volume with N molecules is od.14 Te Lorenz-Lorentz Formula: Origin and Early History 13

molecule and characterised by its electric polarizability. He thought of the material molecule as being situated at the centre of a sphere or cavity, an idea which can also be found in Lorenz’s paper of 1875. Lorentz thus pictured the molecules as embedded in an all-pervading ether, which he, much like Maxwell, regarded as a dielectric substance. He emphasised the necessity of assuming inter-molecular space being flled with ether, a belief he stated was “not open to doubt.” Lorenz, on the other hand, had dismissed the ether as superfuous and even “unscientifc” in his electrical theory of light from 1867 and it played no role whatsoever in his optical theory two years later. Afer a series of complex calculations Lorentz ended up with the following expression:

3 4π 3 ρ 2 ρ 3+ 4πε − 4πε n −1 ( 0 ) 0 = 3 κ (n2 + 2)d ρ3 M(3+ 8πε0 ) − 8πε0 κ

Here ε0 denotes the dielectric constant of the free ether, M is the mass of a molecule, d the density of the body, and κ is the ether’s specifc resistance according to Helmholtz’s theory. Te quantity on the right side of the equation is thus a constant for a particular transparent body. In the last part of his extensive 1878 memoir Figure 6. Hendrik A. Lorentz in 1927. Courtesy the Niels Bohr Lorentz compared his theoretical law of refraction with Archive, Copenhagen. available experimental data from the literature. Unlike Lorenz, he did not perform experiments of his own. In its modern formulation the Lorenz-Lorentz law is ε −1 4π stated as a relation between the refractive index of a sub- r = Nα stance, a macroscopic quantity, and its polarizability α, a ε + 2 3 r microscopic quantity: n2 −1 4π In modern literature this expression, which for = Nα , 2 + N = NA is called the molar refractivity, is ofen used syn- n 2 3 onymously for the Lorenz-Lorentz formula. In a paper of 1910 on the theory of opalescence Einstein appropriately When the polarizability is small, the equation referred to it as the Clausius-Mossotti-Lorentz formula.20 reduces to Modern physicists sometimes use the more cumbersome name Clausius-Mossotti-Lorenz-Lorentz (CMLL) formula. 2 Te declared purpose of Lorentz’s work was to con- n −1≅ 4πNα or n−1≅ 2πNα struct a theory of the optical properties of matter, such as indicated by the title of his memoir, which in English In agreement with the Gladstone-Dale formula, this reads “Concerning the Relation between the Velocity of expression is valid for gases at normal pressure. It fol- Propagation of Light and the Density and Composition lows from the Lorenz-Lorentz theory that the polariza- of Media.” Contrary to his Danish near-namesake, Lor- tion of a molecule in a solid body under the infuence entz considered a molecular or atomic model in connec- of an external electric feld is not only determined by tion with his theory, namely that a molecule consists of the strength of the feld and the number of molecules an electric charge harmonically bound to the rest of the per volume. Tere is also an efect due to the polarized 14 Helge Kragh

Lorenz, he summarised and discussed the two Danish articles whereas Lorentz’s German paper was a substantially reduced and revised version of his 1878 memoir published in Dutch in the proceedings of the Amster- dam Academy.24 In a series of lectures delivered at Columbia Univer- sity, New York, in 1906, Lorentz called the double discovery “a curious case of coincidence.”25 Referring to the Annalen papers of 1880, the British physicist Arthur Schuster wrote a few years later that “two authors of similar name, H. A. Lorentz of Leyden, and L. Lorenz of Figure 7. Lorenz’s unpublished derivation of the relationship Kopenhagen [sic], have almost simultaneously published 2 2 between the dielectric constant (D) and the specifc refractivity (N). investigations leading to the result that (μ – 1)/(μ + 2) D is constant.”26 Again, when awarding Lorentz the Rumford Medal in 1909, Lord Rayleigh said about the neighbour molecules which produce an additional force. formula that it had been “reached simultaneously, along Tis force was in the earlier literature sometimes called diferent special lines, by H. A. Lorentz originally from the “Lorentz-Lorenz force,” a name which should not be Helmholtz’s form of Maxwell’s electric theory, and by L. confused with the well-known Lorentz force acting on Lorenz, of Copenhagen, from a general idea of propaga- an electrical charge moving in a magnetic feld.21 tion afer the manner of elastic solids.”27 Although Lorenz never referred to the electromag- However, given that Lorenz published his result as netic derivation of the Lorenz-Lorentz law in his pub- early as 1869 the curious coincidence does not constitute lications, in an unpublished manuscript from 1887 a proper case of simultaneous discovery. Robert Merton he used his own electrodynamic theory to derive the and other sociologists of science have long ago noted that law.22 Based on his electrical theory of light from 1867 discoveries in science are rarely made by a single scien- he found an expression for the refractive index and its tist or group of scientists. Discoveries are almost always dependence on the molecular currents elements (Figure “multiples,” meaning that the same or nearly the same 7). In this way he concluded wholly independently of discovery is made by two or more scientists (or groups of Maxwell’s theory that the medium’s dielectric constant scientists) working independently of each other.28 Mul- was given by the expression tiple discoveries may be more or less simultaneous, but the important thing is not so much the chronology as it is that they are made independently. Merton proposed, ε ≡ ε ε = 2 r / 0 n somewhat artifcially, that even discoveries far removed from one another in time may be conceived as “simulta- He thus arrived at the very same relationship as neous” in what he called “social and cultural time.”29 found by Lorentz. Applying the notion of simultaneity in its ordinary meaning there is no doubt that priority to the Lorenz- Lorentz law belongs to L. Lorenz and, consequently, 6. LORENZ-LORENTZ OR LORENTZ-LORENZ? that it should not be referred to as the Lorentz-Lorenz law. On the other hand, the discovery was not Lorenz’s Because H. A. Lorentz originally published his paper alone, what Merton called a “singleton.” It can be seen as in Dutch, and L. V. Lorenz published his two papers a classic example of a multiple discovery, in this case a of 1869 and 1875 in Danish, the Lorenz-Lorentz for- “doublet” separated in time by nine years. mula became generally known only when abridged and Te order Lorenz-Lorentz can be found in the lit- revised versions of their papers appeared in German in erature in the 1890s, but with the rising fame of the 1880. Both of the papers were published in Annalen der Dutch physicist the order was soon reversed or Lorenz Physik und Chemie but in two diferent issues and with simply lef out. In an obituary article on Lorentz, Max Lorentz’s as the frst. Gustav Heinrich Wiedemann, the Planck referred to the formula relating refractivity and editor of Annalen, had originally planned to have the density, “which by accident had been established at the two papers published consecutively, but for some reason same time by his namesake, the Danish physicist Ludvig this did not happen.23 Apparently the two authors were Valentin Lorenz, and for this reason has been assigned at the time unaware of each other’s work. In the case of the curious double name Lorentz-Lorenz.”30 Much later Te Lorenz-Lorentz Formula: Origin and Early History 15 we fnd the same usage in the authoritative textbook Tis was also the conclusion of the Austrian physi- on optics written by Max Born and Emil Wolf: “By a cist Franz Exner, at the University of Vienna, who in remarkable coincidence, the relation was discovered 1885 stated that the Lorenz-Lorentz law had been “com- independently and practically at the same time by two pletely confirmed.”34 As Exner pointed out, the law scientists of almost identical names, Lorentz and Lorenz, served as a key instrument for obtaining information and is accordingly called the Lorentz-Lorenz formula.”31 about the size and constitution of molecules and the As an illustration of the popularity of the two terms, range of the unknown molecular forces. For the diam- Google Scholar gives ca. 14,900 results for “Lorentz-Lor- eter of gas molecules he suggested the formula enz” and ca. 3,700 results for “Lorenz-Lorentz.” Te pref- 2 − erence for the frst eponymous term can be seen as an n 1 D = C 2 ℓ , example of a general tendency to associate a discovery n + 2 with the name of a famous scientist even in cases where priority belongs to someone else.32 Te Clausius-Mossot- where C is an empirical constant and ℓ the mean free ti formula is another example. Google Scholar also lists path of the molecules. Combining refractivity and difu- the number of references to the two Annalen papers of sion measurements Exner found D = 10-10m for air mol- -10 1880, namely 192 references to Lorenz’s paper and 644 ecules (N2, O2) and D = 2.7×10 m for CS2 vapour. references to Lorentz’s. Te subject of molecular refractivity belonged as much to chemistry as to physics. Indeed, refractivity studies had been part of theoretical chemistry many 7. A TOOL FOR PHYSICAL CHEMISTRY years before the Lorenz-Lorentz formula. Te new formula further stimulated this kind of work which played Afer the refraction studies of Lorenz and Lorentz a most important role in the new discipline of physical had become widely known they spurred a large number chemistry that emerged during the 1880s. When Lorenz of experiments in molecular refractivity under various and Lorentz fgure in books on the history of chemis- conditions. Te overall result of this work was that the try, and not only in those on the history of physics, it is Lorenz-Lorentz law agreed far better with experimental principally because of their role in the Lorenz-Lorentz data than competing formulae of an empirical nature. formula.35 By the turn of the century the formula and In a review paper of 1888 the British physicist Arthur related refractivity studies had become a staple part of William Rücker referred to the works of L. Lorenz and textbooks in physical chemistry.36 H. A. Lorentz as well as to Prytz’s experimental confr- Te leading Swiss chemist Hans Heinrich Landolt mation of the law named afer them. Rücker found it of and his German colleagues Wilhelm Ostwald and Julius particular interest that the measurements of Lorenz and Wilhelm Brühl were among those who applied the Lor- Prytz indicated that the value of (n2 – 1)/(n2 + 2) did enz-Lorentz formula to calculate the so-called molecu- not depend on whether the substance was in a liquid or lar refractivity (or refractive power) of a particular sub- a vaporous state (Table 1). Having reviewed the experi- stance. Tey defned this quantity as the product of the mental data Rücker concluded that “Te results, on the specifc refractivity RLL and the molecular weight M, whole, confrm the accuracy of the physical meaning of that is, with n determined at a particular wavelength, 2 2 the expression (n – 1)/(n + 2) and tend to show that the 2 diameter of the molecule is the same in the liquid and n −1 M MRLL = ⋅ gaseous state.”33 n2 + 2 d

In cases where the Gladstone-Dale formula was used, the molecular refractivity was similarly given by Table 1. Data for the quantity (n2 – 1)/(n2 + 2) obtained by Lorenz and Prytz. Source: Rücker (1888). M MRGD = (n −1) Substance Formula Work Liquid Vapour d

Ethyl ether (C2H5)2O Lorenz (1875) 0.30264 0.3068 Ethanol C H OH Lorenz (1875) 0.28042 0.2825 Te monochromatic light used in most experiments 2 5 was either the yellow sodium D line (λ = 589 nm) or the Water H2O Lorenz (1875) 0.20615 0.2068 red Hα line in the spectrum of hydrogen (λ = 656 nm). Methanol CH3OH Prytz (1880) 0.2567 0.2559 It turned out that in many cases the summation rule Methyl acetate (CH3)2COO Prytz (1880) 0.2375 0.2399 for mixtures could be carried over to chemical com- Ethyl formate C2H5COOH Prytz (1880) 0.2437 0.2419 16 Helge Kragh pounds, such as suggested as early as 1863 in a paper by 37 Gladstone and Dale. If a compound consists of q1, q1,… elements with atomic weights μ1, μ1,… then the molecular weight is

M = q1µ1 + q2µ2 +…

According to the summation rule the molecular refractivity r = RLL is simply the weighted sum of the individual atomic refractivities given by

2 Figure 8. Based on thermochemical arguments the Danish chemist n i −1 µi ri = 2 ⋅ Julius Tomsen proposed in 1886 an octahedral model of benzene n i + 2 di in which there were no double bonds. Tomsen’s structural model was among those which Brühl dismissed as incompatible with Tat is, refractivity data based on the Lorenz-Lorentz formula. n2 −1 M ⋅ = q µ r + q µ r +… n2 + 2 d 1 1 1 2 2 2 of chemistry, but it is more surprising in the case of the chemically trained Lorenz. Experiments showed that although the rule was In fact, at the end of his 1880 paper Lorenz dealt approximately correct for many compounds it was not with a number of chemical reactions during which the universally true. In several cases the molecular refractiv- refractivity constant changed. From his own and oth- ity difered substantially from the sum of the constituent ers’ experiments he suggested that the change in refrac- atomic refractivities or, diferently phrased, a particular tivity might constitute a measure of the chemical afn- atomic refractivity did not always have the same value. ity in the same way as the change in heat (Q) did in the 39 It was soon recognised that the molecular refractivity thermochemical Thomsen-Berthelot theory. Lorenz is infuenced also by the constitution of the molecule as suspected that exothermic processes were followed by a given by the arrangement of atoms and the presence of decrease in refractivity and endothermic processes by an double and triple bonds. increase. However, he admitted that the case of ammo- The pioneer in this branch of optical chemistry nia was J. W. Brühl, who employed the Lorenz-Lorentz formula in a series of elaborate studies of inorganic as well N2 + 3H2 → 2NH3 +Q as organic substances. By considering the refractivity values of compounds in homologous series he derived the corresponding values for double and triple bonds was an exception to the rule. Te molecular refractivity in molecules. He applied this method to the vexed and of NH3 was known to be 0.3266 and Lorenz’s measure- much-discussed question of the constitution of benzene, ments of a mixture of N2 and H2 in the mass ratio μN : 3μH = 14 : 3 resulted in 0.3116. C6H6. On the assumption of Kekulé’s structural model with three single and three double bonds Brühl found a theoretical value for benzene’s molecular refractivity that only difered 0.6% from the measured value. On the 8. CONCLUSION other hand, he concluded that alternative formulae suggested by H. Armstrong, A. von Baeyer, J. Tomsen and The Lorenz-Lorentz law is a general, non-trivial others did not agree with benzene’s molecular refractiv- relationship between the refractive index of a sub- ity (Figure 8).38 stance and its density. Te origin of the eponymous While the Lorenz-Lorentz formula aroused great law – or perhaps better formula – is traditionally dated interest in the chemical community, none of the found- 1880 and considered an example of a simultaneous dis- ers of the formula took much interest in the chemical covery made independently by the two physicists afer applications. Tis is perhaps understandable in the case which it is named. However, although Lorentz came to of Lorentz, who had neither interest in nor knowledge the law independently of Lorenz and the discovery was thus a “doublet,” it is not a simultaneous discovery since Te Lorenz-Lorentz Formula: Origin and Early History 17 the Danish co-discoverer formulated the law already in 6. W. Sutherland, Phil. Mag. 1889, 27, 141. 1869, nine years before Lorentz. For this reason I pro- 7. W. Johst, Ann. Phys. Chem. 1883, 20, 47. pose that the law should properly be called the Lorenz- 8. L. Lorenz, Phil. Mag. 1863, 26, 81. For Lorenz’s con- Lorentz law although most physicists and chemists pre- tributions to optics, see H. Kragh, Appl. Optics 1991, fer the other permutation.40 30, 4688 and O. Keller, Progress in Optics 2002, 43, Te routes of the two physicists to the refractivity- 195. density law were entirely diferent both as regards for- 9. L. Lorenz, Kgl. Da. Vid. Selsk. Skrifer 1869, 8, 205. malism and physical interpretation. And yet they arrived French translation in H. Valentiner, ed., Oeuvres Sci- at exactly the same formula. In the physical sciences it is entifques de L. Lorenz, 2 vols., Royal Danish Acade- not unusual that the same result can be derived in dif- my of Science, Copenhagen, 1898-1904. ferent ways and therefore is not uniquely determined by 10. J. Jamin, Compt. Rend. 1856, 43, 1191. the underlying theory. From a modern point of view the 11. L. Lorenz, Kgl. Da. Vid. Selsk. Skrifer 1875, 10, 483. theory behind the Lorenz-Lorentz law is simply Max- 12. P. K. Prytz, Ann. Phys. Chem. 1880, 11, 104. well’s theory of electromagnetism, but Lorenz’s original 13. L. Lorenz, Ann. Phys. Chem. 1880, 11, 70. formulation had nothing to do with that theory. Aware of 14. G. Quincke, Ann. Phys. Chem. 1869, 137, 402. the dual origins of the law, Wilhelm Ostwald commented 15. L. Lorenz, Kgl. Da. Vid. Selsk. Skrifer 1890, 6, 1. that “this agreement between two completely diferent Te paper is today recognized as the foundation of approaches increases the probability that the result has a so-called Mie-Lorenz scattering theory. more general signifcance than if it were based on one or 16. R. M. Hawthorne, J. Chem. Educ. 1970, 47, 751. the other of the theoretical foundations.”41 17. H. A. Lorentz, Collected papers, vol. 2, pp. 1-119, Whatever its theoretical background and interpre- Martinus Nijhof, Te Hague, 1934-1936. tation, the Lorenz-Lorentz law was eminently successful 18. H. A. Lorentz, Collected papers, vol. 1, pp. 193-383, and instantly applied to the study of molecular refractiv- Martinus Nijhof, Te Hague, 1934-1936. ity and related branches of chemistry, physics and mate- 19. E. A. Woodruf, Isis 1968, 59, 300; T. Hirosige, Hist. rials science. By the early twentieth century it was pre- Stud. Phys. Sci. 1969, 1, 151. dominantly a resource for the new generation of physi- 20. A. Einstein, Ann. Phys. 1910, 33, 1275. cal chemists rather than a topic belonging to theoretical 21. Handbuch der Physik, vol. 20, pp. 503-505, J. Sprin- physics. ger, Berlin, 1928. 22. Manuscript dated 1 June 1887; Lorenz Papers, Royal Danish Academy of Science. ACKNOWLEDGMENTS 23. Wiedemann to Lorenz, 7 May 1880; Lorenz Papers, Danish Museum of Science and Technology. My thanks to Christian Joas, the Niels Bohr 24. H. A. Lorentz, Ann. Phys. Chem. 1880, 9, 641; L. Archive, Copenhagen, for helpful comments on an ear- Lorenz, Ann. Phys. Chem. 1880, 11, 70. lier version of the paper. 25. H. A. Lorentz, Te Teory of Electrons, Dover Publi- cations, New York, 1909. 26. A. Schuster, An Introduction to the Teory of Optics, REFERENCES p. 284, Edward Arnold, London, 1909. 27. Lord Rayleigh, Proc. Roy. Soc. A 1909, 82, 1. 1. https://www.nobelprize.org/nobel_prizes/physics/lau- 28. W. F. Ogburn, D. To m a s , Pol. Sci. Quart. 1922, 37, reates/1902/lorentz-lecture.html. 83; R. K. Merton, Te Sociology of Science, pp. 343- 2. L. Lorenz, Phil. Mag. 1867, 34, 287. E. Whittaker, A 370, University of Chicago Press, Chicago, 1973; D. History of the Teories of Aether and Electricity, pp. Lamb, S. M. Easton, Multiple Discovery, Avebury 267-270, Nelson and Sons, London, 1958; W. Kaiser, Publishing, Avebury UK, 1984. Teorien der Elektrodynamik im 19. Jahrhundert, pp. 29. Ref. 28 (Merton), 369. 157-162, Gerstenberg Verlag, Hildesheim, 1981. 30. M. Planck, Naturwissenschafen 1928, 16, 549. 3. I. Newton, Opticks, pp. 270-276, William Innys, Lon- 31. M. Born, E. Wolf, Principles of Optics, p. 87, Pergam- don, 1718. on Press, Oxford, 1970; J. D. Jackson, Classical Elec- 4. J. H. Gladstone, T. Dale, Phil. Trans. Roy. Soc. 1858, trodynamics, p. 119, John Wiley & Sons, New York, 148, 887. 1962. 5. J. H. Gladstone, T. Dale, Phil. Trans. Roy. Soc. 1863, 32. T. F. Gieryn, ed., Science and Social Structure, pp. 153, 317. 147-158, New York Academy of Sciences, New York, 18 Helge Kragh

1980; J. D. Jackson, Am. J. Phys. 2008, 76, 704. 37. J. H. Gladstone, T. Dale, Phil. Trans. Roy. Soc. 1863, 33. A. W. Rücker, J. Chem. Soc. Trans. 1888, 53, 222. 153, 217. 34. F. Exner, Sitzungsber. Kais. Akad. Wiss. 1885, 91 (2), 38. J. W. Brühl, Zs. Phys. Chem. 1887, 1, 307; Brühl, Ber. 850. Deut. Chem. Gesselsch. 1891, 24, 1815. For the debate 35. A. J. Berry, From Classical to Modern Chemistry, pp. concerning the structure of benzene, see A. J. Rocke, 88-90, Dover Publications, New York, 1968; A. J. Ann. Sci. 1985, 42, 355 and S. G. Brush, Stud. Hist. Ihde, Te Development of Modern Chemistry, p. 393, Phil. Sci. 1999, 30, 21. Dover Publications, New York, 1984. 39. H. Kragh, Brit. J. Hist. Sci. 1984, 17, 255. 36. E.g., J. H. van’t Hof, Vorlesungen über Teoretische 40. Te case for Lorenz-Lorentz over Lorentz-Lorenz und Physikalische Chemie, part III, pp. 75-76, Vieweg was briefly argued in A. Sihvola, IEEE Antennas und Sohn, Braunschweig, 1903; W. Nernst, Teoreti- Prop. Mag. 1991, 33, 56. cal Chemistry, pp. 306-313, Macmillan and Co., Lon- 41. W. O s t w a l d , Grundriss der Allgemeinen Chemie, p. don, 1904. See also the comprehensive review in S. 133, W. Engelmann, Leipzig, 1899. Ostwald mistak- Smiles, Te Relations between Chemical Constitution enly thought that Lorenz’s optical theory belonged to and some Physical Properties, pp. 239-324, Long- the older tradition of elastic ether theories. mans, Green and Co., London, 1910. Firenze University Press www.fupress.com/substantia

Feature Article 2001: Te Crystal Monolith

Citation: J.M. Garcia-Ruiz (2018) 2001: The Crystal Monolith. Substantia Juan Manuel Garcia-Ruiz 2(2): 19-25. doi: 10.13128/substantia- 57 Laboratorio de Estudios Cristalográfcos. Instituto Andaluz de Ciencias de la Tierra. CSIC-Universidad de Granada. Spain Copyright: © 2018 J.M. Garcia-Ruiz. E-mail: [email protected] This is an open access, peer-reviewed article published by Firenze University Press (http://www.fupress.com/substan- Abstract. In the famous movie “2001: A Space Odyssey”, Stanley Kubrick and Arthur tia) and distribuited under the terms Clarke claim that an extraterrestrial civilization catalyzed the evolution of hominids of the Creative Commons Attribution on our planet. To represent such a powerful civilization, they use a crystal. To date, License, which permits unrestricted it seems that we have not been contacted by advanced civilizations and that we are use, distribution, and reproduction in any medium, provided the original alone to manage our own future. Yet Kubrick and Clarke perhaps intuitively touched a author and source are credited. truth about the power of crystals. An argument is developed here that genuine crystals, mainly quartz single crystals, were the earliest catalysts of the abstract thinking, sym- Data Availability Statement: All rel- bolism, and consciousness. evant data are within the paper and its Supporting Information fles. Keywords. Crystals, crystallography, abstract thinking, paleoneurobiology, Kubrick, Competing Interests: The Author(s) “2001: A Space Odissey”. declare(s) no confict of interest.

THE MONOLITH OF “2001: A SPACE ODYSSEY”

Art, in its very diferent forms, has contributed and contributes almost as much as science and philosophy to create our conception of the outside world. For instance, most citizens believe nowadays that we are not alone in the universe, in spite of the fact that we have no evidence of the existence of extraterrestrial life even in the closest planets and their moons. Te science-fction movie “2001: A Space Odyssey”1 and the novel based on its script is one of the masterpieces of flm that has contributed in large measure to our vision about the great question of the habitability of the universe. Te movie, released in 1968, was directed by Stanley Kubrick based on the script he wrote with the science-fction writer Arthur Clarke. Te screenplay was based, in turn, on Clarke’s 1951 story, “Sentinel of Eternity”.2 Te plot in both cases tells of the tangible existence of alien civilizations much more advanced than our own. How, in a novel or a flm, could such a mighty civilization, capable of modifying humankind’s evolution, be visually represented? What would an alien Big Brother look like? It had to be something that would evoke ultrahuman power, something enigmatic, disturbing and even fearsome, a secular image of God. Clarke and Kubrick described their vision of the star character of the movie, the monolith, with the following words:

It was a rectangular slab, three times his height but narrow enough to span with his arms, and it was made of some completely transparent material; indeed, it

Substantia. An International Journal of the History of Chemistry 2(2): 19-25, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-57 20 Juan Manuel Garcia-Ruiz

was not easy to see except when the rising sun glinted on its edges. As Moon-Watcher had never encountered ice, or even crystal clear water, there were no natural objects to which he could compare this apparition. It was certainly rather attractive, … (Figure 1).3

Tey chose a parallelepiped, a transparent slab, perfectly smooth with sharp edges and dihedral ninety- degree angles. In a word, they chose a crystal. As a completely transparent monolith was ill-suited to the art of flm, Kubrick was compelled to darken the transparent slab, but both authors agreed that the icon had to be a crystal. In fact, in the early versions of the 2001 screen- Figure 1. A frame of the movie “2001: A Space Odyssey” directed play,4 the monolith was specifed as “a cube about ffeen by Stanley Kubrick. Te black monolith was a crystal clear cube feet on a side, and it is made of some completely transpar- in the original script but afer considering the technical problems ent material”.5 Tey wrote: “the hominids watch, wide- Kubrick decided to use a black cube and then a black slab. eyed, mesmerized captives of the Crystal Cube.6 Moreo- ver, the sequence entitled “Killing the lion” ended with Clarke and Kubrick belong to a group of intellectu- the following phrase: “And then one night the crystal als who believed in the existence of advanced alien civi- cube was gone, and not even Moonwatcher ever thought lizations capable of traveling across the universe that of it again. He was still wholly unaware of all that it had could have purposefully altered the evolution of life on done”. Te fnal version of the screenplay also explicitly our planet,10 a theory that can be seen as a secular ver- talks of a “crystalline monolith”;7 and the hypnotizing sion of Christian visitation. More than ffy years afer sound that attracts Moon-Watcher “pulsed out from the the flm’s premiere, we are yet to have any contact with crystal”.8 Clarke also had chosen a crystal for “Sentinel of an extra-terrestrial civilization. We have already visited the Eternity”: the machine lef purposefully on our moon Saturn and Jupiter, we have found in their moons jets of by the alien civilization was a crystal pyramid. He wrote: water11 and clouds of organic compounds12, but no sign “Perhaps you understand now why that crystal pyramid of life yet. Our rockets and probes have traveled beyond was set upon the Moon instead of on the Earth”.9 Te our solar system. It is worth remembering that we have sentinel that will alert that humans have the technol- been emitting radio waves into outer space for some 130 ogy to cross the space was a “crystal pyramid” with hard, years. At the speed they travel, we have already reached scratch-proof “crystal walls”, most probably a diamond. celestial bodies that are 130 light-years away, a distance Te choice of a crystal to represent a supernatural far enough to reach other solar systems such as alpha- intelligence, or of any machine built by it, was inevitable. centauri,13 or the Trappist-1 and its seven planets where In everyday language, the word crystal evokes concepts NASA optimistically pushes astrobiological groups to such as order, purity, transparency, harmony, perfection, search for life.14 Tis means that within a distance of reason, intelligence… and power. Tese are all justifed 65 light years from Earth either there is no life, or the because they allude to the physical and chemical prop- life that there is either has not noticed (it is not capa- erties that have characterized crystals throughout his- ble of hearing us) or does not know how to respond. tory, and to how these properties have been transmuted Or they are a bunch of intelligent but rude aliens that to our cultural heritage through the arts and philosophy. do not want to know anything about us (which, given Humans have held a fascination for crystals since primi- the state of the management of our planet, would be a tive times, and even today it is thought that crystals sign of intelligence). All in all, we need to sort out our hold some enigmatic power. Since the very formation intraplanetary disputes and environmental challenges, of our consciousness, but mainly since the discovery of because it does not seem that anybody out there is going their three-dimensional order in the nineteenth century, to come and lend us a hand crystals have represented the exact opposite of the ata- vistic, the biological, and the human. Tus, the image of a device placed on purpose by an advanced alien civili- CRYSTALS zation or from that civilization had to be a crystal. Te selected shape, a pyramid, or cube or slab, was chang- Having taken the position that the vision of Kubrick ing through the diferent stages of the script – but it was and Clarke may remain science fction for the foresee- always thought to be a crystalline polyhedron. 2001: Te Crystal Monolith 21

Figure 2. Collection of quartz crystals from the Acheulian site of Singi Talav. From reference 18. (D’Errico). able future, I do not dispute the possibility that the two geniuses responsible for the masterpiece 2001 were not © An idea of Juan Manuel Garcia-Ruiz. Photography Javier Trueba. so misguided on the role that crystals have played in Figure 3. A recreation of the collection of crystals by Homo erec- evolution. We now know that the frst objects that homi- tus. ©Javier Trueba/Juan Manuel García-Ruiz. With permission. nids collected without any applied purpose were quartz crystals. A set of varied, irrefutable proofs have been gathered and in part contributed by paleoart experts nor do they have perforations or signs of use as trinkets such as Robert Bednarik15 and James Harrod.16 For or jewels. No, they were objects considered valuable in of example, Pei Wenzhong, who discovered Peking Man, themselves. Tey were esteemed in the Acheulean, they published in 1931 the discovery of twenty quartz crystals continued to be so in the pre-historical and historical in the famous Zhoukoudian cave along with the remains eras. And there is no evidence that the human fascina- of Homo erectus that date from 700,000 years ago.17 One tion for crystals is waning even today. Te question is of them was a perfectly faceted, smoky quartz crystal, a unavoidable: why were those hominids, still without a hexagonal prism, biterminated in pyramids of some six developed consciousness, drawn to those quartz crys- centimeters in length. In 1989, at the famous archaeolog- tals? Why did they value them and carry them as pre- ical site of Singi Talav in India, six practically complete cious treasures? (Figure 3). quartz prisms from the Lower Acheulean strata (300,000 When Homo erectus raised his head and looked at – 150,000) were found (Figure 2). Tese prisms are natu- the African savannah or the Asian forests, everything ral, have not been modifed and measure between 7 and he saw was curved or branched. Te trees, the bushes, 25 mm in length.18 Smaller quartz crystals were exca- the furrows carved by water, the streams, the clouds, vated at the Acheulean site of Gesher Benot Ya’aqov, in the mountains, the animals, and their fellows: there was Israel (240-750 years ago).19 Bednarik discovered a frag- not one straight line,23 no object formed by fat surfac- ment from a sizeable transparent rock crystal also in the es, no polyhedral shapes. Today, thanks to the pioneer- Acheulean, this time at Gudenushöhle, in Austria,20 and ing and quixotic Lewis Fry Richardson24 and the saga- quartz crystals at various levels ca. 276-500 kiloyears ago cious Benoit Mandelbrot,25 we know that the geometry in Wonderwerk Cave (South Africa).21 of nature is fractal geometry. Everything that nature has In summary, almost a million years ago (per- created on the face of the earth is the product of contin- haps even earlier if other discoveries are confrmed), uous branching and curvature. Everything except crys- the Homo erectus brain was so drawn to the shapes of tals. When Homo erectus tried to understand the world quartz crystals that they decided to collect and travel with their preconscious brains, the frst thing they had with them.22 Crystals collected by hominids have been to do was to fnd visual patterns, to separate what is the found alongside hominid fossils far from their place of same from what is diferent. When they found quartz origin. In several cases, the crystals found in these sites or pyrite crystals, they would have understood that (for instance in Singi Talav) came from diferent out- these shiny, polyhedral objects formed by straight lines, crops. Despite being collected in diferent sites, they fat faces and deterministic angles, free of curves, were were identifed as objects of the same type, a formidable utterly unique. We must remember that, with the excep- exercise of pattern recognition. Tese crystals were not tion of crystals, the straight line, the grid, polyhedra tools since they are too small to be used for any practi- and, of course, Euclidean geometry, were all invented by cal purpose. Tey have not been worked on or modifed, humans. Tis exceptional Euclidean nature is the reason 22 Juan Manuel Garcia-Ruiz

Figure 5. A 25 cm long dagger blade of rock crystal from the Cop- Figure 4. Te dolmen of Alberite. A) Te red star shows the loca- per Age found in tomb 10.042-10.049 of the Archaeological Zone tion of Alberite in South Spain. Te red circles show the possible of Valencina de la Concepción-Castilleja de Guzmán (Seville). Pho- provenance of the crystal. B) Te quartz crystal found in the dol- tography: Miguel Ángel Blanco de la Rubia. Courtesy of the ATLAS men. Photograph courtesy of Salvador Dominguez-Bella. C) Picture Research Group (University of Seville). of the dolmen by Pedro Cantalejo; d) Scheme of the dolmen, from the Gabinete de Bellas Artes of the Museo of Cádiz, Junta de Anda- lucía. tal is well faceted but it is not transparent, and it has some deformations. Tis suggests a sophisticated ability to identify single crystals from other rocks and miner- why rock (quartz) crystals were among the frst objects al aggregates. Te collection and location of this crys- collected by hominids, before the creation of our con- tal was purposeful because it is the only singular object sciousness. found in the dolmen. Te use and meaning of this object However, there was something even more enig- are unknown but indeed was not a tool because there matic about these objects. Everything that Homo erectus is no sign of hitting or percussion, no pieces have been saw around them had an origin, a history, a beginning removed from it, and it is too large to be used as a uten- and an end. Plants sprouted and grew, streams rose up sil. Most probably the crystal was considered an icon, a from the rains, as did the forms etched by erosion; ani- means of enormous value, of power. Te reason is that mals were born, and they themselves saw their children this type of pegmatitic quartz crystal does not exist near born: all things, even the rudimentary tools that they the area of the location of the Dolmen. Te closest areas had managed to produce, had an origin, had a begin- from which this type of crystals are known are the Gre- ning and an end. However, those mysterious crystals did dos Range near Madrid, or the Galicia massif northwest not. Who was the creator of something so singular? Tat of Spain.28,29 Terefore, these early mineral collectors question had to have an answer sooner or later. Inevita- had to transport that crystal from at least 500 kilome- bly, the crystals were seen as “machines”, devices that for ters or perhaps 900 kilometers to place it in the dolmen the frst time “communicated” with the beyond, what- of Alberite. Only an icon of great relevance would have ever or whoever the beyond was – any of the versions of been the subject of that efort. Almost coetaneous with the great beyond that the monolith from 2001 encom- 26 Alberite, quartz crystals were labored in many other passes. sites in Europe and elsewhere. Tey worked the crystals Te role of crystals as icons of power grew during of quartz with admirable skill and expertise, capable human evolution and prehistory. A dramatic example of converting, for example, a dagger into a work of art is the case of the Dolmen of Alberite (Cádiz, Spain), a 30 27 rather than a weapon (See Figure 5). funerary installation ca. 6000 years old (Figure 4). Te irresistibly combination of uniqueness, mystery, A smoky crystal of quartz of about 45 cm in length and harmony is the source of fascination for crystals was found within a well preserved dolmen. Te crys- that has been maintained throughout human history. tal has a blocky habit, and consists of a well developed Tose little “monoliths” not only sparked the imagina- {100} hexagonal prism terminated with {101} pyramids tion of our ancestors but have also shaped our culture faces. Its origin is undoubtedly pegmatitic. Te crys- and our thought. Tey become idols in which to trust, 2001: Te Crystal Monolith 23 the objects that triggered the belief in the existence of with time. Among the putative evidence for symbolic extraterrestrial powers, a power able to communicate by behavior, it has been proposed that Oldowan hominids sending storms, stones (meteorites) or energy to make were already shaping rocks with geometric shapes, ca. stones (lightning stones or fulgurites): the baethylus or 1.7 Ma.42 Was the Oldowan (mode I) culture already sacred stones that were worshipped for millenniums).31 prepared to collect crystals? Could the “crystal cen- Te use of crystals as objects with magic power was a tered” view I propose help to understand the technologi- characteristic shared by many primitive civilizations, cal jump from the Oldowan to the Acheulean?43 Oakley including Maoris, Apache, Canadian Indian, Polyne- were among the frst to propose that some of the fner sian and the Malagasy of Madagascar.32 Later on, crys- Acheulian handaxes look like masterpieces of artistic tals and minerals were considered arcane curative as crafsmanship rather than tools, i.e., they were made described in Babylonian tablets33 and the lapidarium of with symmetrical perfection beyond technical neces- the Early and Late Middle Ages and the Renaissance;34 sity.44 A mind that searched for perfect mirror symme- later, the harmony of the universe was explained by try is the frst step to produce harmony and beauty with Kepler on the basis on the Platonic polyhedra, shapes all patterns. Is this search for symmetry linked to the sym- of crystalline solids.35 From the mid-nineteenth century, metry of crystals? Is there any correlation between the crystals underlay the teaching that converted order and collection of crystals and the evolution of the mirror abstraction into the tools to understand the world, not symmetry of the handaxes during the Acheulean? only for science but also for art and philosophy.36,37,38 Te second line of research must focus on neuro- They are not extra-terrestrial. No alien civilization logical studies. It has been proposed that superior pat- put them there. Teir origin is as natural as any other tern processing is the fundamental basis of most, if not object of nature. But their rarity and their allure prob- all, unique features of the human brain, and the belief in ably sparked the imagination of a mind already prepared imaginary entities such as ghosts and gods. Te process to grasp the meaning of that singularity. Nowadays, our of pattern recognition involves the electrochemical, neu- brains are prepared to identify order; in fact, to see order ronal network-based, encoding, integration, and transfer even when order is not there, the origin of some optical to other individuals of perceived or mentally-fabricated illusions. Our brains have evolved to seek geometrical patterns.41 It will also be very important to understand patterns to understand the external world.39 But are our the current perception of crystal symmetry, and frac- brains designed to prefer order? In other words, do crystal order (either mathematical or random fractals) by tals attract us because the crystals were among the frst humans and closely related primates. Tis will be pre- elements that our ancestors, starting with Homo erectus, cious information for establishing whether Homo erec- collected? Or did we gather crystals almost a million tus perceived at the neurological level the fnding of the years ago because our brain was already designed to pre- “monoliths” one million years ago, and how they impact fer order (which benefted the comprehension of nature cognitive milestones in human evolution. Finally, the and therefore could be evolutionarily advantageous)?40 investigation opens a great question. Our understanding Did crystals impact our cultural history because they are of the world is based on a limited abstracted vision of a frmly linked to the birth of art, symbolism, and con- complex physical world. Te current level of understand- sciousness?41 ing has been shaped by the Euclidean reduction that we Te hypothesis that I propose in this paper is dif- began to use almost one million years ago. What would ficult to prove. Only circumstantial evidence can be have happened if there had not been any crystals? Would ofered so far, but the current evidence is strong enough an understanding of the world that had not drawn on to be worth of more thoroughly investigations. I foresee abstraction have been evolutionarily successful? Is there two main lines of research to explore further the role of a way to understand the world as it is, and not as we crystal geometry in the ability of hominids and humans have invented it for ourselves? to develop an abstract vision of the external world. First, we must increase archeological feld studies, particularly in the Paleolithic, to fnd further material ACKNOWLEDGMENTS to be studied with modern analytical methods. New discoveries of collected crystals and other manuports will Te research leading to these results has received help to reconstruct the story of the use of fuorite, pyrite funding from the European Research Council under and especially quartz crystals. Tis will allow revealing the European Union’s Seventh Framework Programme with the highest precision when hominids started to col- (FP7/2007-2013)/ERC grant agreement nº 340863 “Pro- lect crystals and how its signifcance and “uses” evolve metheus”. Te author also acknowledges a grant of the 24 Juan Manuel Garcia-Ruiz program Salvador de Madariaga (Ministry of Economy 23. Tere are some natural straight lines but they are and Competitiveness of Spain). irrelevant for the discussion. For instance, the hori- zon may appear like a curve line almost straight only in some places of the shoreline. Hanging vines REFERENCES are also straight, but they grow in the rain forest, an dangerous ecosystem for the evolution of hominids. 1. S. Kubrick. 2001: A space odyssey. Metro-Gold- 24. L.F. Richardson, O. M. Ashford, P. G. Drazin, Te wyn-Mayer. 1968. Collected Papers of Lewis Fry Richardson, Cambridge 2. A.C. Clarke, Te sentinel, Avon Periodical Inc, 1951. University Press, 2009 3. A.C. Clarke. 2001: A space odyssey. A novel based in 25. B. Mandelbroth, Te fractal geometry of Nature, W.H. the screeplay by A.C. Clarke and S. Kubrick, Arrow Freeman, 1983. Books, 1968. Page 18. 26. Aunque Kubrick and Clarke were atheists they decid- 4. S. Kubrick and A.C. Clarke, 2001, A Space Oddisey. ed to leave the interpreation of the Monolith opens. Typerwritter version of the script written October, See S. Schwam. Te Making of 2001: A space odyssey. 13, 1965. Modern Library. 1998 5. S. Kubrick and A.C. Clarke, 2001, A Space Oddisey. 27. J. Ramos Muñoz, and F. Giles Pacheco (Eds.) El Dol- Typerwritter version of the script written October, men de Alberite (Villamartín). Aportaciones a las 13, 1965, page a14. Formas Económicas y Sociales de las Comunidades 6. S. Kubrick and A.C. Clarke, 2001, A Space Oddisey. Neolíticas en el Noroeste de Cádiz, Universidad de Typerwritter version of the script written October, Cádiz, Cádiz. 1996 13, 1965, page a17. 28. S. Dominguez-Bella and D. Morata. Zephyrus 1995 7. A.C. Clarke. 2001: A space odyssey. A novel based in 48, 129. the screenplay by A.C. Clarke and S. Kubrick, Arrow 29. S. Domínguez-Bella, J. Ramos Muñoz, M. Pérez-Ro- Books, 1968. Page 19 dríguez.): “Productos arqueológicos exóticos en 8. A.C. Clarke. 2001: A space odyssey. A novel based in los contextos de los yacimientos prehistóricos de the screenplay by A.C. Clarke and S. Kubrick, Arrow la banda atlántica de Cádiz. Inferencias de su doc- Books, 1968. Page 20. umentación”, La Ocupación Prehistórica de la 9. A.C. Clarke, The sentinel, Avon Periodical INnc, Campiña Litoral y Banda Atlántica de Cádiz. Aprox- 1951, page XXX imación al Estudio de las Sociedades Cazadoras-Re- 10. G.D. Philips, Stanley Kubrick: interviews, University colectoras, Tribales-Comunitarias y Clasistas Inicia- Press of Mississippi, 2013. les, (Ramos Muñoz, J., editor), Junta de Andalucía, 11. W.B. Sparks, et al. Te Astrophysical Journal, 2016, Sevilla, 2008, 213. 829, 1. 30. A. Morgado, J.A. Lozano, L. García Sanjuan, M. 12. M.Y. Palmer et al. Science Advances 2018, 3, Luciañez Treviño, C. P. Odriozola, D. Lamarca Irri- e1700022. sarri, A. Fernandez Flores. Quaternary International, 13. E. Hand, Nature 2012, 490, 323. 2016, 424, 232. 14. J. De Witt, Nature Astronomy, 2018, 2, 214. 31. Plinio the Elder, Natural History, Volume 36. 15. R. G. Bednarik. Rock Art Research 2003, 20, 89. 32. V. B a r n o u w, Adam, An introduction to Anthropolog: 16. J. Harrod, Arts 2014, 3, 135. Ethnology. Vol. 2 (Homewood, Illinois,: Te Dorsey 17. P. We n z h on g , Acta Geologica Sinica (English Edition) Press, 1971). 228. 1931, 11, 109. 33. K. Reiter, Die Metalle im Alten Orient: Unter 18. F. d’Errico, C. Gaillard, M.N. Misra. Hominidae. Pro- besonderer Berücksichtigung altbabylonischer Quellen. ceedings of the 2nd International Congress of Human Munster:̈ Ugarit-Verlag. 1997. Paleontology. 1989 pp. 237–39. Editoriale Jaca Book, 34. J.L. Amorós. La gran aventura del Cristal. Editorial Milan. Complutense. Madrid. Second edition, 2017 . 19. N. Goren-Inbar et al., Rock Art Research 1991, 8, 133 35. J. Kepler, Te Harmonies of the World. Tr. Dr Juliet 20. R.G. Bednarik, Cambridge Archaeological Journal Field. pub., by American Philosophical Society. 1997. 1992, 2, 27. 36. B. Kahr. Crystal Growth and Design 2004, 4, 3. 21. R.G. Bednarik, Te Artefact 1993, 16, 61. 37. N. Brosterman. Inventing Kindergarten. New York: 22. R.G. Bednarik, Developments in Primatology: Progress Harry N. Abrams. 1997 and Prospects, Te Human Condition. Chapter 3: Te 38. J.S. Rubin, Intimate Triangle: Architecture of Crystals, Hard Evidence, Springer Science, 2011. Frank Lloyd Wright and the Froebel Kindergarten. 2001: Te Crystal Monolith 25

Huntsville, Alabama: Polycrystal Book Service. 2002 39. D. Ackerman I Sing the Body’s Pattern Recognition Machine. New York, NY: Time Magazine. June 15, 2004. 40. B.D. Beltman, Psychiatric Annals, 2005, 39, 5. 41. M.P. Mattson. Frontiers in Neurosciences, 2014, 8. 265. 42. J. Harrod, Arts 2014, 3, 135, page 140. 43. I. de la Torre, L. McHenry, J. Njau and M. Pante, Archaeology International 2011-2012, 15, 89. 44. K. Oakley. In S. L. Washburnand P. Dolhinow (eds), Perspectives on human evolution, 1972 pp. 14−50. Holt, Rinehart, and Winston. New York.

Firenze University Press www.fupress.com/substantia

Feature Article Almost a Discovery – Henri Gorceix, the Mining School of Ouro Preto, the Monazite Sand of Citation: J.H. Maar (2018) Almost a Discovery – Henri Gorceix, the Mining Bahia and the Chemistry of Didymium School of Ouro Preto, the Monazite Sand of Bahia and the Chemistry of Didymium. Substantia 2(2): 27-41. doi: 10.13128/substantia-59 Juergen Heinrich Maar Copyright: © 2018 J.H. Maar. This is Retired, Chemistry Department, Universidade Federal de Santa Catarina, Florianópolis, an open access, peer-reviewed article SC, Brazil published by Firenze University Press E-mail: [email protected] (http://www.fupress.com/substantia) and distribuited under the terms of the Creative Commons Attribution License, Abstract. Te chemical history of the supposed element didymium may well be char- which permits unrestricted use, distri- acterised as a case of collecting empirical data in a period of “normal” science. But this bution, and reproduction in any medi- element’s history also reveals little known facts of the history of chemistry in South um, provided the original author and America, such as the exploration and smuggling of monazite sands, and the difcult source are credited. beginnings of scientifc research and higher education in Brazil. Didymium is also a Data Availability Statement: All rel- curious case: even afer it was shown to be a mixture, it continued to be regarded as an evant data are within the paper and its element. Tis fact alone raises questions about the adequacy of scientifc methodology Supporting Information fles. at the time. In this paper, we consider the history of didymium, and determine how this history’s diferent facts and stories, set in Brazil’s rather unique historical and sci- Competing Interests: The Author(s) entifc context, intertwine thanks to the work of Claude Henri Gorceix. declare(s) no confict of interest. Keywords. History of didymium, Claude Henri Gorceix, Ouro Preto Mining School, monazite sands.

It is absolutely necessary to study the facts, to observe phenomena. Henri Gorceix

Whatever the aim Man establishes for himself to reach, whatever the idea he chooses to develop, they cause a great feeling of pleasure when he succeeds, a great happiness when turned reality.1 Henri Gorceix

TO DISCOVER AN ELEMENT

For the layman, more than for the scientist, the discovery of a new element marks a great event in the history of chemistry. Indeed, discoveries or isolations of elements signify a great scientifc advance, even from a theoretical point of view: the discovery of oxygen, of the frst noble gas, of the elements foreseen by Mendeleev in his Periodic System. Te discovery of an

Substantia. An International Journal of the History of Chemistry 2(2): 27-41, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-59 28 Juergen Heinrich Maar element is also one of the most “democratic” aspects prevented the discoverer from carefully isolating the new of the history of chemistry. While some elements were element from mixtures of elements or already known discovered by the most respected scientifc authorities – elements. Scheele, Berzelius, Klaproth, Vauquelin, Wollaston, Davy Te discoveries of only three elements have been – others were found by diligent and industrious prac- associated with scientifc practice in Latin America: plat- titioners of a “normal” science (in Kuhnian terminol- inum – discovered in gold deposits in Colombia by the ogy) – Mosander or Marignac. Some of these elements Spaniard Antonio de Ulloa (1716-1795); vanadium – dis- perpetuate the names of their discoverers: Gregor and covered as “eritrônio” in a lead mineral from Zimapán titanium (1791), Courtois and iodine (1811), Balard and (Mexico) by the Spanish mineralogist Andrés Manuel bromine (1826).2 del Rio (1764-1849), professor at the Real Seminario de Te discovery, isolation and physical and chemical Minería in Mexico; and tungsten, whose Spanish dis- characterisations of rare earth elements constitute an coverers Fausto de Elhuyar (1755-1833) and Juan José de apparently unbounded harvest for historians of chemis- Elhuyar (1754-1796) were later key personalities in the try. In this endeavor, the marvels of science and the aes- scientifc communities of Mexico and Colombia, respec- thetic pleasure of discovery are painted in the best pos- tively. sible light. Isolation and perfect characterization are nec- In this paper, we will shed light on the little-known essary requirements for defning a “discovery” of a new aspects of two stories that are intertwined by the works element. In the case of the elements of the rare earths, of an important fgure within the history of chemistry all were derived from two pioneering discoveries: Gad- in South America: the French mineralogist and chem- olin’s yttria (1794) and Berzelius’ and Klaproth’s ceria ist Henri Gorceix (1842-1919). Te frst of these stories (1803). Experimental problems were, however, particu- is the foundation of the Mining School of Ouro Preto, larly difcult: their physical and chemical properties are where Gorceix acted as director for several years. Te so similar that chemical separations proved extremely second is the curious case of the didymium (Mosander, burdensome and laborious (thousands of recrystalliza- 1841), a supposed rare earth element that continued to tions are ofen required). Such difculties led chemical be the subject of much research – especially at the chem- annals to occasionally record the same discovery twice ical laboratory of the Mining School of Ouro Preto – or announce discoveries of nonexistent ‘elements’ that even afer it had been shown to be a mixture of neodym- were, in reality, mixtures or already known elements. ium and praseodymium, and not an element afer all.5 Many such situations are discussed in Karpenko`s paper on `spurious elements`.3 For most of the 19th century, chemistry lacked a THE MINING SCHOOL OF OURO PRETO theoretical framework that could guide a targeted or systematic search for new elements. Experimental com- In contrast to Hispanic America, where the frst uni- plexity was later reduced by the introduction of a new versities were founded in the 16th century, Portuguese analytical tool – spectroscopy (Bunsen, Kirchhof, 1859). America was only granted access to higher education But the spectra of alleged newly discovered elements can beginning in the 19th century – unless we consider the easily be mistaken for either combinations of the spectra Hispanic universities as mere colleges and some of the of already known elements or an impure element. For Luso American institutions, such as the Jesuit College in this reason, many ‘discoveries’ of new elements in the Salvador (1557), the Seminario in Olinda (1800)6, and the realm of rare earths were made, much to the perplex- military School of Fortifcations in Rio de Janeiro (1792) ity of the scientifc community. In 1880, the anonymous as institutions of higher learning.7 editor of the Manufacturer and Builder observed that the Te transfer of the Portuguese royal family from the chemists of recent decades had discovered an enormous capital Lisbon to Rio de Janeiro (1808) brought freedom number of new ‘elements’. But, even if one only consid- and favoured the growth of Brazil’s practical and applied ers the last few years, the number of ‘discoveries’ and sciences, which evolved in line with the pragmatism accumulation of unconfrmed empirical facts was sim- established by the Marquis de Pombal (1699-1782) at the ply beyond belief.4 Te editor correctly identifed why: University of Coimbra. But it was only Brazil’s politi- chemists considered evidence from spectroscopic analy- cal Independence (1822) which enabled the foundation ses sufcient to warrant the report of a new discovery, of the frst faculties stricto sensu: the law schools in São ignoring that such evidence could well be a mixture of Paulo and in Olinda (1827), the medical schools in Rio rare earth elements, or a re-discovery. Te desire and de Janeiro and Bahia (1832), pharmacy courses in Rio de haste to claim priority over a new discovery sometimes Janeiro (1832), Bahia (1832) and Ouro Preto (1835), and Almost a Discovery 29

a region rich in ores and geological interest, Ouro Preto proved an obvious choice for a mining school. Potosi, in Bolivia, centralizing immense silver ore extraction, host- ed since 1756 the frst Mining School of the Americas, perhaps the frst worldwide.10 Afer 1699, when the adventurer Antonio Dias discovered the frst gold deposits around what is now Ouro Preto, the original village grew very quickly. It became a municipality in 1711 and was suggestively named Vila Rica (Rich Village); in 1720 the village became the capital of the new captaincy of Minas Gerais. During the 18th century, Vila Rica was responsible for most of the Americas’ gold production, which is refected in the rich religious and profane architecture preserved to our days Figure 1. Emeric Marcier (1916-1990), “Ouro Preto”, 1952. Oil on (included by UNESCO in the World Heritage in 1980), canvas, 64 x 90 cm. Museu de Arte de Santa Catarina/MASC, Flori- as well as a rich literary and artistic activity. A timid anópolis, Brazil. refex of the Enlightenment can also be detected in a movement for political emancipation in 1789 in Ouro Preto, which was the capital city of the Province and the polytechnical school in Rio de Janeiro (1874), which State of Minas Gerais until 1897. was separated as institution from the Escola Central Te Mining School of Ouro Preto was fnally found- (1848), successor of the military school and a cradle of ed in 1875, during Alfredo Correia de Oliveira’s (1835- positivist thinking in Brazil. 1919) tenure as minister of the empire. A personal deci- These institutions created an intellectual elite in sion, almost an imposition, of Emperor Pedro II, in the Brazil and provided higher level education – but only words of the historian José Murilo de Carvalho, “the cre- in a practical and pragmatic sense. Proper scientific ation of the School was, afer all, an act of political will, research in Brazil was the odd exception, not the rule, directed in great part by ideological rather than economic and Brazilian universities were created as predominant- reasons”11. In a nation with a slave-based society and an ly anachronic institutions.8 In this context, the Mining economy based on agriculture and export with “a very School of Ouro Preto occupies a position entirely sui incipient industrial activity”, there was no real need to generis. Since its conception and foundation, the school’s train geologists or mining engineers. Te country had day-to-day life involved not only teaching, but also much no tradition of geological or mineralogical research. It is international-level research in its frst decades of exist- therefore difcult to imagine the creation of the Mining ence. School as a requirement mandated by society, the lower According to S. Figueirôa, the frst proposal for the classes, or even as a necessary step for the country’s eco- creation of a mining school in Brazil was made in 1804 nomic, social and scientifc development. Pedro II (1825- by Manuel Ferreira da Câmara (1764-1835), who took his 1891), constitutional monarch at the tender age of 15, Alma Mater, the Mining School of Freiberg as a model.9 was educated by his tutors, José Bonifácio de Andrada e Te frst attempt to create a mining school in Ouro Pre- Silva (1763-1838), the “patriarch of independence”, and to, Minas Gerais, dates back to 1832, during the period the Marquis de Itanhaém, Manuel Inácio de Andrade of the Regência – which governed Brazil from the abdi- (1782-1867), mentors responsible for awakening, besides cation of Pedro I (1831) until the majority of Pedro II his ever mentioned interests in literature and arts, a (1840). But why Ouro Preto? strong interest in science, particularly chemistry.12 Tere Ouro Preto was chosen to host a new mining school are well preserved notes from Pedro II on Mendeleev’s thanks in great part to the eforts of Bernardo Pereira de periodic system (the evolution of which he documented Vasconcelos (1795-1850), who argued before the Brazil- through scientifc journals), and on scientifc subjects ian Parliament for a law able to facilitate the economic he personally taught his daughters. During his trips to recovery of Minas Gerais, which had been stagnant since Europe, the emperor visited many chemists: Chevreul, 1780. Tis law was also meant to compensate for the fact Liebig, Berthelot, Pasteur, Kelvin, van’t Hof. Te emper- that the province had received none of the schools creat- or was also a member of the Royal Society and of the ed afer Brazilian independence. But the law would only Academies of Paris, Berlin, St. Petersburg and Munich.13 be made a reality 43 years later. Located at the center of We shall not occupy ourselves with well-known contro- 30 Juergen Heinrich Maar

CLAUDE HENRI GORCEIX (1842-1919)

Several foreign scientists exerted a long-lasting infuence on the evolution of chemistry in Brazil. We may begin by remembering the Italian general Carlos Antônio Napion (1757-1814), professor of chemistry at the Military Academy, or the Frenchman Félix d’Arcet (1814-1847), who attempted to produce sulfuric acid in Rio de Janeiro but died in a laboratory fre. Tose who had the most enduring infuence on Brazilian chemistry were the German phytochemist Teodor Peckoldt (1822- 1912), pharmacist of the emperor, and analyst Henri Gorceix, a pioneer of geochemistry in Brazil. Claude Henri Gorceix was born on October 19th, 1842, in the small village of Saint-Denis-des-Murs, district of St. Léonard, Department of Haute-Vienne (which is also the native region of Gay-Lussac, a distant relative of Gorceix) and fathered by Antoine Gorceix and Cécile-Valérie Beaure la Mareille.15 Gorceix concluded his studies at the Lycées in Limoges and Douai, graduated in 1866 at the École Normale, and obtained a diplo- ma of a “Generalist in Physical Sciences and Mathemat- ics”. Gorceix lectured at the French School in Athens while exploring the geology and mineralogy of Greece. Figure 2. Claude Henri Gorceix (1842-1919). Lythography by Daubrée found him in Athens afer Pedro II’s invitation, unknown artist. (courtesy Oesper Collection for the History of Chem- and in March 1874, Gorceix signed a contract with the istry, University of Cincinnati). Brazilian government at the Brazilian embassy in Paris. Te contract mandated he “to organize teaching of mineralogy and geology in Brazil”. Article 12 of the contract versies surrounding his lack of commitment to matters states, ipsis litteris: “Mr. Gorceix promises to go to Rio of State in favour of his pursuit of personal interests14 de Janeiro, at the services of the Imperial Government to – matters possibly made worse by the longevity of his organize the teaching of Mineralogy and Geology”. Reso- reign. We will, however, focus on his interest in science. lute, severe and ill-tempered, rough and rude accord- On his second voyage to Europe (1870/1871), Pedro ing to contemporaries, and arrogant in the opinion of S. II visited the Paris Academy (he would be elected a Figueirôa16, Gorceix was a man tailored to his new func- member in 1875), and upon his return, he invited Gabri- tions: teaching and research in mineralogy and geology. el Auguste Daubrée (1814-1896), a former student of Multiple administrative and bureaucratic difculties also the École Polytechnique and then director of the Min- awaited him – not to mention envy, enmity, unfair spon- ing School in Paris, to visit Brazil – not only to research sorship and competition, lack of recognition and under- our mineral kingdom, but to help with the foundation standing. of a mining school. Daubrée had no interest in leaving While still in Europe, Gorceix organized equipment Europe, and suggested Claude Henri Gorceix for this and materials for teaching and laboratory activities. task. As we will see, Gorceix was the perfect man, in all Afer the School was installed, he had to hire and engage respects, for this work. A sincere friendship linked Gor- teachers, assistants and other staf members, as well as ceix and the emperor, as demonstrated by the surviving, ascertain fnancial assistance for poor students arriving prolifc correspondence between them. Te emperor and from other provinces. About the installation of the labo- Empress Teresa Cristina (1822-1889) became godparents ratory, Dutra comments: to Gorceix’s daughter, Cécile Pierrete Terèse Gorceix (Professor Christiano Barbosa da Silva would later say For us today it is practically impossible to evaluate the that Gorceix’s “frst daughter was the School of Mines”). immense difculties associated with the installation and maintenance of a chemical laboratory in the tropics at the end of the 19th century. How was it possible to fnd in Europe and take to Ouro Preto the necessary chemicals, Almost a Discovery 31

glassware, stoves, distillation apparatus, equipment for preparing samples, and all kinds of paraphernalia? How to hire skilled labour necessary for scientifc and technical activities? [ … ] . We can afrm with certainty that during many and many years the laboratory supplied the incipient demand for analyses, in a country just beginning to know its immense mineral possibilities, and also that the laboratory created a basis for future development of geochemistry in our country.17

With respect to equipment, the laboratory owned a small spectroscope (a detail important for our discussion) that may have been considered a novelty in laboratories and was certainly a rarity in Latin America. In July 1874, Gorceix arrived in Rio de Janeiro to a country that lacked any tradition in mineralogy and geology. Once there, he was tasked with organizing teaching and research ex nihilo. Brazil’s supposed and ofen hailed mineral richness is in great part a myth.18 Geologist Daniel Atêncio of São Paulo University notes that of more than 5000 mineral species known today, only 65 are native to Brazil (23 of them discovered by Atêncio himself). The first to be discovered was chrysoberyl (BeAl2O4), described in 1789 by Christian August S. Hofmann (1760-1814) and Dietrich Ludwig Karsten (1768-1810), both students of Abraham Wer- ner in Freiberg. Chrysoberyl was analyzed by Martin H. Klaproth (1795, before characterization of beryllium Figure 3. Bust of Henri Gorceix by an unknown sculptor in the as an element). José Bonifácio de Andrada e Silva (1763- inner yard of the former Escola de Minas, erected 1926 at the 50th 1838), also a former student at Freiberg and discoverer of anniversary of the Mining School, copyright and photograph by several new mineral species in Scandinavia (1799/1801), René & Peter van der Krogt, Delf, Netherlands. (courtesy René & could have contributed to the creation of a Brazilian Peter van der Krogt, Delf). mineralogical tradition, but he instead devoted himself exclusively to the cause of Brazilian political independence. demic and scientifc activities. So, instead, he decided Gorceix’s frst assignment in Brazil was a visit to to fnish his research career and engage in politics. He the Province of Rio Grande Sul with the botanist Ladi- became mayor of Le Mont par Bujaleuf. Gorceix only slau Neto (1838-1894), director of the National Muse- returned once to Brazil, in 1896, when he was invited as um. Near the end of the same year, Gorceix travelled to a teaching consultant. He died in Limoges at September Minas Gerais to determine the location of the Mining 6th, 1919. In 1926, on the 50th anniversary of the School, School. He opted for Ouro Preto. In the midst of 1875, a bronze bust was erected in the internal yard of the he presented his proposals for the School, which includ- building. In 1970, his remains were transferred to the ed his “teaching philosophy”, to the governor. Finally, on Mausoléu Gorceix in Ouro Preto. October 12th, 1876, the Mining School of Ouro Preto was Besides his continuous eforts to ensure the sur- formally established. Gorceix was named the director vival and quality of the School, several examples of his (until 1891) and was responsible for various disciplines. own scientifc research were published during his time With the fall of the monarchy and the proclamation in Ouro Preto, including analyses of minerals and the of a Republic in 1889, Gorceix and the school lost their discovery of xenotime-Y (YPO4). Tese research papers great protector, Pedro II. In the face of this new politi- were published in part in the journal he founded in cal context, Gorceix departed as director of the school in 1881, Anais da Escola de Minas (since 1936 Revista da 1891 and named Professor Archias Medrado (1851-1906) Escola de Minas, and since 2016 International Journal as interim director. Gorceix returned to France, but his of Engineering), and also in part in the Comptes Rendus long absence made it difcult for him to return to aca- of the French Academy and in the Bulletin de la Société 32 Juergen Heinrich Maar

Minéralogique de France. Gorceix was responsible for defeated the arguments of infuential people, including the section “Geology in Brazil” at universal expositions the Viscount of Rio Branco, José Maria da Silva Para- such as the Exposition Universelle de Paris (1889) and the nhos (1819-1880), who was a former Prime-Minister South American Exposition in Berlin (1886).19 (1871/1875) and very close to the Polytechnic, where he served as dean and professor. Gorceix’s plan remained in efect, with minor adjustments, until 1893 – substan- THE MINING SCHOOL OF OURO PRETO. CUM tial revisions occurred only in 1936, when the School MENTE ET MALLEO became part of the Universidade do Brasil. Since 1969, the School has been part of the Federal University of Once the location of Ouro Preto had been selected, Ouro Preto. the Mining School was installed in the old Governor’s It proved difcult to attract teachers and students to Palace, a partially fortified building erected in 1741 Ouro Preto. Although a provincial capital, the town was under the orders of Governor Gomes Freire de Andrade still relatively small (circa 12.500 inhabitants in 1872). (1685-1763). Te building shares a location with the for- Ouro Preto was also far from Rio de Janeiro, and the mer Casa de Fundição, designed by Portuguese architect school’s admission examinations and coursework itself Manuel Francisco Lisboa (17…-1767) on the basis of a were rigorous (at least compared to the country’s poor project by general and military engineer José Fernandes secondary schooling). Graduates also faced poor pros- Pinto Alpoim (1700-1765), an important fgure in the pects when searching for work in Brazil.24 development of Brazilian mathematics.20 For the disciplines of physics, chemistry, mathemat- In Ouro Preto, Gorceix had to choose between two ics, geology, mineralogy, foreign teachers were hired: models: the Paris Mining School and the Saint Etienne Armand de Paul Bovet, Arthur Charles Tiré (1853- Mining School. He was also, of course, infuenced by his 1924), Paul Ferrand (1855-1895); Brazilian instructors own course at the École Normale.21 Practical reasons led included Archias Medrado (1851-1906), Leônidas Botel- him to choose the St. Etienne model, which proved easi- ho Damásio, and Francisco van Erven (1851-1936).25 S. er to adapt to local conditions. Gorceix’s project difered Figueirôa highlights that Gorceix’s letters expose his substantially from all other higher education courses in contempt for Brazilian teachers, who in his judgement Imperial Brazil. Murilo de Carvalho lists the proposed were bad teachers teaching bad students. facets of Gorceix’s institution: free-of-charge education, Brazil’s general political and intellectual context full time classes (including Saturdays and Sundays) for must be considered when discussing the importance students and teachers, a ten-month long school term of the Mining School. Unlike what occurred at the (seven months was the norm in Brazil) followed by two Polytechnical and Medical schools in Rio de Janeiro, months of feldwork, valuation of creativity and labora- Positivism had no infuence in Ouro Preto. Murilo de tory work, entrance examinations and frequent tests Carvalho describes Gorceix as a Catholic, and his fel- throughout the year, limited enrollments, grants for lows as materialists or evolutionists.26 Provincial and qualifed students (for further studies in Europe), and isolated, Ouro Preto was the right place for study and fnancial assistance to students in need.22 research: students and teachers interacted more than in In Gorceix’s own words: other schools, and both remained longer at the school, the principal point of social life in the city. Student life Time for frivolous discussions about concepts and theories, was much diferent from other schools, and interest in simple speculations of the mind, a legacy from the Middle the nation’s economic and political future led to strong Ages and abandoned by the Old World since a long time, nationalist thought. Many former students ascended to are over [ … ] mines and metallurgical plants will be the best books in our libraries.23 high posts in science, technology and government. Carvalho observes that “it was essential for the ‘Gor- Gorceix’s project was not only innovative, but also ceix spirit’ the concern in translating scientifc knowledge almost an afront to Brazilian academic traditions. Te in developmental policies”. Similarly, in 1970, Djalma project was submitted to the appreciation of a com- Guimarães (1894-1973), pioneer of Geochemistry and mission of the Rio de Janeiro Polytechnic School, and former Ouro Preto student, said: although approval was granted in the end, many sharp Probably former students of the School of Mines exercised criticisms arose, marking the beginning of a famous a great infuence during the frst two decades of this cen- rivalry between the two schools. tury, when political context was not yet prepared to discuss Te approval of his plans confrms Gorceix’s politi- issues related to scientifc and technological knowledge. cal strength during the imperial period. Gorceix Calógeras,27 Pires do Rio,28 Francisco de Sá,29 and other Almost a Discovery 33

eminent former students of the Ouro Preto Mining School Te remnants of the wooden pier built on Cumu- entered the political scenario armed with objective knowl- ruxatiba beach can still be seen today. Te ruin advances 30 edge of our natural resources [ … ]. hundreds of meters into the sea and facilitated the transfer of the “ballast of Brazilian ships” into ship holds. However, the State government assigned the plan- Inhabitants and tourists in Cumuruxatiba, today a sum- ning and supervision of construction of the new capi- mer resort, remain predominantly unaware of the his- tal Belo Horizonte (1897) to engineer Aarão Reis (1853- tory and purpose of this derelict construction project, 1936), a Rio de Janeiro Polytechnic graduate. Orville which has withstood the forces of the ocean, the mer- Adalbert Derby (1851-1915), an American geologist cilessness of time and decay, and even vandalism. Afer active in Brazil, highlighted the quality of research done the pier was intentionally set on fre in 2005, only its row in Ouro Preto and its international recognition in an of pillars survives. 1883 paper published in Science: In 1890, the government of Bahia forbade “export” of the sands. But the trade was liberated again in 1895, At present, the national museum and observatory in Rio, though this time it was at least nominally under the and the school of mines in Ouro Preto, are the principal control of state authorities. In 1900, a record-breaking centres of scientifc activity. Te latter, being a compara- tively new establishment, remote from the centralizing 7120 tons were exported. Aferward, the Brazilian gov- tendencies of the capital, organized on European models, ernment fnally forbade any export of this valuable raw and controlled by an able corps of French specialists, has material. International interest in the illegal extrac- escaped many of the vices of the older institutions.31 tion and smuggling of rare earth can be seen in Alfred Hitchcock’s 1946’s production “Notorious” (launched in Brazil as “Interlúdio”).34

MONAZITE SANDS FROM BAHIA DIDYMIUM AND THE CHEMICAL ANALYSES OF On the southern coast of the Province/State of GORCEIX Bahia, in the municipalities of Caravelas and Prado, dark and heavy sand extends in long strips. Known Te supposed element didymium was frst espoused today as monazite sand, these strips are rich in rare to exist in 1841, when Carl Gustaf Mosander (1797-1858) earth minerals, and, in the case of sands from Bahia, isolated it as an impurity of lanthanum, an element he thorium minerals. Ludwig Camillo Haitinger (1860- had discovered as a contamination of cerium in 1839. 1945) and K. Peters discovered the presence of radium in Didymium remained an element (symbol Di) until 1879, these sands in 1904.32 Later, other deposits of monazite when Paul Émile Lecoq de Boisbaudran (1838-1912) dis- sands were discovered in Espírito Santo (1896) and Rio covered a fraction of samarium in didymium, and as de Janeiro. Orville Derby, director of the geology section such didymium was included in periodic tables (Gme- of the National Museum in Rio de Janeiro, sent samples of sand from the beach of Cumuruxatiba (municipality of Prado) to Gorceix in Ouro Preto (1883). Tis heavy brown sand he received from John Gordon, an English- man (or American?), a manager of the North-American cofee exporting company E. Johnstone & Co. Gorceix analyzed the samples from 1883 to 1885 and realized they contained phosphates and oxides of cerium, lanthanum and didymium (then still believed to be an element). At the same time, Gordon, in partnership with English and German merchants, had exported 3000 tons of monazite sand by 1888. By 1890, the endeavor had “exported” a total of 15.000 tons to Hamburg and was supplying rare earth processing industries in Vienna and Berlin.33 Although there is no confrming evidence, it is believed Gordon personally negotiated with Carl Auer von Welsbach (1858-1929) and collected a fortune with Figure 4. Didymium was incorporated in many periodic classifca- this not entirely legal “trade”. tions, like this draf by Hugo Schif (1834-1915) (courtesy Museo di Storia della Scienza, Florence, reproduced with permission). 34 Juergen Heinrich Maar lin, 1843, Newlands, 1865, Kremers, 1869, Mendeleev, ium (1842-1893)”, which was published in 1903 by the 1869). Didymium possessed elemental status from 1841 Smithsonian Institution afer the initiative of Henry Car- to 1885, when Carl Auer von Welsbach (1858-1929) split rington Bolton (1843-1903). We may suggest a thought the mixture into two elements: neodymium and praseo- provoking question: to what extent is chemical practice dymium. Didymium’s story is not a simple one, and oth- an exact science, considering the many publications er researchers, like Georges Urbain and Henri Gorceix, on the extraction, isolation, purifcation, chemical and came to the same conclusion; in his work on the subject, physical properties of compounds of an element which Gorceix made no claim for a priority, and his efort was does not exist? Should we not consider the scientifc virtually ignored in the “central” scientifc scenario as methodology in place to be questionable or inappropri- well as in the country where he lived and worked. To ate? Has the data been intersubjectively verifed? Should the best of our knowledge, Dutra (2002) was the frst to we accept the sometimes incoherent and inconsistent draw attention to Gorceix’s ‘decomposition’ of didymi- data, and the buildup of information, which would later um. We brought this information to the attention of col- be rejected, as a ‘normal’ step for any scientifc investi- leagues abroad, who mentioned it in recent publications. gation? Or should we explain away such anomalies by Didymium, as well as six ‘authentic’ elements (lantha- appealing to “anthropogenic factors” centered on the num, neodymium, praseodymium, gadolinium, samar- shortcomings of our instruments and research tech- ium, europium) were obtained from cerium, discovered niques? in 1803 by Berzelius, and independently by Klaproth in In fact, until 1885, didymium was widely accepted to bastnaesite, a mineral found in the mines of Bastnaes in be a real element. However, many experiments showed Sweden (Bastnaesite was described for the frst time by contradictory results, which sometimes difered in sam- Vilhelm Hisinger [1766-1852]). ples with diferent origins, were sometimes impossible to replicate, or ofered inconclusive results (such problems are not exclusive to didymium). Since the discovery of didymium, there have been doubts about its elementary nature – this is clear in publications by Hans Rudolph Hermann (1805-1875) as early as 1845, O. Popp in 1864 (erbium and terbium as mixtures of didymium and yttrium), Per Teodor Cleve (1840-1905) in 1885 afer a series of experiments done since 1874, Marc Delafon- taine (1838-1911) in 1878, and Bohuslav Brauner (1855- 1935) in 1885. When, in 1879, Lecoq de Boisbaudran discovered samarium as an impurity in didymium, these doubts seemed to be clarifed in part. Among the chemists that systematically completed research on didymium, we cite Jean Charles Galissard de Marignac (1817-1894), H. R. Hermann, F. Frerichs, Per T. Cleve, Karl Friedrich A. Rammelsberg (1813-1899), and many others. None of these chemists used unorthodox research strategies. No less than nine methods were proposed (and published) to separate lanthanum from didymium, by Hermann, Robert Bunsen (1811-1899) and Zschiesche, Augustin With reference to the great amount of data on the Damour (1808-1902) and Deville, F. Frerichs, Auer von chemistry of didymium, Arthur Comings Langmuir Welsbach, Auguste Victor Verneuil (1856-1913) and (1872-1941) wrote in 1903: “Te voluminous literature Grigory Wyroubof (1843-1913), P. Mengel, Witt, Paul of didymium afords a striking illustration of the pursuit Gerard Drossbach (1866-1903). At the same time, Wil- of science for its own sake, and with no reward beyond liam Crookes (1832-1919), Octave Leopold Boudouard the satisfaction of having advanced the cause of truth”.35 (1872-1923), Eugène Demarçay (1852-1903), Georges Moreover, adding to this “science for science”, to para- Urbain (1872-1938) and G. Dimmer expressed views on phrase parnassiens’ “l’art pour l’art”, the enormous the probable decomposition of didymium, retaining this quantity of publications on didymium can now be found opinion even afer failing to prove it experimentally (not in A. C. Langmuir’s “Index to the Literature of Didym- confrming Brauner’s publication from 1885).36 Adding Almost a Discovery 35 to the uncertainties about rare earths, Boudouard, from brand and Norton were looking for a very pure sample the Conservatoire National des Arts et Métiers, suggested of didymium, like ‘pure didymium chloride’, which is that even Mosander’s cerium from 1839 could be a mix- necessary for electrolysis, and supposedly used the entire ture of two elements (1895).37 supply of lanthanum sulfate and didymium sulfate avail- Our purpose here is to provide detail on Gorceix’s able at the laboratories of Heidelberg University (Hille- chemical experiments with monazite sand and the brand was a student and later an assistant of Bunsen; didymium it supposedly contained. Tese experiments Norton was also a graduate student of Bunsen). In the were performed in Ouro Preto from 1883 to 1886, a electrochemical series, didymium is located between place distant from “central” scientifc research institu- cerium and magnesium. tions. Although the fndings were presented at the Par- Numerous contemporary papers discuss the separa- is Academy and published in French journals known tion of didymium from lanthanum and cerium (Marig- worldwide, they did not attract the deserved attention, nac, 1849; Bunsen, 1875) or independently from cerium perhaps because Gorceix lived and worked at the fringe (Popp, 1864) or lanthanum (Frerichs, 1874), or from still of international academic life and at the “periphery” other species, such as gallium (Lecoq de Boisbaudran, of the academic world. It is interesting to observe that, 1882), thorium (Hermann, 1864), and zirconium (Her- among the hundreds of scientifc communications and mann, 1864). Te separation of ‘didymium’ from cerium papers published by Langmuir in 1903, there is not a or lanthanum is not an absurdity; it could be a separa- single reference to Gorceix. Te so-called “ofcial” sci- tion of a ‘mixture (neodymium + praseodymium)’ from ence solemnly ignored scientifc productions from out- its neighbours. Didymium’s place in the Periodic Table side its geographic limits. Te exhaustive two volume was also a contentious matter (Schiff, 1879; Piccini, texts of Richard Böhm, “Die Darstellung der Seltenen 1885). Te paradox of an experimental investigation of Erden” (1905) make no reference to Gorceix’s work on a non-existent element will be discussed later. Te fgure monazite sand.38 shows a drawing of the Periodic System by Ugo Schif Before coming to Gorceix, it is interesting to see past (1834-1915), which included didymium as an element.40 research on the rare earth “didymia” since its “discov- By way of example, let us discuss the procedure ery” by Mosander in 1841. Marignac (1848, 1854), Her- developed by F. Frerichs for separating lanthanum from mann (1853), Cleve (1875, 1883), Delafontaine (1878), didymium (1874).41 A mixture of lanthanum and didym- Kopp (1879), and Clarke (1881) determined the “atomic ium oxides is heated in a fow of chlorine, then water is weight” of didymium. Mendeleev admitted the value added to the resultant mixture of the oxichlorides, and 138 in his Table. Rammelsberg (1861) found the isomor- the solution is allowed to stand for a while – La and Di phisms of didymium sulfate and other sulfates (“the proportions in the solution are 3:6. Lanthanum chloride isomorphism of the three cerite metals is beyond ques- remains dissolved, and didymium chloride precipitates. tion”), Marignac (1856) determined the theoretical crys- With higher concentrations of lanthanum, the proce- talline form of didymium sulfate, and Nordenskiöld dure must be repeated. Other methods suggested by (1861) determined the crystalline structure of didymium Frerichs consist of dissolving the oxides in nitric acid, oxide, DiO (Rammelsberg proposed DiO2). adding sulfuric acid, and allowing the solution to stand American chemists Francis William Hillebrand for several days. Te sulfuric acid combines with lan- (1853-1925) and Tomas Norton (1851-1941) believed to thanum, creating lanthanum sulfate. In order to obtain have isolated metallic didymium in 1875 via electrolytic pure didymium compounds, sulfuric acid is added to reduction of DiCl2, thus obtaining cerium und lantha- the oxides until all the lanthanum and some of the num in accordance with a method developed by Bunsen didymium is converted in sulfates. Afer evaporation shortly afer the development of the Bunsen cell.39 Using and ignition, a white mass is obtained, from which water electrolysis, Bunsen obtained yttrium, cerium, lan- extracts the lanthanum and part of the didymium. Pure thanum, didymium, thorium, zirconium, calcium and didymium oxide is obtained by dissolving the residue strontium in a “free state”. Bunsen’s process converts in sulfuric acid. Tis example illustrates how the lack of oxides into sulfates, sulfates in chlorides (via oxalates), appropriate experimental (in this case, analytical) meth- which were fnally submitted to electrolysis. Te authors ods may result in false conclusions in chemical practices. remark that, of the three metals, didymium is the most In the case of Gorceix, our interests lie in one aspect difcult to obtain: didymium reacts with oxygen from of didymium’s chemistry: its occurrence in monazite the air to regenerate as an oxide. Tey further mention and monazite sand. Gorceix had already completed some that the physical properties of didymium are more simi- research on monazite before receiving the samples of lar to those of lanthanum, rather than cerium. Hille- Caravelas from Orville Derby in 1883.42 Te frst men- 36 Juergen Heinrich Maar tions of monazite in Brazil came from Gorceix himself visibility among scientists. However, its relative obscu- (1883), when he “tentatively” considered the yellow sand rity may be due to its provenance from the distant and grains from Fazenda Quebra-Galho (São Paulo)to be unfamiliar Ouro Preto. monazite. He was later informed that the sand was actually from Caravelas (1884). Gorceix mentions the occurrence of monazite in other places in Minas Gerais: the diamond-containing depositions in Diamantina43 (1884, 188544), gold places in Casca on the Rio Doce (1885),45 and fnally in Salobro, Bahia (1884).46 In possession of the sands remitted by Derby, Gorceix worked on separations. In 1885, in the Bulletin de la Société Minéral- ogique de France, he writes the following: “the samples [from Caravelas sands] are found in form of yellow bright grains, mixed with some ferro-titanium”.47 Once the iron is completely removed, microscopic examination of the sands reveals a homogeneous aspect containing crystals. Some of these crystals resemble monazite and others suggest the presence of another species of mineral. Te density of the mixture is 5,1. Te sand is then ground into a fne powder, heated with sulfuric acid, dried and dissolved again in a weak acid solution. Te insoluble fraction in the resultant sulfuric acid contains silica and In 1913, Richard Böhm published a more detailed zircona, and the soluble fraction contains cerium and analysis of monazite sand from Bahia, showing the ele- didymium oxides, which are precipitated with oxalic ments neodymium and praseodymium to be decomposi- acid. Oxalates are heated and converted into nitrates. tion products of didymium49: Fusion with potassium nitrate at 360°C separates cerium from didymium. Gorceix’s analysis suggests the follow- Cerium dioxide 31.5 % ing composition for the monazite sand from Caravelas: Phosphoric acid 26.0 Lanthanum oxide 17.52 (a) (b) Neodymium oxide 10.52 Praseodymium oxide 4.9 SiO2 3.4% Torium oxide 1.0 ZrO2 6.3 9.7% CaO 1.1 Other rare earths 9.6 Phosphate 25.7 CeO 28.0 In the same year, 1885, as Gorceix’s publication, DiO + LaO (?) 35.8 Total: 100.3 Bohuslav Brauner (1885-1935) in Prague and Carl Auer (a) Insoluble fraction; (b) Soluble fraction. von Welsbach (1858-1929) in Vienna analyzed didymium due to the well-founded suspicion about the possible Fraction solubilised by sulfuric acid contains: divisibility of didymium. Brauner was initially interested in fnding new evidence for the Periodic System of his Phosphate 28.7 % CeO 31.3 friend Mendeleev, as well as confrming P. Cleve’s work DiO + LaO (?) 39.9 Total: 99.9 signaling the decomposition of didymium into three elements (including Lecoq’s samarium there where indeed three elements).50 Auer decomposed didymium into two Hence, the formula is PO3.3[CeO,DiO,LaO],where 30% is phosphate and 70% are oxides of the three rare elements, neodymium and praseodymium, and after earths. Gorceix emphasizes that his analysis revealed many recrystallizations, fnally obtained the elements greater didymium content in the Caravelas sands when as double nitrates of ammonium – with slight difer- 51 compared with similar sands found in Slatoust (Rus- ences in solubility and with diferent colours. Former sia) and Arendal (Norway). Sands from Slatoust were attempts were unsuccessful, lacking fractional recrys- analysed by Debray at the École Normale. Gorceix’s tallizations of other salts, like double sulfates. Brauner paper was also published in the Comptes Rendus of the did not obtain neodymium and praseodymium com- Paris Academy,48 where it was presented by Jules Henri pounds, but rather obtained diferent spectral data for Debray (1827-1888), which should have given it a greater the two new elements. Six weeks before Auer’s publica- Almost a Discovery 37 tion, Gorceix obtained the same spectral lines as those supposedly divisible (1897).56 The same opinion was shown later by Auer’s compounds. Gorceix found no expressed in 1899 by Friedrich W. Muthmann (1861- sustainable explanation for his experimental results, and 1913).57 Richard Böhm, in 1902, interpreted the spectral did not claim priority over the “discovery” of a new ele- lines of praseodymium as pertaining to three “com- ment. We found no reference to any testing or repetition ponents” of the metal: Prα (λ 596,8 and 589,6), Prβ (λ of Gorceix’s experiments by other researchers. It seems, 481,1 and 440,0, announced by Cleve in 1878), and Prγ judging from his publications, that Gorceix was primar- (λ 469,0), identical to Diη described earlier by Gerhard ily interested in confrming the existence of didymium Krüss (1859-1895) and Lars Nilson (1840-1899).58 in monazite sands. Brauner and Gorceix based their fndings on the decomposition of didymium on spectral data, but in the case of didymium, as Robert Bunsen A POSSIBLE EXPLANATION FOR A PARADOX observed in 1866, the occurrence of “unusually narrow” spectral lines makes for a difcult interpretation. Wil- In 1885, the empirical observation that didymium liam Crookes (1832-1919), in 1886 and again in 1889, did not exist as an element, being confrmed as a mix- discussed the possibility of an even greater divisibility of ture of neodymium and praseodymium, was insufcient didymium than that predicted by Auer.52 to immediately remove it from the practices and activi- Modern chemists may fnd it surprising that even ties of many chemists. We have thus another example afer didymium was found to be a mixture of neodym- of the persistence in science of ideas and facts that are ium and praseodymium, some chemists persevered in no longer acceptable. Such persistence is perhaps better studying didymium as an “element”. Even more surpris- understood as the resistance of some chemists to alter- ing, many chemists still believed, in late 19th / early 20th ing the body of data and beliefs guiding their practices. centuries, that neodymium and praseodymium could It is also an example of how scientifc methodologies be divided into new “elements” on the basis of empiri- may justify anachronisms. “Even formally excluded from cal (mostly spectral) data. Richard Böhm’s book, men- the row of elements – wrote Böhm in his text from 1905 tioned above, presents useful data for that purpose. Fur- – exclusively practical motivations led to a discussion ther, many orthodox chemists investigated new “simple about its preparation since for obtaining its components compounds” as elements. Carl von Scheele, in 1901, afer chemists frequently use materials rich in didymium as many analyses, asserted that praseodymium has in fact a raw material.”59 An 1898 paper by André Job (1870- an elemental nature.53 Konstantin von Chroustschof 1928),60 professor at the Conservatoire National des Arts (1852-1912) announced in 1897 a third component of et Métiers, which described “new chemical compounds didymium, the supposed “element” glaucodidymium, derived from cerite metals”, registers the preparation although this was never confrmed. For Eugene Demar- of oxalochloride of lanthanum obtained from lantha- çay (1852-1903), the discoverer of europium (1901), neo- num sulfate, for which he suggests the formula (C2O4) dymium was, without a doubt, an element (1898).54 Cl2La2. Dissolved in hot water, this salt decomposes into Against current and almost universally accepted lanthanum oxalate and lanthanum chloride. So far, this emerging chemical facts, some chemists insisted on is nothing unusual. However, Job also mentions that it the divisibility of the two new elements isolated from is necessary to start with very pure lanthanum that is didymium. In 1892, Paul Albert Schottländer (1843- free from cerium and didymium (this in 1898), with the 1897), who earned a doctorate in Würzburg and was an “spectroscope showing no more any signs of didymium”, amateur chemist in Berlin (1886/1896), published obser- and various new analytical methods developed by Job vations of the crystallization of double nitrate of praseo- himself showing no signs of cerium. In the same paper, dymium and ammonium: Job refers to the preparation of the same type of salt with cerium and with didymium, alluding to the didym- from the components of praseodymium, one element con- ium oxalonitrate prepared previously by Cleve. Richard sidered to be a metal presented only one absorption line (λ Böhm suggests that this situation could be explained by 468,9) [ … ]. Te other praseodymium components consti- the evolution of chemistry itself: tute two groups, which seem to sufer separation during the crystallization process. One of them we call Prα [ … ] and 55 Since older scientists studied very little or nothing about the other Prβ. spectra, lacking therefore a resource to confrm the absence of lanthanum, they took for didymium oxide all products Also, for the Americans, Louis Monroe Dennis which precipitate as oxalate in an acid solution containing (1863-1936), from Cornell University, and his student didymium and lanthanum, but not cerium.61 Emile Monnin Chamot (1868-1950), praseodymium was 38 Juergen Heinrich Maar

Te situation just described is not unique to the feld in chemistry is the rapid triumph of Lavoisier’s Oxygen of chemistry. But how is it possible for an exact, meth- Teory: afer the death of its most prominent opponent, odologically structured and practiced science to contin- Joseph Priestley (1804), adherents of the old theory rap- ue to operate on obsolete data? Even though there may idly lef the scene. Te “old chemistry” then ceased to be a practical justifcation for the reluctance to reject exist with the death of its last representative, Anders falsifed beliefs, is this not a case of bad science? Or bad Retzius (1742-1821). scientists? Was there an excessive emphasis on empirical In the case of the “element” didymium, we are not facts? Or was practice too distant from theory? discussing theories, but rather the reluctance to accept A possible explanation would be a general unaware- the fact that it is a mixture and not an element. Its fal- ness of the latest research results on didymium. How- sifcation occurred during a period of “normal science” ever, this is not the case here; the international chemi- (in Kuhnian sense) in chemistry. Te general theoretical cal bibliography had already incorporated neodymium framework of chemistry is not being called into ques- and praseodymium, while excluding didymium from the tion, but we may identify some problematic methodolog- series of elements. Alternatively, this could be the result ical aspects of laboratorial work, such as overreliance on of an isolated group of researchers studying the “cerite spectral data. Due to the prolifc amount of work on rare metals” unaware of the newest literature, and therefore earths, we can fnd many situations resembling the case continuing to work on outdated data. Tis situation is of didymium, or cases where scientists were reluctant to ofen observed in the scientifc practice of the so-called accept well documented falsifcation. (geographically) “peripheral science”, where the difu- And we also fnd amongst the history of rare earths sion of new information is ofen slower. discoveries examples situated at the other end of the Tere may be other explanations originating from spectrum; cases where elements were well confrmed the theoretical frameworks prevalent in diferent coun- and widely believed to exist, and yet some were reluctant tries. We know from history that the anti-atomist think- to accept them (not only for scientifc reasons). George ing of positivists like Jean-Baptiste Dumas (1800-1884) de Hevesy’s (1885-1966) and Dirk Coster’s (1889-1950) and Marcelin Berthelot (1827-1907) had a negative infu- hafnium (1923) was unanimously accepted only afer ence on the evolution of several aspects of French sci- George Urbain’s celtium, discovered six months before, ence. But disbelief in the existence of atoms as real and was shown to be the same as lutetium, discovered by concrete entities did not prevent chemists from discover- Urbain himself in 1913, from another mineral, samar- ing new elements, as is proved by Mendelevian eca-ele- skite. ments gallium (1875, Lecoq de Boisbaudran), scandium When discussing rare earths elements, it is impor- (1880, Nilson) and germanium (1886, Winkler). tant to consider the enormous difficulties faced by In other words, science may progress in the sense of chemists. Te properties of rare earths elements are so making discoveries, despite false beliefs or wrong theo- similar that identification, isolation and purification ries. Tis fact could in turn psychologically explain why are extremely difcult. Chemists ofen misinterpreted we experience a kind of inertia or resistance from the mistakes, sometimes thinking a mixture to be a “new” scientifc community when promptly revising beliefs or element, and sometimes concluding that diferent sam- theoretical frameworks in the face of anomalies, as it ples of the same element were diferent elements. Fran- may not be necessary to do so to make new discoveries. co Calascibetta points to theoretical and experimental In this case, the high level of complexity in the experi- problems: mental study of rare earths elements adds to this inertia. Revising theoretical frameworks in light of anomalies Te vast number in the family, their appearance, multi- conficts with the desire to prioritize new discoveries, plication and then disappearance, have several times been which in turns explains why scientists may choose to linked, in both the last and the current century, to impor- ignore such anomalies and carry on with their research tant theoretical aspects of chemistry, including Mendeleev’s periodic system and later Moseley’s discovery and the new programs. defnition of atomic number.62 Te desire for priority also explains the hasty communication of new rare earths discoveries that were Before the invention of spectroscopy, notably that later found to be just mixtures, or rediscoveries. Physi- of Bunsen and Kirchhof (1859),63 chemists had at their cists and chemists know all too well Max Planck’s quote disposal very difcult, troublesome and labour-intensive on the triumph of new theories: it is not caused by the methods for identifcation and characterization of these strength of their arguments, but by the death of the elements. Notwithstanding, experimented chemists like defenders of the older views. Te best example of this Brauner, Auer, Marignac or Urbain were still able to Almost a Discovery 39 characterize new elements using only “classical” analytic cent colours, producing objects much appreciated during methods. the Art Nouveau and Art Déco periods. In the 1920s, the An initial limit to an apparently unlimited num- chemists of Karlovy Vary again mixed neodymium and ber of possible elements was Mendeleev’s number of praseodymium salts, a kind of ‘synthesis’ of didymium, empty spaces (elements still unknown) in his Peri- obtaining several pigments. odic Table, which indicated that there is indeed a limit One more reason has been presented to explain to the number of possible rare earth elements (1869). didymium’s longevity. F. Szabadváry and C. Evans Convinced Mendelevians suggested new graphical rep- (1996) ofered a fanciful explanation for the permanence resentations of the table that allowed the correct loca- of the term “didymium” in the chemical literature: that tion of the elements still to be found, as in the tables of Auer published his paper on the isolation of neodym- Brauner (1902)64 and of Alfred Werner (1905)65. Brauner ium and praseodymium from didymium in an obscure included 19 rare earth elements in his table, seven of chemical journal, the Monatshefe für Chemie.68 But in which were still unknown at the time. Werner proposed Auer’s time, and for many decades, this journal was all 15 rare earth elements, two of them (between neodym- but obscure. ium and samarium) not yet discovered. Afer Henry G. J. Moseley’s (1887-1915) studies with X-ray spectroscopy in Oxford (1913) it was fnally possible to defne EPILOGUE – RUDIMENTS OF AN INDUSTRY the “atomic number” of elements and conclude that the maximum theoretically possible number of rare earth Unfortunately, rare earths constitute an additional elements is 14. Two of these were not yet known: atomic example of underused Brazilian natural resources, thanks number 43 (technetium, 1937) and atomic number 61 to lack of strategic policies of management69 is the sad (promethium, 1945). Finally, by combining empirical conclusion of chemist Osvaldo Serra (*1943) on these knowledge with theory, it was possible to determine resources, which are known for their history, occurrence which ones exist in the vast universe of possible rare and prospection, but were never really considered serious- earth elements – exhaustively described by Marco Fon- ly by the academy and industry. At the same time, Serra tani and coworkers.66 mentions a phenomenon that is not restricted to Brazil, In other words, elements confrmed based on empir- but that occurs worldwide: the dismantling of mineral ical facts alone, but in disagreement with theory, could processing industries not only causes unemployment, but fnally be discarded. Joseph William Mellor (1869-1938), also the disappearance of knowledge and rare skills and in his extensive treatise on Inorganic Chemistry, pre- abilities necessary for mineral processing procedures. sents a table with 73 supposed rare earth elements dis- Te recent resumption of extraction, separation and covered between 1794 and 1920.67 Out of these, 15 were purifcation of rare earths in Brazil is actually a new mixtures and 25 were never confirmed. Empiricism start, as all that was previously known about these pro- without proper theoretical foundations leads to error. cedures was lost. Until 1914, Brazil was the world leader Tis explanation, of course, does not apply just to the of rare earth extraction, but it brought no beneft to the case of didymium. country: “the ballast of Brazilian ships” created richness Still other (non-scientific) motivations exist for elsewhere. Decades later, Brazil began a timid indus- didymium’s persistence, and these are exclusive to trial exploration of monazite sand with the foundation, didymium. Didymium, when still believed to be an ele- in 1942, of the ORQUIMA S.A. (Organo-Química, later ment, was used in the production of special glasses for Indústrias Químicas Reunidas) in São Paulo, with a fac- the protective goggles of glassblowers. Even afer its fal- tory in Santo Amaro/São Paulo.70 In 1946, ORQUIMA sifcation, the term didymium continued to be used, began to process monazite sand not from Bahia (because meaning the fraction remaining afer the removal of all of the long distance) but from the States of Espírito San- cerium content from monazite (this fraction contains to (where it was discovered in 1896 and has been pro- 46% lanthanum, 34% neodymium, 11% praseodymium, cessed since 1906 by MIBRA –Société Miniére et Indus- and some samarium and gadolinium). Mixtures of neo- trielle Franco-Brésilienne) and Rio de Janeiro. dymium and praseodymium, like a ‘false didymium’, ORQUIMA Company developed and used with per- were used as catalysts and in the glass industry. Ludwig fection all the know-how necessary for the extraction, Moser (1833-1916) founded a factory that produces glass- processing, isolation and purifcation of rare earths – ware (Josef Moser & Söhne) in 1857 in Karlsbad (Bohe- especially during the 20 years in which the Polish chem- mia), now Karlovy Vary, which used neodymium and ist Pawel Krumholz (1909-1973), a former assistant of praseodymium salts to obtain a great variety of irides- Fritz Feigl (1891-1971) in Vienna, was in charge as tech- 40 Juergen Heinrich Maar nical director. Krumholz, along with an accomplished nas, 2000. Manuel Ferreira da Câmara (1762-1835), group of coworkers, produced at ORQUIMA an inter- member of the National Assembly, proposed in 1823 nationally acclaimed work on rare earths.71 ORQUIMA the creation of an University in Rio de Janeiro, the processed 2000 tons of monazite annually, producing “Instituto Brasílico” (Campos, E. S., “História da compounds of cerium, lanthanum, neodymium, samar- Universidade de São Paulo”, Editora da Universidade ium, thorium, zirconium, with a purity of up to 99.99%. de São Paulo, 2004, p.27). It was nationalized in 1949 and acquired by the Comis- 10. Habashi, F., Bulletin of the Canadian Institute of são Nacional de Energia Nuclear – CNEN – in 1960, Mines, 90, 103-114 (1997). during the nationalistic and state-centralizing policies 11. Carvalho, J. M. de, “A Escola de Minas de Ouro Pre- of the governments of that period, and transferred later to – o Peso da Glória”, Editora da Universidade Fede- to the NUCLEMON (Nuclebrás Areias Monazíticas). All ral de Minas Gerais, Belo Horizonte, 2002. ORQUIMA’s activities were sadly paralyzed in 2002; 12. Schwarcz, L., “As Barbas do Imperador”, Cia. das lacking the deserved support, the enterprise failed. Letras, São Paulo, 1998. 13. Santos, N. P. dos, Revista da SBHC, 2, 54-64 (2004); Filgueiras, C. A., Química Nova, 11, 210-214 (1988). AKNOWLEDGEMENTS 14. Schwarcz, L., op. cit., pp. 125-158, 416-436. 15. Silva, C. B. da, Revista da Escola de Minas, Ouro Pre- Te author wishes to express his gratitude to the to, 67, 319-340 (2014). following: René and Peter van der Krogt, Delf, Neth- 16. Figueirôa, S., “As Ciências Geológicas no Brasil. Uma erlands; William Jensen, University of Cincinnati and história social e institucional, 1875-1934”, Editora Oesper Collection for the History of Chemistry; Organic Hucitec, São Paulo, 1997, 123-124. Chemistry Department ‘Ugo Schif’, University of Flor- 17. Dutra, C. V., op. cit., 186. ence; Marco Fontani, University of Florence; Laura Colli 18. Atêncio, D., Brazilian Journal of Geology, 45, 143-158 and Museo di Storia della Scienza, Florence; Museu de (2015). Arte de Santa Catarina/MASC, Florianópolis, Brazil. 19. Costa, A. R. da, Santos, P. C. M., Revista da Escola de Minas, Ouro Preto, 59, 347-353 (2006). 20. Ambrosio, U., Anais do IV Seminário Nacional de REFERENCES História da Ciência e da Tecnologia, Caxambu, 1993, 96-99. 1. Gorceix, H., Inaugural Speech, Ouro Preto Mining 21. Carvalho, J. M., op. cit., 50-51. School, October 12th, 1876. 22. Carvalho, J. M., op. cit., 92-97. 2. Maar, J. H., “História da Química, vol.II, de Lavoi- 23. Gorceix, H., Anais da Escola de Minas de Ouro Preto, sier ao Sistema Periódico”, Papa-Livro, Florianópolis, vol. 1 (1881), preface. 2012, 544. 24. Carvalho, J. M., op. cit., 56-57. 3. Karpenko, V., Ambix, 27, 77-102 (1980). 25. Carvalho, J. M., op. cit., 98. 4. Anonymous, Manufacturer & Builder, 122 (7), nº 26. Carvalho, J. M., op. cit., 66, 96. 976, 84. 27. João Pandiá Calógeras (1870-1934), engineer and 5. Cláudio Vieira Dutra, professor at the Federal Uni- geologist, was Minister of Agriculture (1914/1915), versity of Ouro Preto, was the frst to draw attention Finance (1915/1917) and War (1919/1922), and pre- to these facts, in 2002, and to him pertains the prior- sided the Brazilian delegation to the Versailles Con- ity of including this novelty into the history of chem- ference (1918/1919). He authored several books on istry. Dutra, C. V., Revista da Escola de Minas, Ouro History, Economy, International Relations. Dias, M. Preto, 55, 185-192 (2002). M., “O Brasil em Versalhes – Calógeras e a Política 6. Alves, G. L., “O Pensamento Burguês no Seminário Internacional”, Anais do 3º Seminário Nacional de de Olinda”, Editora da Universidade Federal de Mato História da Historiografa, Ouro Preto, 2009. Grosso do Sul/Editora Autores Associados, Campo 28. José Pires do Rio (1880-1950) served as government Grande e Campinas, 2001. minister and as mayor of São Paulo 1926/1930. 7. Maar, J. H., Scientiae Studia, 2, 33-84 (2004). 29. Francisco Sá (1862-1936) was minister in 1909 and 8. Cunha, L., “A Universidade Temporã”, Editora Civili- 1922/1926 and senator from 1906 to 1927. Calógeras, zação Brasileira, Rio de Janeiro, 1980. Francisco Sá and Pires do Rio, were typical represen- 9. Figueirôa, S., “Um olhar sobre o passado”, Edito- tatives of the so-called “First Republic” (1889/1930); ra da Universidade Estadual de Campinas, Campi- all had technological training as engineers and exert- Almost a Discovery 41

ed strong infuence on economic questions. 45. Gorceix, H., Anais da Escola de Minas de Ouro Preto, 30. Guimarães, D., apud Carvalho, J. M., op. cit., p. 132. 4, 29-48 (1885). Djalma Guimarães (1894-1973), Emeritus Professor 46. Gorceix, H., Compt. Rend., 94, 1446-1448 (1884). of the Universidade Federal de Ouro Preto, was per- 47. Gorceix, H., Bull. Soc. Min. France, 8, 32-35 (1885). haps the most important Brazilian geologist, discov- 48. Gorceix, H., Compt. Rend., 100, 356-358 (1885). erer of the great niobium reserves near Araxá/Minas 49. Böhm, R., “Die Verwendung der Seltenen Erden”, Gerais (1931/1932). Veit & Co., Leipzig, 1913. 31. Derby, O., Science, 1 (8), 211-214 (1883). 50. Cleve, P., Bull. Soc. Chim. France, 21, 196- (1874). 32. Haitinger, L., Peters, K., Sitzungsberichte Akademie 51. Auer von Welsbach, C., Monatshefe für Chemie, 6, Wien, 1904, 159-160. 477-491 (1885). 33. Leonardos, O., “A Monazita do Estado da Bahia”, 52. Crookes, W., Chemical News, 60, 27- (1889) Departamento Nacional de Produção Mineral, Rio 53. Scheele, C. von, Z. Anorg. Allg. Chem., 27, 53-57 de Janeiro, 1937. (1901). 34. For a critical review and production history of 54. Demarçay, E., Compt. Rend., 126, 1039-1041 (1898). “Notorious”, see “Notorious” (1946): Hitchcock’s 55. Schottländer, P., Berichte, 25, 378-394 (1892). mature and intricate espionage masterpiece”, https:// 56. Dennis, L., Chamot, E., J. Am. Chem. Soc., 19, 799- cinephiliaandbeyond.org. Te screenplay was written 809 (1897). by Ben Hecht (1894-1964). 57. Muthmann, W., Stützel, L., Berichte, 38, 2653-2677 35. Langmuir, A., “Index to the Literature of Didymium”, (1899). Smithsonian Miscellaneous Collections, Washington, 58. Böhm, R., Z. Angew. Chem., 15, 1282-1299 (1902). 1893. 59. Böhm, R., op. cit (1905), 469. 36. Herzfeld, O., Korn, O., “Chemie der Seltenen Erden”, 60. Job, A., Compt. Rend., 126, 246-248 (1898). Springer, Berlin, 1901. 61. Böhm, R., op. cit. (1905), 470. 37. Fontani, M., Costa, M., Orma, M., “Te Lost Ele- 62. Calascibetta, F., VII Convegno Nazionale di Storia e ments”, Oxford University Press, 2015, 40. Fondamenti della Chimica, L’Aquila, 1997, 259-272. 38. Böhm, R., “Die Darstellung der Seltenen Erden”, 2 63. Bunsen, R., Kirchhoff, G., Annalen der Physik volumes, Veit & Co., Leipzig, 1905. (Pogg.), 110, 161-189 (1860). 39. Hillebrand, W., Norton, J., Annalen der Physik 64. Brauner, B., Zeitschr. Anorg. Chem., 32, 1-30 (1902). (Pogg.), 155, 633-639 (1875); ibid., 156. 456-474 65. Werner, A., Berichte, 38, 914-921 (1905). (1876). 66. Fontani, M., Costa, M., Orma, M., “Te Lost Ele- 40. Selleri, S., Fontani, M., “Cent’anni dalla scomparsa ments”, Oxford University Press, 2015. di Ugo Schif”, Consiglio Regionale, Florence, 2016, 67. Mellor, J., apud Calascibetta, op. cit., p. 264. 63-80. 68. Szabadváry, F., Evans, C., “Episodes from the History 41. Frerichs, F., Berichte, 7, 331-366 (1878). of Rare Earth Elements”, Kluwer Academic Publish- 42. Gorceix, H., Anais da Escola de Minas de Ouro Pre- ers, Dordrecht, 1996, 64. to, 4, 29-48 (1885). 69. Sousa Filho, P., Serra, O., Química Nova, 37, 753-760 43. Gorceix, H., Compt. Rend., 98, 1446-1338 (1884). (2014). 44. Gorceix, H., Bull. Soc. Min. France, 7, 179-182 70. Serra, O., J. Braz. Chem. Soc., 22, 811-812 (2011). (1884). 71. Vichi, E., Química Nova, 6, 152-156 (1983).

Firenze University Press www.fupress.com/substantia

Research Article Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 Citation: J. Elliston (2018) Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1. Substantia 2(2): 43-71. doi: 10.13128/substantia- John Elliston 60 Elliston Research Associates Pty Ltd, 10B Te Bulwark, CASTLECRAG 2068, New South Copyright: © 2018 J. Elliston. This is Wales, Australia an open access, peer-reviewed article E-mail: [email protected] published by Firenze University Press (http://www.fupress.com/substantia) and distribuited under the terms of the Abstract. Te crystalline forms of quartz and that it is insoluble are well known. Te Creative Commons Attribution License, hydrolysis of silica and formation of polymeric silicic acids in slightly alkaline (pH 8.1 which permits unrestricted use, distri- - 8.3) sea water is less well known and not recognised by many geologists. Silica is one bution, and reproduction in any medi- of the most abundant components of ordinary sediments and the amorphous forms, um, provided the original author and source are credited. the silica gels, have been closely studied. Teir properties and behaviour are perhaps better known than those of many other colloids. Quartz veins are abundant in all types Data Availability Statement: All rel- of sediments and in rocks and mineral deposits derived from them. Clearly if we study evant data are within the paper and its how silica is mobilised into these veins and lodes, how it silicifes wall rocks, forms Supporting Information fles. opal, replaces shells and tree trunks, etc. then we may have a better basis for understanding how sulphide particles might similarly be mobilised into veins and lodes, per- Competing Interests: The Author(s) declare(s) no confict of interest. meate shales, form framboids, replace fossils and plant fragments or fne shale bands, etc. Tis article briefy summarises recent developments in the aqueous chemistry of silica. It emphasises the particulate nature of the amorphous silica species in order to update rather simplistic views that quartz veins and many other natural forms of quartz “crystallise directly from solution”. A number of features of quartz veins which are due to the particulate nature of natural polymeric silica are illustrated.

Keywords. Sediments, accretion, concretion, charged particle, surface energy, porphyroid, granite.

INTRODUCTION

Quartz veins and silicifcation are the simplest and most frequently encountered of all mineral concentrations. However, current geological theory is still unable to explain: -

1) Te occurrence of insoluble oxides of iron and silica in large intrusive and brecciated quartz-magnetite lodes. Chemically, iron silicates should have been formed from soluble silicates and ferrous or ferric salts, or similarly if the oxides had been fused together. 2) Te suspension of heavy brecciated wall-rock fragments or angular magnetite fragments in quartz veins and lodes. 3) Quartz veins and veinlets or occasionally large lodes of quartz which arise and die out locally in the rock in which they occur without any

Substantia. An International Journal of the History of Chemistry 2(2): 43-71, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-60 44 John Elliston

sign of the passage of large volumes of “hydrother- quartz veins are attributed to “hydrothermal solutions”, mal fuid” such as would have been needed to dis- which are supposed to have deposited the crystalline sil- solve “precipitated” silica. ica from solution. All experienced geologists know that 4) Te multiple injection and re-injection of quartz in this is too simplistic. Quartz veins occur in many situa- the same vein channel. tions where this direct crystallisation from solution does 5) Te occurrence of crystal lined cavities in quartz not explain the observations. In fact, most commonly, veins. the quartz of the veins appears to have been intruded or 6) Te occurrence of chert veins in granite. injected (as discordant squirts) into the enclosing rock in 7) Te less frequently observed but very real occur- bulk, as a fuid mass of silica, as if it had been molten rences of ptygmatic quartz veins, colloform banding except that there is no evidence of temperatures such as in quartz veins, oolites and concretionary structures would melt quartz. It melts or fuses gradually like glass in quartz veins, or spherulitic structures, rosettes, at 1723°C. Frondel1, 1962, p. 3; Iler2, 1979, p. 15). Molten axiolitic overgrowths, and acicular crystals in vein quartz does not crystallise on cooling but hardens to an quartz. amorphous glassy form of silicon dioxide called lechat- 8) Quartz oolites found in granite and in porphyroids. elierite. 9) Fluid inclusions containing brines, petroleum, bio- Most feld geologists tacitly accept that quartz veins genic organic matter, or carbonates in quartz veins. seem to “sweat out of the rocks”. We clearly believe it has 10) Boehm lamellae or lines of fluid inclusions that to do with the movement of aqueous fuids. In the light extend across crystal boundaries in quartz veins. of the currently published aqueous chemistry of silica, 11) Silicifcation of wall rocks, vein margins, and occa- the earlier reasoning that attributed vein quartz accu- sional shells, tree trunks, etc. by silica which appears mulations to vast quantities of solution based on meas- to be associated with and occur contemporaneously ured solubilities of silica is clearly false. Barnes3, 1967, with quartz veins. p. 392 suggests fve cubic kilometres of water would be required to deposit 100,000 tonnes of quartz in a moder- Tis paper provides adequate explanations for all ate sized vein approximately 128m x 75m x 4m. these observations. Tey are a logical and expected out- Tere is a signifcant disparity between the modern come when current hydrolysis reactions and the physical chemist’s and the classical geologist’s view of the “solu- chemistry of polymeric particulate species of silica are bility” of silica and silicates. Tis article provides the applied. information to bridge that gap. Simple quartz veins are the most common of all mineral deposits. Tey are found in all rock types with the exception of congealed molten lava fows and associ- THE FORMATION OF POLYMERIC SILICA IN ated pyroclastic debris. Tey mainly occur in sediments, NATURAL SEDIMENTS metamorphics, porphyroid and granite related intrusives. Quartz veins are so commonplace with such a wide Quartz is virtually insoluble in water. Dispersion of variety of associations and vein types and encountered hydrated silica is very variable because: - in every feld situation that questions relating to details (a) it takes a long time as measured experimentally (but of their genesis and to the mechanism of their emplace- very quickly indeed if measured in units of geologi- ment are seldom raised. cal time) for equilibrium to become established; It is clear that most geologists don’t really under- (b) the solubility of quartz is a function of temperature, stand how quartz veins are formed. One cannot go pressure, ionic strength, pH, and the presence of straightforwardly to a standard textbook and look up complexing ions; the genesis of a quartz vein. All textbooks contain abun- (c) silica disperses in water to form three classes of sol- dant references to quartz, its mineralogy, varieties, crys- utes (Iler2, 1979, pp 172-248; Stöber4, 1966, pp 161- tallography, properties, features, and associated miner- 163): - als. Information has been published relative to the solu- i) simple monomeric complexes such as Si(OH)4 bility of quartz under various conditions of acidity, tem- aq. perature, pressure, etc., and analyses of the silica content ii) oligomers or polynuclear complexes such as 2- of various natural waters can also be found. Si4O6(OH)6 which represent condensed chains Why then is the classical origin of quartz veins, the of Si(OH)4 tetrahedra (Figure 1). Tese chains most common and abundant of all mineral deposits, appear to grow to a maximum of about six shrouded in mystery? Generally, in geological teaching, Si(OH)4 tetrahedra. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 45

iii) poly silicic acid where very long chains are produced with cross-linking to yield partly condensed three dimensional networks. These polymeric species are ‘colloidal’ in that they are frequently loose spherical aggregates of the size that produces scattering of light or gives a “milky” appearance (Figure 2). (d) the higher polymeric forms of silicic acid tend to occur in solution, re-adsorbed on surfaces, or in suspension as colloidal sol particles, depending on hydrolysis and ionisation.

Surface charge

All solid and crystal surfaces are charged including very small gelatinous natural sediment particles such as clay and the minute globular particles of polymeric silica. Figure 3 is a TEM image (from Stumm5, 1992) that shows the “bumps” due to charge sites where additional molecules would join the surface to result in crystal growth. Tese electrical charges are arranged in minis- cule steps on the crystal surface. Te surface reactivity difers according to the way the units of the lattice are Figure 1. Tis diagram represents silicic acid molecules. Te sphe- exposed at its surface. res represent oxygen atoms and the dots represent hydrogen atoms. If we are to understand why the globular particles or Silicon atoms within the oxygen tetrahedra are not visible. “little balls” of silica interact with each other and form the precipitates, chains, coatings of adsorbed particles, and gel mesh-works as described by Iler2, it is necessary to consider the basic hydrolysis of the mineral- water interface. Similar equilibria relate to a great many

Figure 2. Tis diagram shows the larger polymers that develop into “little balls”. (A) is trisilicic acid, (B) is cubic octasilicic acid that forms larger particles (C) and (D) by condensing monomer to form closed rings until the original species is surrounded by a layer of deposited silica bearing silanol groups. Diferent kinds of incomple- Figure 3. On anhydrous crystal surfaces the lattice units are arran- tely condensed oligomers are thought to form the cores of colloidal ged in layers terminating in ‘steps’. Tis STM image shows the step- particles but above pH 7 the inner silica contains relatively few sila- like layers on a crystal face at which further crystal growth or disso- nol groups. lution are believed to occur. (From Stumm5, 1992.). 46 John Elliston

irregularities but in natural basin sediments clay minerals present really enormous areas of Si-OH terminations to equilibrate with pore fuid sols and solutions. Tese immense surfaces are fully loaded with adsorbed ions and charged particles. Tey all carry adsorbed water monolayers and difuse charge layers in which the co- ions and counter-ions reside in a narrow zone out from the surface (Hemholtz double layers – see Glossary).

Te growth of natural polymeric silica particles

Silicic acid polymerises to discrete particles which then aggregate into chains and mesh-works. As Iler2 Figure 4. If crystalline silica is broken the “broken bonds” or expo- (1979, p. 173) points out, three stages are actually recog- sed surface charges in an aqueous system are immediately satisfed nised. by dissociation of water molecules. If the surface is large it will car- 1) Polymerisation of monomer to form particles. ry an ‘overall’ or residual charge which adsorbs polar water molecu- 2) Growth of particles. les as an adsorbed water monolayer. 3) Linking of particles together into branched chains, then mesh-works, fnally extending throughout the medium thus bestowing a certain degree of rigidity surfaces and certainly to silica and the rock-forming upon it. silicates. It is also important for sulphide minerals. It Below 200°C amorphous silica precipitates from applies as soon as precipitating insoluble sulphides are supersaturated solution under various conditions and in able to form a lattice. Tis surface hydration occurs on several physical forms. Tese range from colloidal par- the minute ‘curds’ or crystallites that develop when sul- ticles so small that the suspension or sol remains trans- phide particles are frst precipitated. parent or so large that it looks milky. Depending on par- In the simplest case, if quartz or a grain of sand is broken, the silicon - oxygen chemical bonds that form the lattice are exposed as indicated in Figure 4. If the surface is large it will carry an ‘overall’ or residual charge which adsorbs polar water molecules as an adsorbed water monolayer. Tis efect can be considered as ‘wetting’ or it can be manifest in the development of a meniscus like that in a test tube. If the surface is small as in Si(OH)4 or the oligomeric silicic acids at equilibrium, there is a situation very like that of water dissociation which is:- - + H2O ⇄ OH + H + + H2O + H ⇄ H3O Terefore for silica the dissociation is:- Si OH ⇄ SiO- + H+ and the net charge on the surface is controlled by the equilibrium for:- + + Si OH + H ⇄ Si OH2 Only a few sediments like BIF’s, deep sea oozes, quartzites or cherts have silica as the dominant component. Clays are the major component of most ordinary Figure 5. Tis diagrammatic representation of the polymerisation basin sediments. However, the surface of most clay par- behaviour of silicic acid indicates the general conditions under which the colloidal particles develop and ‘chain’ into three dimen- ticles is the tetrahedral layer or net-like Si–O–Si linkages sional gel meshworks. In alkaline solution the particles are negati- with Si-OH terminations to the exterior surfaces of these vely charged and remain discrete as they grow in size and decrease large plate-like macromolecules. Te net charge on clay in number. Below pH 7 or in the presence of electrolyte (as in sedi- particle surfaces is due to ‘unsatisfed’ or ‘broken bond’ ment pore fuids) precipitates or gels are formed. [From Iler2, 1979, p. 174]. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 47 ticle size, it precipitates as gelatinous masses or as hard water content. Such polymerisation reactions into glassy flms. semi-rigid gels are ofen promoted by increased elec- Figure 5 (from Iler2, 1979) shows diagramatically the trolyte concentration and slightly decreased water polymerisation behaviour of silicic acid in the pH range content. Tey may be strongly sensitive to pH. 7-10 and in relation to the presence of salts (the condi- b) disordering or hydrolytic degradation of crystal lat- tions which exist in most pore fuids). tices into high porosity mesh-works with which the Polymerisation is slowest at about pH 2 and increas- solvent, water, is intimately associated. Ionic difu- es in proportion to the hydroxyl ion concentration up to sion at 20 - 30% of its rate in pure solvent is ofen about pH 10. Higher polymers are formed mainly by the observed in such gels. further addition of the monomer. It must be emphasised c) the ‘stacking’ or aggregation of colloidal sized plate- that the ionisation constants of the polymers are greater like, rod-shaped, or globular particles into random, than that of the monomer. Tus, the monomer reacts house of cards-like, or semi-ordered mesh-works more rapidly with the dimer and higher polymers than and networks. with another monomer. Tis is the reason why the natu- A similar electrical potential gradient exists within ral precursor silica gels in veins are “inflled” and den- gels as that on the planar surface (the surface potential). sifed and can be regarded as “collectors” of additional On its internal surfaces where water molecules interface silica from the seepage of pore fuids draining from sedi- with oxide or silicate surfaces, centres of charge develop ments or sheared sedimentary materials that form meta- such as: - + - morphic rocks. Si - OH2 or Si - O Te cohesive matricies of consolidating sediments, Tese sites exist on the surface of the gel and within or those which form the precursor silica gels in vein, its pore structure. Te attraction-repulsion due to sur- dyke, breccia pipe, and lode outfow systems are essen- face charge on colloidal particles has now been meas- tially mixed gels. Figure 1 shows molecular models ured. In 1998 a team of researchers led by Professor T. of the various short chain silicic acids which are small W. Healy (personal communication) at the Particulate enough to permeate through these porous systems. Fluids Processing Centre, University of Melbourne, were Tese short chain acids grow by condensation of the successful in adapting an atomic force microscope to monomer to form closed rings until the original species measure the interparticle forces in relation to minute is completely surrounded by one layer of deposited silica distances between them. Tis brilliant work confrms bearing silanol groups. Diferent kinds of incompletely DLVO theory (Figure 6). condensed oligomers may form in the cores of particles As for the simple planar surface of a broken crystal, as these silica polymers aggregate to colloidal dimen- the net charge on the gel surfaces or the surfaces of its sions. component particles is controlled by the equilibrium for:- + + SiOH + H ⇄ SiOH2 Tis equilibrium defnes the net surface charge. THE NATURE OF ADSORPTION PHENOMENA As in the case of water where: - + + OH + 2H ⇄ H3O Te elementary surface chemistry of the mineral- and the concentration of H+ equals the concentration of water interface is simple but basin sediments are largely OH- at pH 7, so for oxide and silicate surfaces a pH will comprised of clays and other components that consist defne for each oxide a condition where the concentra- of small colloidal particles with a very large surface-to- tion of negative sites equals the concentration of positive volume ratio. sites. Tis pH is called the point-of-zero-charge or p.z.c. If these basin muds were dispersed in river water as Figure 7 (from Healy6, 1972, p. 38) is a diagram rep- when they were transported to the sea, they may form resenting the electrical double layer on colloidal particles suspensions and metastable sols. But suspensions set- and the variation of their net surface charge with pH. tle and sols coagulate in sea-water so that the resulting Tese particles may be dispersed as a sol but the same net sediment is a semi-solid three-dimensional meshwork of surface charge to solvent equilibria apply when particles high porosity containing solvent. are coagulated. In a gel some of the charge sites are satis- Such three dimensional semi-rigid, largely amor- fed by mutually near-contacting or interacting particles. phous, heterogeneous mesh-works are technically gels Where pH values are more alkaline or greater than which are produced by:- the p.z.c., the particles have a net negative surface charge a) polymerising small molecules [like Si(OH)4] into and where pH values are more acid or less than the p.z.c. random, cross-linked network structures of high the net surface charges are positive. Te p.z.c. values of 48 John Elliston

Figure 6. Te general form of the curve of potential energy is Figure 7. Tis is a schematic representation of the electrical double plotted as a function of particle separation (interparticle distance layer on colloidal sol particles and the variation of net surface char- 6 . or concentration of the paste) for interaction between particles in ge with pH. [From Healy , 1972, p. 38] a colloidal system. Vr and Va are repulsive and attractive energies respectively and the curve represents their sum at a given interparticle distance. Tis is characteristic of sediment particle interactions. Te general behaviour of particles at various distances from each other is indicated at the top of the diagram. Note the critical interparticle separation at which accretions begin to form. Tis curve describes the general behaviour of particles resulting from the interaction of these forces and is not specifc to any particular type of colloidal particle. Tis succession of repulsion, strong attraction and resistance at exceedingly close proximity is called the DLVO theory that has now been confrmed by direct measurement with an adapted atomic force microscope.

simple oxides vary from pH 2 for SiO2, pH 6 for TiO2, and pH 9.1 for Al2O3. Some p.z.c. values for common minerals are listed in Table 1 (From Healy6, 1972, p. 39). Te existence of surface charges means that surfaces in contact with a solvent must take up or adsorb ions from solution. Tis occurs on all sediment grains but mainly on the clays and other sediment colloids where surface-to-volume ratios are very much larger. It follows that:- a) Positive ions adsorb strongly above the pH of the p.z.c. and negative ions adsorb below the p.z.c. b) Te larger the valance or size of the ion the more strongly it adsorbs. c) Metal ions which hydrolyse such as Fe3+, Zn2+, Cu2+, For example, for SiO2 where the negative charge 2+ Pb , etc. adsorb most strongly at a pH value where persists above pH 2 and is constant from pH 2 to 8, it hydrolysis of the metal ion begins. is found that zinc adsorbs from solution onto the silica Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 49

surface more strongly at pH 7 than at pH 4. At pH 7, centrations with SiO2 as the adsorbent, ions where Zn2+ is beginning to hydrolyse: φchem has a fnite numerical value adsorb in order: 2+ - + 2+ 2+ 2+ 2+ Zn(H2O)6 + OH ⇄ Zn(OH)(H2O)5 + H2O ... etc. Ca < Ba < Zn < Co = It appears that the strongly charged and un-hydro- Ni2+ < Cu2+ < Al3+ < Cr3+ << Fe3+ lysed ion is so heavily solvated that it cannot approach Te order of adsorption of ions and charged par- the interface. On hydrolysis, the average charge per ion ticles at given concentrations of electrolyte deter- decreases and desolvation upon adsorption is more read- mines the order in which species will be desorbed or ily achieved. exchanged if the electrolyte concentration is slowly Similar principles apply when small charged par- increased as it is at increasing depths in diagenetic ticles are adsorbed on surfaces of sediment substrates basin sediments. and the hydration or strong solvation of the more highly 6) Tere is a change in the adsorption characteristics charged surfaces of very small particles is quite signif- at pH values and concentration of species just below cant. It is believed that such highly solvated surfaces pre- that at which precipitation occurs throughout the vent or impede such particles closely approaching the bulk sol or solution. Species tend to ‘coat’ or prefer- surfaces on which they are adsorbed until desolvation entially precipitate on surfaces. Fe(OH)3 coating on is achieved. Tis is the reason why very small hydrated silica, smectite coating on illite, Si(OH)4 on quartz metal sulphide particles or hydroxy-carbonate particles grains, successive layers on botryoids, etc., appear to form “sof” framboids and oolites when these concre- be manifestations of this phenomena. tionary precipitates frst develop. 7) Polyelectrolytes, that is polymers where repeating Colloidal chemists have closely studied adsorp- units are ionised or ionisable in water, are strongly tion phenomena and many more details are available in surface active. Tey may show some preference for standard texts. Te main conclusions which are relevant adsorption on oppositely charged surfaces but more to the behaviour of silica particles migrating and re- usually they adsorb on any surface. Tis is due to depositing in basin sediments are:- entropy considerations in that one adsorbing poly- 1) Higher valent ions carry a higher charge, hydrolyse mer frees to solution many previously bound water at pH levels usually encountered in basin pore fuids, molecules to yield a net increase in the number of and adsorb more strongly on substrate surfaces. kinetic units in the system. 2) Adsorption is concentration dependent. Surfaces 8) Te simplest way of desorbing species, and particu- adsorb more charged particles or ions per unit area larly metal species, is to lower the concentration of from higher concentrations of sols and solutes. the surrounding soltion. Dilution simply washes 3) Difuse (Hemholtz) double layers thin at higher con- most surface adsorbed species of (eg. soap). centrations of electrolyte. Colloids readily coagulate 9) Swamping adsorbing surfaces by increasing the elec- because van der Waal’s attractive forces more easily trolyte concentration or concentration of competing exceed repulsion between like charges. Aggregation ions in the surrounding solution is equally efective and accretion are more readily achieved. in desorbing species. Increasing electrolyte concen- 4) Ions or charged particles can have a chemical or a trations within particle aggregates due to syneresis, ‘site’ afnity for certain surfaces or sites on a sur- or simply by the increase of salinities at depth dur- face. Tis so-called ‘specifc adsorption’ is expressed ing diagenesis, are important mechanisms for releas- mathematically for the free energy of adsorption ing ionic and charged particle species from basin ΔGads as: sediments. With metal ions the desorption by this ΔGads = Z+e(ψδ - φchem) exchange process as salinity steadily increases is in Where Z+ is the valency of the adsorbing ion approximately the order: adsorbing on a surface where the potential at a dis- Ag+ ~ Hg2+ > Pb2+ < Cu2+ > 2+ 2+ 2+ 3+ 3+ 3+ tance δ out from the surface is ψδ volts, e is the elec- Zn > Co > Ni > Al > Cr > Fe tric charge, and φchem is the adsorption potential of Te order of desorption for charged metal hydroxy- the ion concerned. If there is a strong chemical or sulphide species which determines the order of ‘site’ afnity ions can adsorb on like-charged surfac- their mobility in the outward seepage of pore fuids 2- es. For example, SO4 adsorbing on Al2O3 in water (the paragenetic sequence) has not been confrmed lowers the positive charge through zero to negative experimentally and is far less certain because of the 2- so that SO4 must then adsorb on surfaces of mod- complex mixture of substrates. Te adsorbed metal erate net negative charge. sulphide species also remain undefned but it is clear 5) From solutions at about pH 8 and equi-molar con- that the hydration is variable and that a complex 50 John Elliston

series of double sulphides is involved. Te original paragenetic sequence suggested by Lindgren7 and his contemporaries before particle - ion interactions and DLVO theory had been developed was based entirely on feld observations. Tey studied the order of deposition of sulphide minerals in veins and lodes. Lindgren7 refers to sulphides and says (1937, p. 361) that on the whole the order of deposition of the ore- metal sulphides seems to follow: Hg, Ag, Cu, Pb, Zn, Ni, Co, Fe Despite lead and cobalt being displaced when these lists are compared, Lindgren’s7 suggested order of sulphide deposition from observations many Figure 9. Generally larger particles form weaker gels. However, mixtures of large and small particles form very strong dense gels. years ago does seem to have much more than a co- Te infll ‘necking’ and silanol linkages are not so well developed incidental resemblance to the order which has now between larger particles. Smaller intergranular particles in the inter- been determined for base metals to be desorbed by stices act as strong ‘bridges’ linking the whole into a denser mass. increasing salinity. [From Iler2, 1979, p. 372].

AGGREGATION OF GLOBULAR POLYMERIC SILICA PARTICLES IN NATURAL SEDIMENTS

Liquefaction and fow or intrusion of natural sediments provides the shear to aggregate the sediment particles into close packed clusters as in Figure 8. Te extremely small size of the ultimate particles in this form of silica must be kept in mind. A small granule of typical silica gel about 1 mm in size contains about 50 x 1012 particles. Silica spheres forming ‘close packed’ accretions in re-liquefied natural sediments probably involve both large and small particles, but as Iler2 (1979, p. 372) points out, a strong dense gel is formed from a mixture of large and small particles (Figure 9). Te ‘active’ silica, Figure 10. An electron micrograph of opal structure that has been lightly etched with hydrofuoric acid shows internal rings within the Si(OH)4 molecules and the oligomeric silica (Figure the ordered silica spheres. Iler2 (1979, p. 401) records as many as 1), are much smaller again. Tese can pass through the fve internal layers. Te size of the particles comprising the layers is 50nm but the opaline spheres have to be over 300nm before the interplay of colours can be seen.

fnest membranes. It is important to recognise the wide size range of the colloidal silica spheres and the ability of the monomeric and oligomeric species to difuse through the fnest pore spaces. In ‘close packed’ and ordered structures of silica spheres such as opal (Figure 10), this ‘active’ silica can difuse in to fll the spaces between particles. Te opaline spheres themselves are in fact, built up from much smaller silica particles or spheres. Concentric rims round a central nucleus can be seen in the lightly etched Figure 8. Tis diagram shows the way natural polymeric silica glo- 2 bules closely pack together in the shear regime to reduce surface opaline spheres in Figure 10 but Iler (1979, p. 401) energy and form denser, physically stronger clusters that aggregate records as many as fve shell-like layers on cross sections to accretions. of opaline silica spheres which show their growth from Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 51

Figure 11. Silica accretions in porphyroid are usually about half to 1 cm in size but in some fuid muds they aggregate to much larger chert or agate nodules. Tese are called geodes. As they condense and loose water a peripheral region may loose water to form a much denser porous media. Similar to that in the opaline cavities, the polymeric silica particles settle into horizontal layers to form Uruguay banding as illustrated.

Figure 12. Sediment as initially focculated has its silica particles much fner particles like tree rings. Tese minute opa- ‘coating’ other sediment particles and interspersed as random lin- line spheres apparently build up like small concretions, king chains (From Elliston9, 2017, p.101). framboids, or oolites and with the aging of the gels the build-up continues to infll the ‘necks’ in the chains of silica spheres or the interstices in the ordered opaline- type ‘close packed’ structures. Opal is usually formed in irregular cavities in the soil profle where the ground water is alkaline. Smaller silica gel particles flling the cavities fnally dehydrate to form chert, potch or ‘milky’ opal. Te concretions of some gem quality opal appear to develop in dilute mobile gels. In some cavities the opaline spheres settle in horizontal layers to Uruguay Banding as Illustrated in Figure 11.

Te accretion of polymeric silica in mobile sediments

Te development of polymeric silica accretions in mobilised sediment is illustrated in Figures 12 to 14. Fig- ures 15 and 16 also show how impure silica accretions break up and divest impurities as they are kneaded when the sediment is further re-liquefed and involved in fow. Te development of siliceous accretions can be seen in many rock exposures and drill cores and in a wide variety of rock types. Unfortunately, many geologists remain unaware of current developments in colloid and Figure 13. Liquefaction or shear by pasty fow of the sediment surface chemistry and remain unable to correctly inter- allows some ‘close packing’ of silica particles into micro-accretions pret the porphyroid or small globular quartz accretions and orientation of the clusters and chains of particles to result in a schistose or “tufaceous-looking” greenstone texture as in Figure 18 or larger synerectic quartz ovoids that are commonly (From Ellistonr9, 2017, p.101). found in sediments that have been mobilised during diagenesis. Figure 17 is a typical example of a “quartz clot rock” of the accretion of very small polymeric silica globules from drill core at the Dalgaranga Prospect in Western aggregating to quartz ovoids was found in drilling the Australia. Some of the small quartz ovoids can be seen Olive Wood Prospect at Tennant Creek NT. Tis is illus- to be aggregates of smaller pieces. A very good example trated in Figure 18. 52 John Elliston

Figure 16. Accretions comprising only silica spheres can form dense ‘close packed’ aggregates. In this confguration each silica sphere Figure 14. With continued laminar fow the micro-aggregates of is densely packed in contact with twelve others such that the accre- silica spheres combine to form larger macro accretions that survive tion has sufcient internal cohesion to survive in the shear regime 9 in the shear regime in accordance with their internal cohesion. Tis (From Elliston , 2017, p.101). is dependent on their ‘purity’, the degree to which they have been “kneaded” (broken and re-formed) in the laminar fow (From Elli- ston9, 2017, p.101).

Figure 17. Tis shaly porphyroid intersected in drilling at the Dal- garanga Prospect west of Mt Morgan in Western Australia contains small ovoids of quartz. Some of these can be seen to be aggregates of small ovoids or fragments of the pre crystalline polymeric silica. Tis is clear evidence of their accretionary origin (From Elliston, Figure 15. Accretions of silica spheres containing impurities tend to 2017, p.129). break up in the shear regime during laminar fow (From Elliston9, 2017, p.101). Shear by slight lateral movement in a highly siliceous sediment intersected by drilling at the Wagga Siliceous accretions develop in various types of sedi- Tank prospect near Mt Hope in NSW has developed the ment containing polymeric silica gel particles that have shreds and small round accretions of khaki chert illus- been liquefed or sheared by pasty fow. Mobilised beds trated in Figure 19. Similar shear in a jasper bed inter- in the Western Australian banded iron formations form sected in the Explorer 7 prospect drilling at Tennant accretions of gelatinous ferric hydroxide (crystallises to Creek, NT, has developed small white quartz accretions. haematite ovoids) and denser silica gel that crystallises Tese shown in Figure 20, have been ‘kneaded’ or rolled to white quartz ovoids. Accretionary chert ovoids occur to a greater extent to divest the larger ferric hydroxide in some of the porphyroids at Tennant Creek and those particles that form the bright red pigment in jasper. Very at the Black Angel Prospect initiated research into the small clay particles or muscovite fakes are the usual col- longstanding problem of porphyroid and granite genesis. ouring impurity in khaki chert but the very common Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 53

Figure 20. Formerly fuid (thixotropic) jasper band were found in the lode section when drilling the Explorer 7 prospect at Tennant Creek, NT. During fow the fuid jasper gel developed denser accretions of paler ‘blue’ (leucoxene tinged) and white quartz (From Elli- 9 Figure 18. Te accretionary origin of the small quartz ovoids in the ston , 2017, p.159). Olive Wood re-liquefed sediment lens at Tennant Creek, NT, is clearly indicated by the many examples of the aggregation of sub-units of former polymeric silica. Te enlargement shows this aggregation but most small accretions in this rock have crystallised to small round globules of quartz (From Elliston, 2017, p. 153).

Figure 19. Slight lateral movement or shear in former siliceous ooze from the Wagga Tank drilling Mt Hope, NSW, has developed an obvious ‘porphyroidal’ texture. Te elongation, distortion, and Figure 21. Tese siliceous accretions were formed experimentally aggregation of the ovoids indicate deformation and plasticity occur- in a gel paste and were also produced in a number of other col- red at the sof stage in consolidation of the ooze. Wisps and shreds loidal materials. Te formation of accretions requires thick pastes of cherty sediment are similarly deformed. Some quite large (1cm) and achieving steady rates of uniform shear throughout concen- accretionary nodules or ‘gel balls’ have developed (From Elliston, trated thixotropic gelatinous materials that are normally gelatinous 2017, p.311). solids. Tis imposed some limitations to the laboratory work [From Healy11, 1973 Accretion of colloidal material, p.9, fg 2(c)]. and abundant white quartz in accretions and veins is usually a dispersion of exceedingly small inclusions of so-called phenocrysts, irregular fuidal contacts with water or even residual hydroxyl groups trapped in the feldspar and central shrinkage cavities flled with chlo- crystal lattice. rite characterise the siliceous accretions that occur in Before measurement of interparticle forces with an the porphyroids at Tennant Creek, NT. From these and adapted atomic force microscope had confrmed DLVO similar features in the feldspar ovoids it became quite theory in 1998, Professor T. W. Healy at the Particulate clear that these ovoids had hardened or crystallised from Fluids Processing Centre, University of Melbourne, in common components of ordinary sediments. Tey were 1973 was commissioned to demonstrate the formation of accretions of clays, hydrous ferromagnesian mineral and accretions experimentally in a range of colloidal pastes polymeric silica particles. Te rather irregular interbed- and slurries. His denser silica gel accretions, similar to ded and formerly fuid layers and lenses of porphyroid those in mobilised siliceous shale (Figure 17) or jasper had been mudfows. Te Caroline mudfow conglomer- (Figure 20 are illustrated in Figure 21. ate, Mammoth intrusive sandstone and the True Blue Tubular syneresis crack patterns, plastic distor- and the Plum Mine slip complexes outcropping in the tion, aggregation of sub units, chert and jasper in the turbidite sequences at Tennant Creek contained clay and 54 John Elliston

Figure 22. Tis illustrates the characteristic features of siliceous accretions in the Tennant Creek porphyroids. Tey include tubular synerectic crack patterns, plastic distortion, aggregation of sub units, chert, jasper, white and ’blue’ quartz rolled together, irregular fuidal contacts with feldspar and central shrinkage cavities flled with chlorite. Tis evidence for a non-molten or volcanic origin is conclusive (From Elli- ston10, 2014, p. 69). quartz accretions like those in the porphyroids. A selec- of the charge sites on a crystal surface. Tese charge sites tion of quartz accretions from the Tennant Creek por- are arranged in ‘steps’ and Figure 23 (from Stumm5, phyroids is illustrated in Figure 22. 1992) is a highly simplifed model of a surface to which lattice units are added as cubes. It indicates the diferent activation energies that would pertain at the fve possi- THE CRYSTALLISATION OF QUARTZ FROM ble surface sites. Sites 1 and 2 are most reactive and face MONOMERIC, OLIGOMERIC AND GLOBULAR site 5 is least reactive. Since precipitation/solution will POLYMERIC DISPERSIONS occur preferentially at sites 1 and 2, the surface confguration is continuously converted towards 3 and 4 giving Quartz is virtually insoluble and quartz crystals a ‘stepped’ growth to the crystal face. do not grow directly from solution. Tey grow through Crystallographically oriented striations on the faces a disordered solvated surface layer (coating of poly- of natural quartz crystals refect this “stepped growth” meric silica particles) that are replenished by difusion of the respective lattices. Tis is commonly observed on from the surrounding fuid as polymeric silica parti- large hexagonal quartz crystal faces but it is also respon- cles adsorbed at the surface charge sites loose water and sible for “hopper structures”, internal growth layers, com- become part of the crystal lattice. pound crystals, negative and pyramidal forms, etc. in a In the geological context equilibrium between the variety of natural crystals like pyrite, galena, quartz, etc., monomeric, oligomeric (Figures 1 and 2) and polymeric Details of textures observed in the Waitara Sinter on silica globules is maintained according to the pH and the Coramandel Peninsula, New Zealand, are illustrated concentration of other salts present in the surrounding in Figures 24 and 25. Tey are from a student honours solution as shown in Figure 5. Figure 3 is a TEM image Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 55

Figure 24. Tis euhedral crystal of clear glassy quartz was formed in a cavity in the Waitara Sinter, Coramandel Peninsula, New Zealand. It is coated with “little balls” (about 70 to 80 nm) of amorphous polymeric silica. Te silica globules are irregular in size but they tend to be arranged in ‘chains’ or lines along the crystal Figure 23. Tis is a diagram of a simplifed square crystal lattice striations and they form a crude linear pattern on the crystal sur- model. It shows the fve possible types of surface exposure for the face in accord with the atomic geometry of the crystal lattice (from unit cells in such a lattice. Te activation energies for each site dif- Newton12, 2000). fers as the units are bonded to 1, 2, 3, 4, and 5 surfaces respectively. Sites 1 and 2 are most reactive and site 5 is least reactive. Since precipitation or solution will occur preferentially at sites 1 and 2, the surface confguration tends to be continuously converted towards tal striations and they form a crude pattern on the crys- site 3 giving a ‘stepped’ growth to the crystal face. tal surface. Te atomic geometry of the crystal lattice has resulted in a pattern of electric charges distributed on the crystal surface to which the much larger charged thesis by Z. J. Newton12, University of Auckland, in May particles coagulating from the sol both as short chains of 2000. Euhedral quartz crystals are observed project- globules (focs) and as individual charged particles have ing into vughs, miarolitic cavities, central cavities in been attracted. Te globular polymeric silica particles geodes, and into a cavity in the Waitara Sinter illustrated can be directly observed to be coating the crystal sur- in Figure 24. Water is a product of the silanol-siloxane face and the charge they carry can be inferred from the condensation by which natural polymeric silica crys- crude pattern in which the charged particles have linked tallises. Water pockets, fuid inclusions or cavities that to each other in ‘chains’ and to the crystal surface. are still or were formerly flled with fuid are therefore Figure 25 at 10 times greater magnifcation shows a found in most occurrences of natural quartz. Such cavi- small part of one of the non-striated faces of the quartz ties are mainly due to the syneresis of the surrounding crystal in the Waitara Sinter. Without the alignment of polymeric silica but while the particulate gel remains ‘steps’ the polymeric silica globules have been depos- porous, additional Si(OH)4 difuses inward as the con- ited in more random patterns on the surface and this densation reactions proceed. Te syneresis cracks and enlargement also illustrates the wide variety in size of various forms of shrinkage or fuid cavities can be flled the “little balls” of polymeric silica. Colloidal particles with ‘secondary silica’. range in size over 4 orders of magnitude from molecular Te euhedral crystal of clear glassy quartz illustrat- dispersions (2 to 7Å) to about 1,500 nanometres. ed in Figure 24 has grown in a fuid cavity in the Wait- ara Sinter. “Little balls” (about 70 to 80 nm) of amorphous polymeric silica have been deposited from the CRYSTALS GROW LIKE CONCRETIONS IN last stages of dilution in the surrounding sol. Te actual GELATINOUS MEDIA silica globules that can be observed in the SEM image form an incomplete coating on the crystal surfaces. Te Surface catalysis or the role of the surface interface polymeric silica globules are irregular in size but they in diferent gel structures does not yet appear to be ful- tend to be arranged in ‘chains’ or lines along the crys- ly established. It seems that the enhancement of crystal growth operates in a very similar manner to the way 56 John Elliston

6) Particles that difuse to the growing crystal face are charged hydrated molecular or macromolecular species. 7) On surfaces the diffusion rates of particles are increased (Henisch13, 1972, p. 56). 8) Adsorbed ions and surface hydration water on particulate species joining a crystal lattice must be released in cases where water or these ions cannot become part of the lattice. 9) Te higher concentrations of nascent water and ions liberated at the surface may be responsible for the increase in surface difusion rates and destabilisa- tion of the dispersed particulate species. 10) Gelatinous media act as ‘molecular sieves’ where they efect chromatographic separation and band- Figure 25. At 10 times greater magnifcation the polymeric silica ing of particulate species (as in agates, geodes, etc. globules on one of the non-striated faces of the quartz crystal in - see Figure 26). Most large geodes develop cavities the Waitara Sinter are seen to have been deposited in more random where crystals grow from the internal surface. Since pattern. Without the alignment of ‘steps’ the polymeric silica globules (“little balls”) have been deposited in a wide variety in sizes the structure of the ‘coating gel’ in which enhanced (from Newton12, 2000). crystal growth is occurring has a specifc orientation in relation to the crystal face, it suggests that particles with similar non-homogeneous charge disin which concretions form, that is by entrapment of the tribution (like those dictating the internal ordering mobile particulate species at the surface layer where a in framboids or the radial orientation of chlorite higher concentration of electrolyte is maintained by des- or chamosite layers round oolites) are controlled to orption from the particles joining the nucleus. In the approach the surface in a certain way. case of concretion, oolite, or framboid formation, the particles are presumably larger and retain their hydration. Concretionary nuclei are synerectic, semi-ordered, and ‘close packed’ denser gels. Tey exude electrolyte to the surface as Helmholtz double layers impinge due to van der Waal’s interparticle attraction. The enhanced crystal growth in gels or surface catalysis by a ‘gel coating’ appears to involve the following steps: 1) Crystallite formation by coalescence of molecular species. 2) Crystals nucleate by coalescence of crystalline molecular clusters of less than unit cell size. 3) Enhanced crystal growth (surface catalysis) occurs on crystal faces having a certain orientation in relation to the long-range ordering in the coating gel. 4) Solubility in the vicinity of surfaces is greatly reduced and colloids in particular tend to “plate out” on surfaces. Tis process of “difusophoresis” is also used in surface coating technologies where colloidal particles are deposited onto substrates like plastics or roofng materials. Figure 26. Te fne concentric bands of ferric hydroxide stained chert, the growth of coarse crystals into the crystal cavities in the 5) Species added to the surface of a crystal growing centre and the fuid ‘out-burst’ channel in this geode are positive in a gel move to the liquid-solid interface by difu- evidence for the syneresis of its precursor silica gel. Tis clear empi- sion down a concentration gradient created by their rical evidence positively supports theoretical work that shows that removal from dispersion. synerectic aging of a uniformly mixed gel must result in separation of component particles into Liesegang type concretionary bands. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 57

sition at around 300°C, a geothermal gradient of 35°C/ km, and a pressure gradient of 100 atm/km. It suggests an unrealistic 5000 cub/m of solution would be required to deposit each tonne of quartz. If temperatures indicated by the fuid inclusion data are assumed, the listed solubilities generally result in a requirement of some 4000 to 5500 cub-m of solution per tonne of quartz deposited. Where would this volume of water come from? Sediments generally are deposited with average total water content around 75% by weight. In the consolidation of sediment to a rock some 3.5 to 4 cub.m of pore water, surface adsorbed water, and water from chemical dehydration reactions must be released. Each cubic Figure 27. Long slender ‘needles’ of rutile develop in the gelatinous meter of fnally compressed and anhydrous rock (weigh- polymeric precursor of a large single crystal of vein quartz. Growth ing 2.7 tonnes) could be expected to express silica dur- on the small newly nucleated ‘seed crystals’ of rutile in their supporting silica gel medium is enhanced in accord with the orien- ing the stage where it could be dissolved and mobilised. tation of the crystalline surface with the ‘structure’ or meshwork It is suggested that water outfow of this magni- pattern within the gel. Tis surface catalysis can therefore be con- tude could be sufcient to accumulate and deposit the fned to one face of the ‘seed crystal’ that grows taking all available observed volumes of vein quartz. However due regard dispersed rutile to the exclusion of other faces. must be given to all classes of solutes of quartz, their mutual equilibrium, and the permeability, thixotropy, rheopexy, syneresis, and enhanced crystal growth prop- 11) If the difusion of particulate species through the erties of the precursor vein gels. surrounding gel and transfer to its internal surface Quartz vein development is related to the water out- against the growing crystal resulted in their orienta- fow system, crosscutting the sediments initially by clas- tion (as it does in certain micelles and concretionary tic diking and hydraulic fracture. Te crosscutting frac- structures) such a favourable orientation in relation tures start flling frst with water, dilute sediment slurry, to the crystallographic axis and the surface ‘growth or watery silica gel, as earthquake shocks or disturbance steps’ may signifcantly increase the rate of growth. of the whole pile initiates “outbreaks” from over-pres- 14 15 Hatschek and Simon (1912), Boydell (1925), sured entrapped fuids. Te crosscutting nature of earlier 16 13 Healy (1969), Henisch (1970) and many others have veins is shown in Figure 28. All these “outbreak” or vein demonstrated enhancement of crystal growth in gels injection phases is followed by a “static” period in which experimentally. It is developed dramatically in the spec- pore fuid seeps through the fabric of the crosscutting, tacular needle-like crystal growth of rutile in the precur- more watery vein gel, introducing new silica both by sor silica gel of vein quartz (Figure 27) and similar acic- the fuid seepage and by difusion (“sweating out of the ular interlocking needles of tourmaline are occasionally rocks”). also developed in this type of quartz. Te polymeric silicic acids of the vein gel remove the monomer by polymerisation. Tis creates a difusion gradient in respect of this species. Like the replacement THE FLUID TRANSPORT OF SILICA INTO VEINS reaction in petrifying wood, monomeric Si(OH)4 dispersed in sediment pore fuids difuses into the precur- Calculations of water volumes required to transport sor vein gels during the extended times involved. Exam- the tonnage of quartz observed in various vein systems ples of silicifed wood plainly demonstrate that substan- and lodes have tended to discount the oligomeric and tial volumes of silica can be transferred and localised by polymeric silicic acids (colloidal silica) with which the cumulative difusion. Te polymerisation of the mono- monomer and ionic acids must be in equilibrium. mer depletes this species within the vein and “draws” It is quite apparent that calculations based on meas- in more silica, possibly as much by difusion as by seep- ured solubilities of silica alone give unrealistically large age of solution. Denser and “purer” precursor vein gels volumes of solution. Tis leads to the postulation of re- are then involved in each successive mobilisation. Later circulation or additional quantities of water from deep injections are of dense silica gel slurry capable of sup- mantle sources. Unrealistically large volumes of solution porting wall fragments and other minerals as in Figure would be required to deposit the observed large volumes 29. Tey involve relatively small volumes of water. of vein quartz (Barnes3, 1967, p 392). He assumes depo- 58 John Elliston

Figure 28. Several types of early diagenetic crosscutting sediment veins ranging from irregular quartz veinlets ‘splashed’ within the ooze, ‘dirty early quartz’ veinlets, and later quartz-chlorite veins re- injected beside original veins, illustrate stages in “the birth of quartz veins” in this sof diagenetic siliceous ooze intersected in the Wagga Figure 29. A discordant quartz vein in basic rock carries pieces and Tank prospect drilling in Mt Hope district, NSW. slabs of wallrock “suspended” in the quartz. Some crystal cavities or vugs are developed near the centre of the vein. If the quartz in the vein were deposited from solution it introduces the problem of suspending the fragments while sufcient quartz is precipitated to Requirements for vein emplacement support them. Injection of the vein material as thixotropic silica gel is clearly indicated. Rheopexy (instant resetting as gelatinous fuids Veins and veinlets ranging in all sizes from minute approach cessation of fow) would support fragments and preserve micro-veinlets to large ‘quartz blows’ constitute the most their delicate projections. Afer injection and resumption of a gela- common of all mineral deposits. Smaller veins persisting tinous condition the crystal cavities indicate synerectic shrinkage for relatively shorter distances through the rocks usually during dewatering and fnally crystallisation. Te vein outcrops on consist only of quartz but many quartz veins contain a the south coast of N.S.W. between Bungen and Goalen. variety of other minerals such as sulphides, oxides, fel- spars, chlorite, carbonates etc. Te veins appear to be emplaced by the injection of entrapped fuid and com- er mineral masses, it must have some substance or ponents from the sediments that these fuids carry with ‘body’ and be able to rapidly re-solidify. them. Fluids escaping upwards through semi-consolidat- 4) Te material of the vein must be some component or ed sediment piles re-liquefy and mobilise sediment, indi- mixture of components of the local materials from vidual sediment components, or previously re-mobilised which it originates. sediments. Tese then rise to higher levels within the 5) Te host materials injected by veins must be plastic, pile. Such vein injections are usually episodic and trig- both plastic and cohesive, or cohesive but readily frac- gered when de-watering basins are disturbed or shaken tural so that veins can hydraulically penetrate them. by earthquake shocks. 6) Veins that show evidence of multiple intrusion or To be injected through semi-consolidated sedimen- several stages of liquefaction and re-intrusion must tary materials during diagenesis and metamorphism, be capable of isothermal re-liquefaction by repeated certain essential requirements for the emplacement of mechanical disturbance such as fault movements, any vein must apply. Tese are: earthquake shocks, etc. 1) Entrapment of a fuid or a readily liquefed semi- 7) Te mineral or minerals comprising the vein must solid at hydraulic pressure that is supporting part of have simple mobile precursors which can crystallise the weight of overlying materials. When disturbed it directly (with only water loss) to the crystals or crys- must be able to fow to efect equalisation of this load. talline assemblage of vein minerals. 2) If the vein is to intrude upward, its fuid-rich mix- Veins are therefore logically part of the diagenetic ture of components must be lighter than any plastic ‘plumbing system’, the means by which fuids leave con- material that it invades or confned within a crack solidating sediment piles, large re-mobilised sediment or opening in a cohesive fractural solid. masses, or the accumulated precursor mineral gels and 3) If the vein material is to preserve the shape of its sludge which constitute newly deposited lodes and ore- injection and suspend wall rock fragments or oth- bodies. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 59

Figure 30. A ptygmatic mud dike is injected into hematite shale at the sof sediment stage. Te highly fuid nature of mud between haematite layers is indicated by the irregular folding in its bed- Figure 32. Laminose fow or lateral slippage along overpressured ding but the dike which has broken across the sof bands is highly fuid-rich bedding planes during diagenesis has allowed some of the contorted into a ptygmatic pattern due to the rheology of the non- hematite shale bands to fold and disrupt. Te suspending ‘fuid’ was Newtonian fuid paste. (From Explorer 8 core, Tennant Creek, NT). probably flled initially with dilute polymeric silica as a “dirty early quartz vein” and then ‘inflled with additional Si(OH)4 to crystallise as whiter quartz. (From Gecko Mine drill core, Tennant Creek, NT.).

Figure 33. A ptygmatic-type clastic mud dyke or “dirty early quartz vein” was intruded into diagenetic sof and banded hematite shales. Tese early “out wellings” of entrapped fuid through fne oozy sediment are episodic with each successive reactivation becoming more quartz rich until the flling material resembles vein quartz. (From Explorer 8 drill core, Tennant Creek, NT.).

ica gels escaping from normally consolidating sediment piles. Tey are very much a part of the normal “sediment plumbing system” augmenting the slow seepage and diffusion of the fuids uniformly upward through the sediment layers as a whole. Figure 31. A flow banded mass flow unit, one of the Tasma- Once established, crosscutting veins and dykes then nian west coast porphyroids outcrops on Whip Spur above South become channel-ways by which more fuid seeps round Queenstown. A clastic dyke or fluid outflow channel intruded or through their less permeable host sediments. Because shortly afer its deposition over other wet sediments. Tis dyke has of this continuing fuid seepage, veins are ofen rejuve- a ‘sandy’ matrix similar to that of its porphyroid host but it carri- nated by subsequent movements, efect extensive altera- es rotated pieces of disrupted siliceous fow bands and a ‘jumble’ of these is concentrated in the upper part of the dyke. tion of adjoining sediments, or act as feeder channels to larger ‘alteration zones’ or overlying mineral deposits. Tey are ofen concurrent with, or occupy faults and Veins emerge from partly consolidated sediments shear zones which act in the same way to channel fuids at all stages of diagenesis. Some examples of early veins and hydrothermal fuids draining from depth. through newly deposited sof sediments and mass fow units are illustrated in Figures 30, 31, 32, and 33. Veins and sediment dykes could be regarded as the “muddy water” or the more fuid pastes and watery sil- 60 John Elliston

EVIDENCE FOR THE FORMATION OF QUARTZ VEINS autoclave or an “aqueo-igneous magma”, iron silicates FROM SILICA GELS would be formed. As observed in many highly aqueous environments, Quartz veins formed at low to moderate temperatures very abundant hydrous iron silicates, iron magnesium and calcium silicates, ferric and ferrous alumino-sil- Tere are many examples of injected, intruded, and icates, iron hydroxides, etc. are ofen associated with “remobilised” type vein quartz in close association with sedimentary ‘chemical sediments’ like banded iron for- granites, greenstones, andesites, metadolerites, pyrox- mations. But these under-saturated ferromagnesian- enites, amphibolites, metabasalts, spilites, and even ser- rich hydrates and iron hydroxides are stable in the deep pentines and other ultrabasic rocks. Tis has suggested ocean and aqueous environments because they are com- an origin related to high temperature. Melting has in pletely hydrolysed (fully reacted with water). fact been put forward in the past as the mechanism for In most cases the concept of minerals such as quartz mobilising the quartz in some of these settings. Howev- and magnetite having been deposited contemporaneous- er, the clear evidence is that a single and usually relatively from hydrothermal solutions is not possible. Silicates ly pure mineral is involved, a high temperature is neces- or silicic acid in solution together with ferrous or ferric sary to melt quartz itself, quartz veins lack evidence of ions should deposit iron silicates from solution rather any extreme temperatures, and molten silica is highly than the respective oxides. reactive with such things as undersaturated basic rocks, High temperature hydrothermal or “aqueo-igne- magnetite, haematite, calcite, etc. ous” quartz veins in serpentines, andesites, metabasalts Te data from homogenisation of fuid inclusions and dolerites, pyroxenites, amphibolites, etc. must also in vein quartz repeatedly indicate low to moderate tem- be somehow constrained to stop what would be their peratures usually less than half that assumed for mag- expected reactions with these basic and under-saturated ma melting. Tis has even led committed magmatists rocks. It is most noticeable that in a true basalt feld the to abandon the idea of molten silica injection to form molten rocks are fully reacted with all their available quartz veins. quartz. Iron (olivine) or ferromagnesian silicates pre- For an exploration geologist, observing quartz veins dominate and there is not a quartz vein in sight! in the feld, in drill cores, in granites, in greenstones To ft the case of quartz veins in ordinary sediments, (even ultrabasic rocks), in andesites, in sediments, in limestone, and marbles there is clearly a need to suggest metamorphics, and as massive vein systems in faults a low temperature form of mobile quartz. Quartz fuidity and shear zones is so commonplace, and ofen appears due to simple fusion of dry quartz is no longer seriously to be of so little relevance to the discovery of major ore proposed, even in supposedly igneous rocks. Terefore, deposits, that very seldom is any consideration given to in all cases, vein quartz, quartz mobility, and vein quartz their origin and genesis. In fact, afer having worked for crystallisation are envisaged as being from some aqueous so many years in older sedimentary, metamorphic, and phase such as a solution or hydrothermal solution, from granitic rocks, the usual habitat of mineral deposits, it silica gel, or from “magmatic quartz” which is thought- takes something like a visit to the mountains of lava and lessly considered some type of higher temperature low- volcanic debris such as comprise the islands of Hawaii, water eutectic. to realise how impressive and unusual are vast rock exposures totally devoid of quartz veins! Actual inclusions of calcite, dolomite, haematite, The entrapment of organic materials and salt in fluid magnetite, rhodochrosite, siderite, chlorite, dawsonite, inclusions in quartz veins etc. which are ofen found in quartz veins preclude the possibility of molten fuidity of the silica. For example, Te mechanism by which either primary or second- the ferric hydroxide and polymeric silica precursors ary inclusions of organic liquids and salt could become of jasper do not form iron silicates when intimately trapped in minerals such as quartz has always been a mixed together because these substances are already problem (Kvenvolden and Roedder17, 1971, p 1224). Te fully reacted with water. Te hydroxyl groups satu- problem arises entirely from the assumption that an ion- rate and ‘blind’ what would be the normal reaction ic solution is the only medium from which quartz, such of these ions to form iron silicates. Mixtures of the as the vein quartz in which the inclusions are observed, gelatinous forms of these materials are stable as mani- crystallises. fest by banded iron formations and bedded jaspers. If Kvenvolden and Roedder17 (1971) review various these same materials were heated so that there is dis- studies of organic fuid inclusions in vein quartz from sociation of the iron and silica hydroxides, in say an sediments and from vein quartz in granites. Tese con- Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 61 tain “vaselinous oil”, possibly olefnic diferentiates as ers and Ottemann20 (1952) describe how in the newly dark resinous fakes, and black globules of aromatic dif- laid out Moll Tunnel which supplies the Stau Lakes of ferentiate. Tere is no essential diference between the the Kapruner Tales with water from the Moll River, inclusions in granite quartz veins as opposed to those in Horninger mapped a zone about 1400m from the tun- veins in sedimentary and metamorphic rocks. nel mouth which was rather wet and in sof material From the fuid inclusions themselves, Kvenvolden (1261.6m above sea level). In this zone an oily white and Roedder17 determined homogenisation temperatures mass of silica gel was found in a loamy strongly fow- from 70 to 160°C with most in the range 110-150°C. ing fault. Te water had a temperature of 2.7°C and a Te entrapped material was in most cases a moderately pH of 7.5. Te main mass of the silica gel was a very saline aqueous liquid but organic liquids include meth- fine-grained deposit weakly broken into two tabu- ane, ethane, butane, n-alkanes and isoprenoid hydrocar- lar masses and spindle shaped white to yellow forma- bons, and bituminous substances. tions. Some small crystals of calcite and glass-clear or Te molecular composition and distribution of the faintly milky opaque columnar prismatic crystals of hydrocarbons suggest biological precursors for these quartz were found in parts of the gel mass. Te quartz components. was either well-formed single crystals or small radiat- Te problems arising from the studies are: ing clusters of quartz prisms (up to 2mm in length x (a) the low temperature indicated by the homogenisa- 0.5mm diameter). Many of these were found entirely tion of the inclusions, which apply equally to quartz suspended within the gel from which they had been veins in sediments and in granites; formed. (b) the biogenic origin of the amines and hydrocarbons, In Australia Levings21 (1912) noted gelatinous silica when these were found included in granite veins in veins at the Great Australia Mine at Cloncurry and (Petersilie and Sorensen18, 1970, p. 59); the Victorian Department of Mines issued a report (K. (c) the means by which the inclusions could be Bowen22, 20th February 1967) on the “mutton fat” which entrapped in the quartz crystals when these were is the gelatinous silica containing aluminium hydroxide assumed to have crystallised from solution. found in the quartz veins of the Woods Point district. Tese problems do not exist for quartz veins having Te “mutton fat” sections of the quartz veins in several a precursor of gelatinous silica. Since such precursor sil- of the mines in the Woods Point area are associated with ica gel accumulates in, and is associated with fuid out- quartz “foors” and fssures in the dykes and are regard- fow channels, it is logical that some hydrocarbons and ed as a good indication of gold. saline pore fuid may become mixed with or included Analysis of the “mutton fat” solids (the material is in it. Tis could occur particularly when such precur- 73-75% water) indicate about 53% SiO2 and 27% Al2O3 sor gels may be subjected to periodic re-liquefaction. and although the jelly like material contains small crys- Entrapped organic liquid or inclusions of saline fuid tals of calcite and quartz it is apparently the dispersion within vein silica gel are preserved as entities while the of aluminium hydroxide which has prevented complete gel further densifes and is “inflled” by difusion into crystallisation to quartz. Bowen’s22 report indicates that it of additional monomeric and short chain oligomeric the small crystals of quartz have crystallised in situ silicic acids. Tese tend to maintain concentrations as within and from the “mutton fat” silica gel and that a exothermic crystal growth depletes and removes the dried-out sample preserved in the Mines Department mobile species with release of water. Museum (Reg. No. 15188) had formed a mass of small crystals of calcite and quartz embedded in a transparent flm like material. Tis suggested that the crystallisation Silica gel directly observed in veins occurs on drying out and that the original material in its natural condition in the mine may have been entirely One of the earliest excavations in a younger mountain gelatinous. fold belt, the driving of the Simplon tunnel through the It is clear that the quartz veins merging to silica Italian-Swiss Alps in the 1890’s, encountered gelatinous 19 gel or with parts of them still preserved in a gelatinous silica in some of the quartz veins. Spezia (1899) records condition are not found very frequently. However, these how in driving this tunnel, over record lengths for that examples demonstrate that certain quartz veins can and time and deeper beneath the surface above than ever do have this silica gel precursor and that quartz can before, “steamy conditions” were encountered and a vein crystallise within and from such gelatinous precursor of silicic acid in the gelatinous condition was discovered! material. A number of similar quartz veins containing gelatinous silica have been discovered since. Hellm- 62 John Elliston

Figure 34. Highly fuid or “poured in” vein quartz or a disjoin- ted plastically contorted fragment of vein quartz occurs in a large Figure 35. Discordant, ruptured, and re-healed fold patterns occur mineralised lode intersection. Tis core is from Woodlawn Mine, in this multi layered quartz vein from the Laurentian Shield, near NSW, and a number of almost ptygmatic quartz veins and frag- Sudbury in Canada. Tis plastic deformation is consistent with a ments of veins of this type were intersected in this lode. polymeric amorphous silica precursor in the vein before crystallisation, but it would remain difcult to explain if it is contended that the vein formation was by “crystallisation from hydrothermal solu- Plastic quartz in veins tion” and then subsequent folding of the crystalline quartz. In this type of random discordant folding, analysis of the stress felds that Tere are many examples of plastic quartz veins or cause the folding, shows it to be isotropic or fuid deformation. Any bulbous ptygmoidal plastic folding of vein quartz. Tis deformation pattern in a solid phase must be anisotropic. type of folding clearly refects the plasticity of the precursor silica gel of the vein prior to densifcation and crystallisation of the quartz. In the Woodlawn drill core, Figure 34 illustrates some highly plastic quartz that appears to have been ‘poured in’ among the sof precursor lode minerals in a highly mobile condition. It is certainly not intruded along a crack or some sort of fracture and it is quite obvious that this vein quartz displaced its pliable and yielding host at the time that it was intruded. Figure 35 is an example of a “wrigglyite” or plastically contorted multiple quartz vein mass that was deformed while the vein material was in the precursor gel stage. Figure 36 from a quartz-magnetite chlorite rock Figure 36. A quartz-magnetite-chlorite rock from the lower part of contains highly mobile irregular ptygmatic quartz vein- the Peko mineralised complex at Tennant Creek, NT, has concretionary rims developed round siliceous accretionary nuclei. Tese lets. Tese appear to have been irregularly ‘squirted in’ successive concretionary overgrowths on synerectic nuclei unmista- so that they became somewhat disrupted and discon- kably indicate the colloidal environment. Later ptygmatic and ‘clea- tinuous in the semi-fuid precursors of their chlorite ner’ quartz veins have sinuously intruded the sof precursor quartz- and magnetite host that also contains rimmed siliceous chlorite-magnetite. synerectic nuclei (concretions).

of other associated lode minerals of similar origin to Vein quartz as a breccia matrix the quartz itself (Figure 37), such that they appear as an injected or intruded breccia (Figures 29, 38, and 39). Many quartz veins, particularly discordant types, In many instances the quartz veins merely bifur- suggest a type of hydraulic fracturing, a movement in cate and anastomose round slabs and pieces of wall or bulk of the quartz material itself and its ability not only host rock such that these slabs become isolated in the to fragment the wall rocks, but to entrain pieces of wall vein material. Tese observations are inconsistent with a rock within the injected quartz. Quartz veins therefore hydrothermal solution theory of quartz vein genesis. As frequently contain angular fragments of wall rock, or calculations of the volumes of water necessary to trans- Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 63

port quartz in solution indicate, solutions of silicic acid are very dilute even at optimum conditions favouring solubility. If the resultant quartz vein had been originally injected as a dilute solution, what is to suspend the wall rock slab or the breccia pieces while further volumes of solution arrive to introduce the amount of quartz necessary to “cement up” the entire vein to the point where it could support the fragments? Figure 37. An intrusive siliceous breccia of grey chert contai- Te observations of angular fragments in quartz ning angular and partly digested fragments of chloritic wall-rock vein breccias include basic rocks of density 3.5 (as in was intersected in drilling near the margins of the gold orebody Figure 29) and even angular fragments of minerals like at the Peak Mine in Cobar, NSW. Tese earlier grey cherty ‘elvans’ magnetite (density 5.0 - as in Figures 38 and 39) which or extensive silicifed zones are ofen found intruded round larger would certainly sink in any opening or hydraulic frac- mineral lodes. ture flled with a dilute solution. Te habit of many quartz veins has been described aptly as “discordant squirts”, a high degree of fuidity involving a very mobile material is indicated by such complex injection structures Figures 40 and 41. How could such a mobile precursor to vein quartz maintain heavy wall rock fragments suspended in it, if it were a solution? Te phenomena of repeated injection or liquefaction, high fuidity and mobility during actual injection, yet “instant refreezing” once in place to “set” the breccia and wall rock fragments suspended within the material, is a clear indication of a silica gel precursor of the vein material. Only a non-Newtonian fuid, that is a material Figure 38. Syneresis of the gelatinous lepidocrocite precursor in the having a Bingham yield point at which it reverts to a fu- massive magnetite ore envelope at Warrego Mine, Tennant Creek, has developed a strong pattern of shrinkage cracks that have been id isothermally in response to shock or shear (thixotro- flled with quartz. Tere are several episodes of mobility in the sili- py) could display this phenomenon. Correspondingly, in ceous matrix that has suspended the heavy magnetite fragments by order to account for the features commonly observed in its rapid rheopectic re-setting. quartz veins, it must also revert suddenly to a gel condition (visco-elastic solid) as fow declines to a critical rate when the intrusive fow nears completion (rheopexy). At the time of injection, the silica gels that are the precursors of vein quartz must have a somewhat similar density to the wet sedimentary materials (like those in the Moll tunnel) that they invade. Tis means that any wall fragments or slabs isolated by branching and re- joining of veins are able to remain suspended in a fuid of almost the same density while it is mobile and the fuid silica gel slurry would preserve the sharp conchoidal fragments and slivers of the sof fractured wall materials. Injection of quartz veins is by hydraulic fracture, the materials injected being fractured by the diferential pressure or fuid “overpressure” but the incompressible fuids carry the main compressive load of deep burial Figure 39. Colloform banding in the quartz magnetite and sulphide hydraulically (Phillips23, 1972). Tis “hydraulic fuid” is lode cored at New Cobar Mine, has been further disrupted at the thixotropic polymeric silicic acid accumulating in the gel precursor stage. Vein quartz ‘wells’ across the colloform bands normal water outfow or “plumbing system” by which and one of the magnetite layers has been brecciated so that highly angular and delicate fragments of now very heavy magnetite are water escapes from sediments during lithifcation. Te suspended in a matrix of more fuid and rheopectic silica gel. Te silica originates from the natural amorphous silica in rheopectic re-setting of the precursor polymeric silica has “frozen” the sedimentary materials themselves. the delicate magnetite precursor fragments in place. 64 John Elliston

Figure 40. Fluid polymeric silica net veining intrudes a sparsely mineralised section of the Kaiser prospect core near Orange, NSW. It shows a number of features resembling the intrusion of a clastic Figure 42. Several successive ‘layers’ of additional quartz have been dyke. It has been formed by hydraulic expansion of the vein space injected at the precursor stage beside the frst of these quartz veins so that angular fragments are rotated and slivers of cohesive sedi- in the Wagga Tank prospect drill core. Chlorite septa on their mar- ment parted from the margins. Tis type of fuid vein is transitional gins separate the diferent additional intrusions but this has largely to a clastic dyke. been disrupted by infux of polymeric silica. Tis has dispersed and disoriented pieces of the chloritic vein walls through the later intruded silica. Small chlorite ‘needles’ have crystallised in the sof polymeric silica before it fnally also crystallised.

in the non-mobilised static condition. Periodic earthquake shocks, slumping, or mechanical disturbance to the sediments, particularly to the wetter seepage zones, episodically re-liquefes those parts of the sediments or lodes that are thixotropically sensitive to that level of disturbance. Each disturbance therefore triggers another episode of reactivation of the dykes, new injections, and extension of existing hydraulic fracture systems. Repeat- ed injection of veins into pre-existing veins results in a “layered” vein complex. Tey are common (Figure 42). An early injection of vein quartz precursor fuid Figure 41. Vein quartz is “splashed in” and partly “soaked into” the along oversaturated bedding planes is illustrated in Fig- host rock mass to give the impression of a very mobile penetrative ure 32 from the Gecko Mine drilling at Tennant Creek. fuid. However, some wall rock pieces appear isolated and suspen- Tat these disrupted shale beds were sof and pliable yet ded in the vein material. Tis example is from the Macdonald Mine, Bancrof, Ontario. cohesive at the time of disruption is indicated by their plastic and random distortion in response to the movement which apparently involved some lateral slippage. Re-brecciated Quartz Veins in Sediments Early clastic diking and “water outfow” structures are closely related to the formation of quartz veins. The water outflow channels from consolidating Such dykes tend to be flled with silica gel that devel- sediment piles are seen as various types of “dirty early” ops an increasing density and purity with each succes- quartz veins, clastic dykes and sills and ptygmatic out- sive sequence of injection followed by seepage. Early welling in the sof sediments (Figure 30). In many of the quartz veins at their gelatinous stage are similar to and clastic dyke type “outfow” structures the hydraulic frac- ofen form part of the clastic dykes. Like the semi-con- turing breaks and re-breaks the “gelled” semi-consolidat- solidated sediment, the gel of the early vein quartz can ed sediments into splintery conchoidal fragments typical be fractured by an episode of mobility and the disrupted of gel fracture, such that they fll the dyke (see Figure 31). fragments or “wisps” of the original gel are ofen found Te fuid outfow through the transecting channel- dispersed in the new matrix. ways and dykes is continuous, being by slow seepage A large body of discoloured impure quartz occurs in through the fabric of the “gelled” sediments themselves or near the lode zone of the Peak Gold Mine at Cobar, or the “re-gelled” dyke and vein fllings when these are NSW. Tis ‘elvan’ or peripheral lode quartz was inter- Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 65 sected in the 1986 drilling. It is a body of siliceous min- 401, and Figure 10 (on page 12)]. Frondel1 (1962, p. 300) eral matter that has been episodically intruded as a dia- lists its occurrence in a much wider variety of so-called piric or vein-like mass of wet thixotropic precursor silica igneous rocks. It is recorded from the basaltic rocks of gel where the high fuid content of material in the lode Northern Ireland, the Faroe Islands and Iceland; from zone made it lighter than surrounding semi-consolidated the trachytes of the Siebengebirge in Germany; from sediments. Te sof plastic mobile nature of its precursor serpentine at Kosemutz in Silesia; from cavities in basal- silica gel is strikingly indicated by the breccia phases (see tic rocks in the Puy-de-Dome, Plateau Central, France; Figure 37) in which angular fragments of sof chloritic from reddish brown to pinkish so-called rhyolite in the wall-rock are randomly rotated and dispersed or ‘disinte- state of Queretaro, Mexico; and from trachytic rocks grated’ in their milky fuid siliceous matrix. near Erandique, Honduras. One other observation worth noting is that jasper, chalcedony, or chert veins are rarely found in or origi- Veins of recognised meta-colloidal silica such as jasper, nating from bedded cherts or massive bedded jaspoidal opal, and chalcedony formations. Tese are frequently crowded with veins but the mobilised silica is almost invariably white vein Jasper, chalcedony, and chert are among the forms of quartz. Te reason is that the later episodes of mobil- cryptocrystalline quartz that are regarded as originating ity, during which the more mobile precursor vein gel from colloidal silica. Tere is such a strong body of opin- is injected into the semi-consolidated parent chert bed, ion to this efect that when agates are found in supposed efects a “cleaning” of the silica. As in the development volcanic rocks or jaspers found in rhyolites, these are then of a white quartz vein in sediment, or a coarsely crys- postulated as secondary or “introduced later infllings”. talline galena vein from a fahl ore bed, the thixotropic What the occurrences of agate, chert or jasper imply about liquefaction of the precursor gel allows a separation of the rock in which they occur is not ofen considered. particles of diferent macromolecular shapes (discard of Frondel1 (1962, p 299) records the occurrence of impurities by rejection from closer packed aggregates). gelatinous silica itself in some so-called “igneous” and Te “clean” precursor sol is mobilised into the vein. sedimentary rocks. He lists occurrences of gelatinous sil- It reverts rheopectically to a gel and most of the silica, ica in the diabase sills of New Jersey, with hydrophane in which eventually flls the vein with crystalline milky porphyry at Hubertusberg, Saxony, in the Chalk forma- white quartz, is added by difusion. Further monomeric tion in England, and in cavities in marly and finty rocks and oligomeric silica is added into this precursor vein in France. gel during its aging, densifcation, and crystallisation. Flow banded red jasper forms the main bulk of some parts of the “rhyolite” outcropping in the creek at Box Range Farm, near Lochiel in southern N.S.W. As Colloform Quartz Veins banded rhyolitic lava under atmospheric or limited con- fning pressure this rock contains far too much silica for Colloform banding is observed in many quartz viscous fow other than at very high temperature (near veins, ofen with gold or complex selenide or telluride the melting point of quartz - 1723°C). Jasper is not a gold-silver ores (Lindgren8, 1933, pp. 497 and 505). Te product of fusion or an “aqueo-igneous magma”. banded colloform quartz frequently contains thin con- Chalcedony veins and amygdales containing un- centric crusts of fbrous quartz (Figures 43 and 44). Tis weathered pyrite framboids occur in an outcrop of so- fne radial crystallisation is typical of that developed called hypersthene andesite at Allendale, 19 km west between the concretionary bands in many oolites and of Maitland in NSW (Ostwald and England24, 1977, p. orbicules. It can be seen in the orbicules from the Gold- 113). Un-oxidised pyrite occurs as irregular patches en Plateau Mine quartz vein (Figure 45). about 10 mm in size within the chalcedony that also has Figure 46 shows colloform banding in a quartz vein Liesegang-type banding surrounding the pyrite patches. from the Abra Prospect core, Jillawara, WA. Some small Both this type of agate banding in the chalcedony and crystals growing out from some of the colloform bands the concretionary framboidal structures indicate difu- are over-grown with further gelatinous precipitate. Te sion and reducing conditions in the precursor polymeric irregularities, disrupted layers, and ‘oozy’ nature of the silica which occurs in this hypersthene andesite as veins. precursor silica together with a later small veinlet of Opal has been defnitely established as a form of similar “pearly” quartz that cuts across the colloform colloidal silica with “packed” amorphous silica spheres bands, refect the sof and highly hydrous nature of the of about 180 to 250nm in diameter clearly visible in the original deposition. Te textures clearly indicate a gelati- electron micrographs [Dana25, 1932, p. 475; Iler2, 1979, p. nous stage in the accumulation of this vein quartz. 66 John Elliston

Figure 43. A quartz vein from Creede, Colorado, shows successive crystallisation of bladed quartz towards a central vug from colloform banded quartz deposited near the vein walls. Te history of aging of the polymeric silica in this vein is plainly the same as that in geodes where a similar pattern (Figure 26 on page 56) of cleaner crystals develops. Tese grow from colloform outer layers toward a central shrinkage cavity. Te chain of shrinkage vugs where it occurs in quartz veins, is usually central along the length Figure 45. Orbicular spherulites occur in vein quartz from the of the vein but a pattern of crystallisation from colloform banding Golden Plateau Mine at Cracow in Queensland. Concentric zones confrms the initial introduction of polymeric silica gel. of radial quartz crystals surround central orbicular cores which are richer in chlorite. Tese concretionary structures depend on syneresis in the polymeric silicic acid precursor and the radial crystallisation pattern is typical of the outgrowth round nuclei in gels.

Figure 44. Contemporaneous veining and re-mobilisation of colloform quartz banding occurs in core from the New Cobar Mine, Figure 46. Colloform banding in quartz veins sometimes occupies NSW. Tree sets of quartz veinlets in the centre bands cut each the whole vein space as in this example from the Abra prospect at other and one of them parallels and merges into a colloform layer. Jillawarra, west of Mt Morgan in WA. Some post depositional oozy Cleaner ‘cloudy’ white quartz welling out as an expanded band and movement can be seen between the bands and towards the lef a cutting other colloform layers and the irregular wisps of ‘poured in’ small dark stylolite-like vein has a pearl-grey branch veinlet outli- quartz in the upper core confrm the mobility of the newly deposi- ned by pale silicifed margins as it passes up through the colloform ted silica. layers.

Druse and Miarolitic Cavities in Veins and mineral lodes. In all these cases they refect the condensation and crystallisation of the original gelati- Miarolitic crystal cavities, vughs, and druse cavi- nous precursor. Tis formerly flled the whole space ties are found in geodes, agates, in the centres of large but on condensation from the ‘rind’ or outer surface quartz accretions, in septarian nodules, fint nodules in the centres reduce to very watery gels or water pockets limestone, and they commonly occur in quartz veins (still found in some geodes) from which the remaining Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 67

meric silica gel precursors. Exceptions are sandstones, quartzites, gravel beds and conglomerates but these are silicifed or ‘cemented’ to hard erosion resistant rocks by the mobile forms of natural polymeric silica. Difusion of monomeric and oligomeric silica ‘densifes’ and hardens chert, jasper, fint, chalcedony, agate, etc. Quartz in veins and mineral lodes and in accretions in porphyritic, metamorphic and granitic rocks crystallises from gelatinous polymeric precursors. Part 2 will present further evidence of the features, behaviour and crystallisation of quartz in veins. Tese provide further support for the conclusions from Part 1 and for the new and revolutionary conclusions from the industrial research programs recorded in Elliston9, 2017.

GLOSSARY

Figure 47. A typical pattern of druse cavities in a quartz vein illu- Accretion: is rapid formation of clusters of similar strates how the polymeric silica initially silicifes wall rocks and flls shaped particles to form ‘close packed’ and pre-ordered the whole vein space. It then contracts to central watery silica gel aggregations at net lower surface energy in any remobi- into which the large quartz crystals grow until essentially water or salt water is lef in the cavities. lized concentrated fuid paste containing colloids. Crys- tallisation of these pre-ordered aggregates occurs subsequently to then form a ‘porphyroblastic’ texture where silica crystallises in coarse prismatic crystals (see Fig- the large crystals are set in a fner grained matrix of ure 26). crystallised sedimentary material. A similar condensation and crystallisation pattern Adsorption: is the adherence or fxation on a sur- (usually outward from the wall of the vein) is found in face (usually but not necessarily a colloid because of the many druse cavities and vughs, lines of cavities, and enormous area, surface energy, and charge) of an ion or comb structures in quartz veins (Figures 43 and 47). Tis charged particle. Te uptake by a surface of a solute or refects the gel condensation and crystallisation of the dispersion can occur by electrostatic, dipolar, quadrapo- original silica gel precursor of the vein so that it can also lar, linkages or hydrogen bonding, etc. Where a chemi- suspend heavy fragments (as in Figure 29 on page 58). cal linkage is involved, the surface-controlled reaction is called chemisorption. Te dispersed ions and charged particles compete for adsorption sites on all available CONCLUSIONS FROM PART 1 surfaces. Changes in concentration, pH, in the availability of surfaces, and in the permeability (spacing of the The mobile forms of natural silica are the oligo- meshwork through which the ions and particles can dif- meric silicic acids and short chain polymers. Depending fuse) ofen have quite marked efects in exchanging and on conditions of pH and other salts present, these tend replacing surface adsorbed species. to establish equilibrium by slow difusion in pore water Aggregate: refers to a mass or body of any sub-units such as smaller gelatinous accretions or concretions. and through veins and openings. Te monomer Si(OH)4, is the smallest neutral hydrated form of silica and these Tese can crystallise as a mosaic of small interlocking molecules can difuse through the fnest pore spaces, gel crystals, as a composite of complexly intergrown crys- meshwork, or porous material. Any reaction that removes tals, or in optical continuity as a single ovoidal crystal. monomeric silicic acid from the fuid phase creates a dif- Rounded or irregular zones and patches of granular sed- fusion gradient that in time provides the inflling silica to iment or matrix cemented by infll concretion have also efect replacement or to complete crystallisation. been referred to as aggregates. Te observations and diagrams illustrated in Part Concretion: this is the slow or step–wise accumu- 1 and in Elliston9, 2017, provide completely conclusive lation of material about a central nucleus to produce a evidence that natural crystalline quartz in rocks, quartz banded–textured spherical or elliptical accumulation of veins and mineral deposits has crystallised from poly- higher particle density and compaction than the medi- 68 John Elliston um in which the particles are difusing. A concretion similar particles or macromolecules but natural gelati- may be homogeneous, being self–nucleated, homoge- nous sediments are complex mixtures of charged parti- neous but nucleated on a foreign body, or heterogene- cles having a wide range of sizes, shapes, and composi- ous (i.e. banded) with or without a specifc nucleus. Te tions. Hydrolysis reduces most particles to the colloidal active process of concretion depends on colloidal par- size range but these can form a matrix to larger residual ticles individually difusing towards the precipitating grains. Gelatinous ferric hydroxide, hydroxy carbonates, surface represented by the boundary of a higher density hydrated organic matter and silica gel occur in most gel aggregate with the less dense surrounding medium sediment and sometimes as major constituents. How- through which the particles are difusing. Concretion ever, the most common particles in pelitic sediments are could be considered to represent “adsorption” of ions or clays, amorphous silica, and hydrous ferromagnesian colloidal sol particles onto a growing nucleus, and fnally minerals. Tese common particles are shaped as plate- onto a growing macroscopic aggregate of particles or a lets, spheres, and rods respectively and in the “gelled” or denser gel surface. Removal of such particles from dis- coagulated condition they link together to form ‘house persion by precipitation at a nucleus or at an interface of cards’ or ‘book-house’ structures, ‘strings of beads’, between random open meshwork gel and denser ordered and ‘scafold-like’ structures of rods. Natural sediment gel creates a difusion gradient (fewer particles in that can be thought of as these several types of structures vicinity). To equalise the concentration, other particles randomly intermeshed and securely cross-linked togeth- under Brownian motion arrive in turn to precipitate er by the mutual satisfaction of coulombic charge sites (adsorb) and accumulate on the surface. and by van der Waal’s attractive forces where particles Crystal growth: requires particles to move to the are appropriately packed or close enough. It is not sur- surface - Removal of a mobile species from dispersion in prising that wet sediments have shear strength! the gel by its crystallisation creates a difusion gradient Semi-solid gelatinous sediments are thixotropic and towards the crystal surface. Henisch13 (p. 51) notes that have a defnite yield value (Bingham yield point). Te convection currents are suppressed in gelatinous media strength of the sediment fabric is very sensitive to water and therefore movement must be essentially by difusion content and to the presence of focculating or defoccu- but a very important function of the gel media is also to lating agents. Liquefaction is isothermal and mechani- suppress nucleation. Without growth on many closely cally induced but the linkages between particles tend to spaced competing nuclei, the faces of crystals supported re-form during viscous fow of the mud. Te systems are in the gel medium are supplied by a steady difusion of “self-healing” but there is a time delay in reverting to the particles or ions so that large and well formed-crystals original gel strength called hysteresis. Cross-links are are able to develop. more readily broken at higher temperatures. Transition Crystal lattice: describes the stable meshwork of from an elastic gel to a liquid of relatively lower viscosity chemical bonds that hold the atoms of a crystal together in occurs revertably over a narrow temperature range. Te an ordered repetitive pattern of unit cells so that the com- more concentrated gels require higher temperatures but pound that has crystallised achieves a low energy state. the thermal energy “sofens” the paste and makes it eas- Crystallisation of feldspar: a number of natural ier to disrupt the fabric of particle linkages. Gels melt! clays in close packed aggregates react spontaneously Helmholtz double layer: in an aqueous electrolyte with alkali metal ions and monomeric silica to feldspar solution in the vicinity of a charged surface the aqueous and water with the liberation of heat. Feldspathoids are phase is divided into four regions of distinct dielectric sometimes formed as an intermediate product. Reactions behaviour. Te innermost region) consists of preferen- are described on page 214 of Elliston9, 2017. tially oriented water molecules in contact with the solid Crystallisation of quartz: most natural quartz has surface and where specifc ions are adsorbed without crystallised from compact polymeric species to which a their hydration shells. Tis is called the inner Helmholtz further and continuing supply of the monomer is availa- layer. Te region further from the surface (β in Figure ble. Some details are set out on pages 10-13, pages 21-26, 1.7 in Elliston9, 2017) contains both free water molecules pages 126-130, and pages 215-216 of Elliston9, 2017. and molecules attached to hydrated ions. Tis is called Gel: a gel is essentially a semi-solid meshwork of the outer Helmholtz layer and is defned by the by the fne particles coagulated or focculated by the inter-tan- closest approach that a fully hydrated charged ion can gling of long chained polymers where the particles are make to the solid – liquid interface. Further out from linked to form a visco-elastic permeable solid by interac- the surface the concentration of counter-ions (having tion between electric charges on their surfaces. Synthet- a charge of opposite sign to the surface) decreases with ic gels of pure clay, silica gel, gelatin, agar, etc. contain increasing distance in the Gouy-Chapman difuse layer. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 69

Te outer and inner Helmholtz layers are referred to as years in even greater quantities of slightly alkaline sea the Helmholtz double layer. water. Reversible chemical reactions are driven by the Hydrated silica polymers: a diagrammatic repre- quantities of reacting substances. In the late stages of sentation of the polymerisation behavior of silicic acid is diagenesis when natural sediments are losing water, the shown in Figure 5. hydration reactions reverse and siloxane linkages pre- Mobilisation: means the liquefaction of a body of dominate. semi-consolidated sediment or other particulate mate- Silanol: Is a fully hydrated form of silicic acid that rial usually by earthquake shock or gravity sliding can condense to a direct chemical silicon-oxygen-silicon downslope. linkage by loss of a water molecule. Rheopexy: is the accelerated resetting to a gel condition in a fowing colloidal dispersion subjected to shear by laminar fow of a thick paste. Tixotropic liquids may rapidly revert to a higher viscosity condition when linkages establishing between particles throughout the fow- Siloxane: the direct silicon-oxygen-silicon chemical ing mass overcome the momentum of the moving mass. linkage is called siloxane. Tis “instant re-freezing” preserves fow foliation, the Sol: is a homogeneous suspension or dispersion of shape and form of intrusions, suspends fragments, etc. colloidal particles in a liquid or gas. In the glossary of Silica: Ordinary sand or crystalline quartz (SiO2). geology, a sol is also defned as a completely mobile mud Silica Polymers, where do all the natural silica that is in a more fuid form than a gel. polymers come from? Quartz is not soluble in water Small particle systems: are materials or substances including normal ground water and stream water in the made up of small particles. Tey are independent of the cycle of erosion. It is transported by the streams and riv- chemical composition of the particles but the very small ers as gravel and sand grains and generated by coastal size of the component particles means that the surface erosion as sea sand. Quartz does not dissolve in seawa- charge enables their interaction with other charged par- ter by dispersion of anions and cations as a solution but ticles and ions in the pore fuids surrounding them. Mud it does hydrolyse (react with water) in slightly alkaline is “sticky” because the particles cling to each other and conditions (seawater pH 7.9 to 8.3) by a process called to surfaces they come in contact with. Examples of small “proton promoted dissolution” (Iler2, 1979, fg. 1.11.) Tis particle systems are mud, clay, silica gel, thick paints, is shown diagramatically as: - food colloids (like yoghurt, cream, soup, etc.). Molecular dispersion from crystalline silica in water Surface charge: See Figures 3 and 4 and captions as Si(OH)4 is catalysed by hydroxyl ions of an alkali or and description in associated text. base. Seawater is slightly alkaline and therefore silica Surface chemistry: is the study of the special chem- (and most silicate surfaces) “disperse” by these surface istry that is related to the solid-water interface. Surface reactions. In sea water and within marine sediments chemistry and colloid chemistry are closely interrelated the small neutral Si(OH)4 molecules polymerise to short because the behaviour of colloidal particles is dependent chain polymeric silicic acids called “oligomers”. on the properties of the very large surfaces they present Silicic Acid: See Figures 1 and 2 and captions. to the solvent in relation to their very small volume. Te Where do all the natural silica polymers come solvent, ions, complexes, and other charged particles from? Te answer to this is that there are immense interact with all surfaces but are especially important in quantities of sand and silicates soaking for thousands of their interactions with colloidal particles. Surface energy: is the diference in energy per unit area between the surface of a given crystal lattice or substance and the energy of the same number of atoms (comprising the unit area) situated within the bulk of the crystal or substance. Surface energy is clearly dependent on the atomic geometry of the atoms exposed within the unit area of surface. Te atoms exposed at the surface are able to interact with particles, ions, solvent, or other substances. Tey have ‘dangling bonds’ or charge that can compensate each other, hydrate, adsorb surface species, or form new chemical compounds. Bonds or linkages of atoms comprising a comparable 70 John Elliston area within the crystal or substance are in equilibrium since 1984 and ensuring that the conclusions were with those surrounding them. reached by the prescribed scientifc method. Syneresis: is the spontaneous aging or contraction of a gel meshwork within itself by the establishment of a greater density of cross–linkages and elimination of REFERENCES water. Te particles or particle–chains achieve greater co–ordination. Te total surface energy is lowered, and 1. Frondel, C., 1962. Dana: Te System of Mineralogy. the internal surface and adsorptive capacity are reduced. 7th Ed. Vol. III, Silica Minerals. Wiley and Sons, New Te contraction and greater gel density causes shrinkage York, 334p. cracks or a pattern of holes or channels (like those in 2. Iler, R.K., 1979. Te Chemistry of Silica. John Wiley cheese) which is independent of whether or not the gel is and Sons, New York, 483p. immersed in water. In syneresis the particles move closer 3. Barnes, H.L., 1967. Geochemistry of Hydrothermal together under the infuence of van der Waal’s attractive Ore Deposits, Holt, Rinehart and Winston, New forces so that the less dense, sparse, weak “watery” gels York, 670p. tend to be less or non–synerectic. Te crystalline state is 4. Stöber, W., 1966. Formation of Silicic Acid in Aque- the low energy state of matter. ous Suspensions of diferent silica modifcations, in Tixotropy: is the isothermal reversible re-lique- Equilibrium Concepts in Natural Water Systems, faction of a gel or coagulated sol. Tixotropy is due to Chap. 7, Advances in Chemistry Series, (Amer. mechanical shock or shear which disrupts the gel parti- Chem. Soc.) pp. 161-182 cle linkages allowing the colloids of the system to revert 5. Stumm, W., 1992. Chemistry of the solid-water inter- to a dispersed sol or more fuid gel at the same fuid face: Processes at the mineral-water and particle- content. Tis isothermal gel to sol or to more-fuid-gel water interface in natural systems. A Wiley-Intersci- transformation is reversible and repeatable. Thixot- ence publication, New York, 428 pp. ropy is mainly induced by shock. A short sharp oscilla- 6. Healy, T.W., 1972. Physico–Chemical Processes in tion throughout the gelatinous mass is more efective in the Diagenesis of Sediments. Unpublished Paper, destroying all, or sufcient of, the interparticle linkages Dept. of Physical Chemistry, University of Mel- at the one time so that the meshwork structure will col- bourne, Vic. 102p. lapse and the material revert to a fuid. Diferential liq- 7. Lindgren, W., 1937. Succession of minerals and tem- uefaction depending on hydration (difering Bingham peratures of formation in ore deposits of magmatic yield points) of diferent colloidal components allows afliations, Trans. Amer. Inst. Min. Met. Eng., 126: separation of more mobile hydrous materials which can 356-376. then simply fow out of the disturbed mixture. 8. Lindgren, W., 1933. Mineral Deposits. McGraw-Hill, Rheopexy: is the accelerated resetting to a gel condi- New York, 930 pp. tion in a fowing colloidal dispersion subjected to shear 9. Elliston, J., 2017. Te origin of rocks and Mineral by laminar fow of a thick paste. Tixotropic liquids may Deposits - using current physical chemistry of small rapidly revert to a higher viscosity condition when link- particle systems. Connor Court Publishing Pty Ltd, ages establishing between particles throughout the fow- Brisbane, 706 pp. ing mass overcome the momentum of the moving mass. 10. Elliston, J., 2014. Australians Successfully Exploring Tis “instant re-freezing” preserves fow foliation, the Australia –and developing the science and technol- shape and form of intrusions, suspends fragments, etc. ogy to do it. Elliston Research Associates Pty Ltd, Castlecrag, NSW. 11. Healy, T.W., 1973. Accretion of Colloidal Materials. A ACKNOWLEDGEMENTS report to Peko-Wallsend Limited, from Department of Physical Chemistry, University of Melbourne, Te permission of CRA Exploration Pty Ltd to pub- Parkville, Victoria, June 1973. lish the content of this company report on the hydration 12. Newton, Z.J., 2000. Waitaia Silica Sinter: Occurrence of silica and the formation of quartz veins is gratefully and aspects of genetic history. Honors thesis, Depart- acknowledged. Permission from Connor Court to publish ment of Geology, University of Auckland, N.Z. photographs and information from the treatise, Elliston9, 13. Henisch, H.K., 1970. Crystal Growth in Gels. Penn. 2017, is also thankfully acknowledged. Te author is also State Uni. Press, University Park, Pennsylvania, 111p. appreciative and thankful for the interest and fnancial 14. Hatschek, E., and Simon, A.L., 1912. Gels in Relation contribution of AusIndustry for monitoring this research to Ore Deposition, Trans. Inst. Min. Met., 21: 451-479. Hydration of Silica and Its Role in the Formation of Quartz Veins – Part 1 71

15. Boydell, H.C., 1925. Te Role of Colloidal Solutions 20. Hellmers, J.H., and Ottemann, J., 1952. Crystalliza- in the Formation of Mineral Deposits. Bull. Inst. Min. tion of Quartz at Low Temperatures. Silikattechnik 3 Met., 243: 1-103. Jg. Hef 2 February 1952. 16. Healy, T.W., 1969. Te efect of polyelectrolytes on 21. Levings, J.H., 1912. Gelatinous Silica in the Great PbS crystal growth in silica gel. Report on Research Australia Mine, Cloncurry, Queensland. Trans. Inst. at the University of Melbourne, Geopeko Technical Min. Met., 21: 478. Seminar, Mount Morgan, February 1969, (Unpub.) 22. Bowen, K., 1967. Notes on the Occurrence of Mutton 17. Kvenvolden, K.A., and Roedder, E., 1971. Fluid Fat at Wood’s Point, Victoria, Report from Depart- Inclusions in Quartz Crystals from South West Afri- ment of Mines, Victoria, (unpub.). ca. Geochim. Cosmochim. Act., 35: 1209-1229. 23. Phillips, W.J., 1972. Hydraulic Fracturing and Miner- 18. Petersilie, I.A., and Sorensen, H., 1970. Hydrocar- alization, Jl. Geol. Soc. London, 128: 337-359. bon Gases and Bituminous Substances in Rocks from 24. Ostwald, J., and England, B.M., 1977. Notes on fram- the Ilimaussaq Alkaline Intrusion, South Greenland. boidal pyrite from Allendale New South Wales, Aus- Lithos 3: 59-76. tralia. Mineral. Deposita, 12: 111-116. 19. Spezia, J., 1899. Gelatinous Silica in the Simplon 25. Dana, E.S., 1932. A Textbook of Mineralogy. Wiley Tunnel. Atti. Acad. Torino, 34: 705. and Sons, New York. 4th Edition, 1947, 851 pp.

Firenze University Press www.fupress.com/substantia

Research Article Visualizing Chemistry. Te Application of Chemical Imaging to Address Scientifc Citation: V. Lusa, A. Franza (2018) Visualizing Chemistry. The Application Challenges in Space Research of Chemical Imaging to Address Sci- entifc Challenges in Space Research. Substantia 2(2): 73-80. doi: 10.13128/ substantia-61 Vincenzo Lusa1, Annarita Franza*2 Copyright: © 2018 V. Lusa, A. Franza. 1 JD, Pontifcia Università San Bonaventura, Via del Serafco, 1 - 00142 Rome (Italy) This is an open access, peer-reviewed 2 PhD, Dipartimento di Scienze Biomediche, Sperimentali e Cliniche “Mario Serio”, Uni- article published by Firenze University versità di Firenze, Largo Brambilla, 3 - 50134 Florence (Italy) Press (http://www.fupress.com/substan- E-mail: [email protected]; [email protected] tia) and distribuited under the terms of the Creative Commons Attribution License, which permits unrestricted Abstract. Chemical Imaging helps to answer difcult questions, especially when those use, distribution, and reproduction questions occur in complex environments. For instance, forensic neuroradiology plays in any medium, provided the original an important role in the courtroom to understand a defendant’s personality. But could author and source are credited. this branch of science be essential in human exploration of space? Even if no emergen- Data Availability Statement: All rel- cy has happened so far, NASA established a partnership in 2002 with the U.S. National evant data are within the paper and its Institute of Justice to promote the knowledge of investigative techniques in the case Supporting Information fles. of a crime being committed on a space mission. Based on forensic neuroradiology and behavioral genetics, this article presents a brand-new study protocol for creating Competing Interests: The Author(s) security procedures designed to safeguard astronauts engaged in long-duration space declare(s) no confict of interest. travel. Since 2009 in Italy, some individuals have been prosecuted who, although con- victed of murder, beneftted from reduced sentences through the verifcation of some genetic polymorphisms and Computed Axial Tomography (CAT), Positron Emission Tomography (PET), and functional Magnetic Resonance Imaging (fMRI) results which showed brain malformations that may produce manifestations of violence. Te protocol specifcally uses chemical imaging and behavioral genetics to show how cerebellar anomalies and biological markers predictive of criminal behavior can trigger impulsive reactions in response to stress. Tis protocol may prove critical when space agencies are evaluating candidates for extra-orbital fights of long duration.

Keywords. Chemical imaging, forensic radiology, NASA, behavioral genetics, space research.

INTRODUCTION

In 1958, President Eisenhower proposed creating a civilian space agency for the United States. Afer considerable congressional debate, the National Aeronautics and Space Administration (NASA) began its operations on October 1, 1958. No one could ever have foreseen what would occur over the next 50 years as humans discovered that the sky was no longer a limit and it is now about to enter a new era of space exploration with a 365-day stay on the ISS, the establishment of a lunar landing out-

Substantia. An International Journal of the History of Chemistry 2(2): 73-80, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-61 74 Vincenzo Lusa, Annarita Franza post, and the future missions to Mars and to a near- flight and has listed them in the Human Bioastro- Earth asteroid (NEA).1,2 nautics Roadmap. Te level of risk refers to three specifc As missions to NEA and Mars are not just ordinary parameters: a 1-year tour of duty on the ISS, a month- space fights, the analysis of physiological and psycho- long stay on the moon, and a 30-month mission to Mars. logical risk factors of past missions will not be relevant Health risk factors are then categorized into: (1) behav- bases for predicting the risk factors astronauts may ioral health and performance; (2) human health counter- encounter during space fights.3 At the Hawaii Space measures (e.g., bone metabolism and physiology, nutri- Exploration Analog and Simulation center, NASA con- tion, immunology, cardiac and pulmonary physiology); cluded in 2016 a year-long study regarding the use on (3) space radiation; (4) space human factors and habit- a shuttle of the forensic sciences utilized in examining ability; and (5) exploration medical capabilities. Te risk crime scenes.4 In fact, the probability of criminal con- factors have been rated as Controlled, Acceptable, Unac- duct in space is not so remote. For instance, astronauts ceptable and Insufcient Data. Te Roadmap then high- have carried weapons into space since the Soyuz fights lights as unacceptable the risk of behavioral issues and in 1965.5 Tus, forensic scientists are prepared to face psychiatric disorders.14 the eventuality of assessing crime scenes and identifying In 2007, NASA developed various health standards evidence during space travel. So it is no coincidence that (i.e., Crew Health Concept of Operations and Medical NASA established a partnership with the U.S. National Operations Requirements) directed towards “a healthy Institute of Justice to implement investigative techniques and safe environment for crewmembers during all phas- for use in future crime scenes in space.6 es of space fight”.15 A review of all the NASA health Over the last two decades, several studies have been standards is conducted every 5 years or any time new published regarding the impact of long-duration space scientifc data suggest that an update is needed. fight on the health of crewmembers.7 Moreover, while In addition to the standards described above, NASA the physiological efects of space fight are well docu- has identifed specifc health criteria for crew selection mented,8 a paucity of knowledge exists on the potential including a medical screening and annual examinations behavioral and cognitive issues that can afect the astro- reported in the Astronaut Candidate Program and sup- naut’s psychophysical performance during fights. For ported by the NASA Aerospace Medicine Board.16 It is instance, the neurobiological consequences of long-dura- worth mentioning that health risk and stressor param- tion space fight and long-term exposure to microgravity eters may be diferent from mission to mission. Con- can lead to alterations in cerebral morphology (e.g. the cerning behavioral health and performance risks, NASA Visual Impairment and Intracranial Pressure Syndrome) has identifed three stressors associated with long-dura- that are poorly understood.9.10 tion space fight: (1) adverse behavioral conditions and Even if no emergency has happened so far, we need psychiatric disorders;17 (2) performance errors due to to think the unthinkable: how would we deal with a cat- fatigue;18 and (3) performance decrements due to inad- astrophic scenario during a long-duration space fight? equate cooperation, coordination, communication, and In this paper, we present a study protocol that aims at psychological adaptation within a Team Gap.19 Tese cri- certain security procedures designed to safeguard astro- teria were identifed by analyzing observational data and nauts engaged in long-duration space travels. Te pro- experimental studies in analogous settings, such as sub- tocol is based both on chemical imaging, e.g. forensic marine,19,20 space simulations,21 mountain survival pro- neuroradiology, to evaluating the cerebral regions whose grams and polar expeditions.22 anomalies can lead to antisocial behaviors,11 and on It is important to note that behavioral health prob- behavioral genetics where it has been demonstrated how lems may be underestimated due to the reluctance of specifc predictive biomarkers of criminal behavior can crewmembers to report them.23,24 In 2000, a survey lead to an unpredictable response in very stressful situ- documented a 2.86 per person-year incidence of behav- ations.12 Tese evaluations can be crucial for space agen- ioral problems among the 508 astronauts engaged in cies at a time in which space tourism is going to become space shuttle fights between 1981 and 1989.25 In 2009, an achievable dream.13 observational information collected for 28.84 person- years described anxiety and irritability as the most common behavioral symptoms (0.832 cases per person- MATERIALS AND METHODS year).26 Furthermore it was pointed out that behavioral health issues are frequent during long-term periods of NASA’s Human Research Program has identi- isolation and confnement.27 A 2004 study reported a fed 32 space-related health risks associated with space 5.2 percent probability of behavioral issues in aspiring Visualizing Chemistry. Te Application of Chemical Imaging to Address Scientifc Challenges in Space Research 75 astronauts afer extended stays in Antarctica. Among the in space for astronauts. Its absence could lead crewmem- most common risk factors of behavioral health problems bers to an increased sense of isolation, emotional insta- are mission duration, circadian rhythm disorders and bility, hypersensitivity, depressive reactions, and possibly microgravity efects on human physiology, social isola- psychiatric disorders.32 tion, cultural and organizational issues and personal- As has already been mentioned, many unknown ity traits.28 Risk management procedures have focused health risks may infuence long-duration space fight. In on providing social and psychological support to crew- this regard, it is noteworthy to emphasize that chemi- members through ground control. However, even if cal imaging can make a positive contribution to foren- these countermeasures have proved to be efective on the sic space research. For instance, a recent clinical survey ISS, it is unknown whether they could be useful during tested via fMRI the neuronal functions of a 44-year-old long-term space fights at greater distances from Earth.2 male cosmonaut engaged in a long-duration space fight Cognitive and psychiatric tests are administered dur- (169 days) on the Russian segment of the ISS. Even if ing both the astronaut enrollment procedures and the his physical state and performance showed no relevant annual medical examinations.29 Neurocognitive and anomalies, the fMRI scans indicated abnormalities behavioral genetic testing have also been developed, but in the vestibular and motor-related cerebral regions.10 they serve mainly to point out phenotypic and geno- Moreover, results from the MARS105 study have shown typic variations to manage sleep-wake disruptions dur- a deterioration in the astronauts’ psychological state during space missions.30 To date, few policies have been ing a long period of confnement, which was accompa- established for long-duration space fights. In analogous nied by a decrease in brain cortical activity.33 Again, in a settings, the National Science Foundation’s Division of study carried out using low-resolution brain electromag- Polar Programs require a psychiatric evaluation con- netic tomography (LORETA), neuropsychiatrists have ducted by a civilian contractor of all aspiring astronauts demonstrated how the microgravity phases of parabolic engaged in winter-over duty at Antarctica’s Amundsen- fights induce changes in frontal lobe activity, a cerebral Scott South Pole and McMurdo stations. Limits have region that is involved in the regulation of emotional also been placed on the number of continuous seasons processing.34 a candidate can stay at the same station.29 Astronauts Despite what has been described above, the contri- are then informed that ordinary social and personal bution of forensic neuroradiology and neurobiology to problems can become clinically signifcant during peri- space research seems to be underestimated. Terefore, ods of extended isolation and confnement. For instance, increased focus on brain imaging and behavioral genet- the Mars Society conducted a 4-month simulated ics can lead to the development of adequate counter- Mars exploration mission at the Flashline Mars Arctic measures to safeguard astronauts involved in long-dura- Research Station in 2007. Analyzing the questionnaires tion space missions that crewmembers completed on fve diferent occasions during the simulation, scientists found an increase in stress factors as well as higher levels of excitement, lone- Brain imaging liness and tiredness. Tese results thus prompt serious questions about psychological issues during missions Cognitive neurosciences have as a study subject an to the outer solar system where a mission’s total dura- understanding of the mechanisms that are the basis of tion is expected to last 10 years or more (e.g., interstellar human aggressivity; they fnd their correspondence at travel to the Oort Cloud, a broad spherical shell of comet an organic level and in particular cerebrally. Te main nuclei located 0.63-0.94 light years away from the Sun).31 objective of this branch of research is to investigate vio- Tese kinds of human expeditions will cause astro- lent behavior in humans by studying certain cerebral nauts to undergo psychological and interpersonal stress- areas using neuroimaging techniques, which constitute a ors that they have never before experienced, such as an major source of scientifc acquisition in the neurocrime unknown level of monotony and isolation, impossible feld. It follows that cognitive neurosciences, linked to real-time communication with the Earth, which will the growing technological development in the feld, are even disappear from their view. Such would be the case destined to evolve over the next few decades infuencing such felds as psychiatry, forensic psychology, and crimi- of missions to Mars where a crew of six or seven people 35 would be on a space fight lasting at least 2.5 years. For nal law in studying the mens rea. the very frst time in human history, they would experi- The development of multiple imaging techniques ence the “Earth-out-of-view phenomenon”. Gazing at the provides powerful tools to probe multiple aspects of Earth has been one of the major positive factors of being brain anatomy. New technologies not only allow the study of structural features but also reveal brain con- 76 Vincenzo Lusa, Annarita Franza nectivity, neurotransmitter receptor profles, and other plished by injecting radioactive isotopes that have short important aspects of brain function. Tese sophisticated half-lives. Isotopes used frequently in PET research allow imaging systems are divided mainly into the following a variety of radiochemical approaches to ligand synthesis. areas: EEG (electroencephalography); CAT (computer- Of particular importance, isotopes of carbon and nitro- ized axial tomography) scan; functional magnetic reso- gen may be directly incorporated, and 18F can be sub- nance imaging (fMRI); positron emission tomography stituted for hydrogen or a hydroxyl substituent in many (PET); single-photon emission computed tomography compounds without loss of bioactivity. Blood fow indeed (SPECT); and magnetoencephalography (MEG). increases in areas of the brain that are in heavy use and In detail, EEG allows direct assessment of the brain’s a fair portion of the injected isotopes will end up in the electrophysiology by displaying the temporal and spa- active part of the brain. As the isotopes decay, a positron tial pattern of the neuronal populations generating the is released. Tis positron will collide with an electron underlying neuroelectric and neuromagnetic felds. Given and they will annihilate each other, sending two photons, its temporal sensitivity, EEG is useful in the evaluation or γ-rays, in opposite directions. Tese γ-rays are picked of dynamic cerebral functioning such as suspected sei- up by the PET scanner, which then determines where zures and unusual spells.36 CAT scan is a computerized they came from the brain. Tomographic techniques X-ray imaging technique where a motorized X-ray source analogous to those utilized in CAT scanning, are used rotates around the circular opening of a gantry, pro- to reconstruct the image from the rays. Te resulting ducing signals that are then processed by the machine’s PET images are spatial maps of radioactivity distribution computer to generate 2-D cross-sectional tomographic within tissue slices.39 SPECT is another nuclear imaging images (ofen called “slices”). Te thickness of each slice technique for imaging molecules, metabolisms, and bio- usually ranges from 1-10 millimeters. When the number chemical functions of organs and cells, and like PET, the of desired slices is collected, image data is recorded in use of radioisotopes is required. As its name suggests, it DICOM format and images can be displayed separately involves the emission of a single γ-ray per nuclear disin- or stacked together by the computer to produce high-res- tegration, which is measured directly. Numerous single olution 3-D images (c.50μm).37 CAT scan and magnetic γ-rays are detected by rotating gamma cameras to recon- resonance imaging (MRI) have allowed for the frst time struct an image of the origin of the γ-rays, which identi- the noninvasive evaluation of brain structure. In particu- fes the location of the radioisotope. Before a test is per- lar, fMRI is a modern noninvasive imaging technique formed, the patient is injected with a radiopharmaceu- to measure and localize specifc functions of the human tical that emits γ-rays and can be detected by the scan- brain without the application of radiation. Indeed, fMRI ner. Te most common radioisotopes used in SPECT are takes advantages of the diferences in magnetic suscep- Iodine-123, Technetium-99m, Xenon-133, Tallium-201, tibility between oxyhemoglobin and deoxyhemoglobin. and Fluorine-18. SPET scan is primarily used to measure When a task is performed, oxygenated blood in excess the regional cerebral blood fow. One of its major advan- of the amount needed (luxury perfusion) is delivered to tages is that SPECT provides improved contrast between the active area. Te diference in magnetic susceptibility regions of diferent functions, and also give better spa- between deoxyhemoglobin concentrations and oxyhemo- tial localisation and greater accessibility because it uses globin concentrations creates the signal in functional radioisotopes with longer half-lives.40 Finally, MEG is imaging. Brain function is indirectly assessed with high the measure of magnetic felds generated by the electrical spatial resolution via detection of local hemodynamic activity of neurons. When neurons are activated synchro- changes in capillaries and draining veins of the so-called nously they generate electric currents and thus magnetic functional areas, e.g. regions of the human brain that felds, which are then recorded by MEG outside the head. govern motor, sensory, language, or memory functions. Once generated, magnetic felds are relatively invulner- Tus, fMRI not only ofers a variety of novel options able to intervening variations in the media they traverse, for research but also opens up a new diagnostic feld of so they are not distorted by the skull, grey and white neuroradiology, by shifing paradigms from strictly mor- matter, and cerebrospinal fuid. But these neuromagnet- phological imaging to measurement and visualization ic signals are extremely small, about 10−15 T. Tus, MEG of brain function.38 Other frequently applied methods scanners require superconducting quantum interference for the identification of functionally important brain device (SQUID) sensors. To detect and amplify the mag- structures are PET and SPECT, which detect changes of netic felds, the SQUID sensors are bathed in a large liq- cerebral blood fow and glucose metabolism. Most sig- uid helium cooling unit at about -270°C. A magnetically nifcantly, PET measures generally refect the functional shielded room houses the equipment and mitigates inter- biochemistry and physiology of the brain. Tis is accom- ference.41 Visualizing Chemistry. Te Application of Chemical Imaging to Address Scientifc Challenges in Space Research 77

Te brain areas that are relevant to the present study and the jury agreed that Hinckley was mentally incapa- are the amygdala, hippocampus, thalamus, midbrain, ble of understanding the crimes he had committed.50 and prefrontal cortex. Te amygdala is involved in pred- In 1991, while engaged in a heated argument, Her- atory and afective attacks. Te thalamus connects the bert Weinstein strangled his wife Barbara and then tried limbic emotional areas and cortical areas. When acti- to cover the murder up by throwing her body from the vated, the midbrain manages impulsive aggression due window. Before the trial began, and because he had no to emotions.42 Te literature has found that reduced hip- history of violent behavior, Weinstein had an MRI fol- pocampal function can be associated with high levels of lowed up with a PET scan. Exams revealed an anomaly psychopathy.43 Te hippocampus also plays a primary in Weinstein’s prefrontal cortex and an arachnoid cyst role in fear conditioning and in emotional responses.44 growing in his lef frontal lobe. Tese fndings showed Te almond-shaped amygdala is also involved in the how the defendant was not able to control his aggressive generation of emotions. An amygdala dysfunction leads behavior in a very stressful situation and thus he was to manifesting impulsive, even violent behaviors.45 For found not criminally responsible due to mental defect.51 instance, a decrease in the amygdala’s volume equal to On October 18, 1992, Johnny Hoskins was arrested 18% can instead be the basis of sociopathic behaviors.46 for having raped, beaten and strangled Dorothy Berger, As regards the midbrain, a disorder in the posterior cin- an 80-year-old who lived in Brevard County, Florida. He gulate cortex can trigger anger. In addition, it distorts was sentenced to death on April 4, 1994. Seven months the possibility of understanding how this behavior can later, Hoskins fled a Direct Appeal with the Florida infuence others.47 Te same applies to the anterior cin- Supreme Court (FSC), claiming that the trial judge had gulate, the cerebellar anatomical region designated to improperly barred neurological testing. Upon review, inhibit automatic and instinctive behaviors and to regu- the FSC afrmed Hoskins’s convictions, but remanded late instinctive reactions.48,49 him to the State Circuit Court for a PET scan. Results Te most important evidence capable of translat- showed Hoskins as having frontal lobe lesions and thus a ing into reality the literature attesting to the biologi- lack of control of his inhibition restraints. Consequent- cal nature of the anti-social acts is found in the judicial ly, the FSC vacated the death sentence.52 practice of both Italian and US criminal trials, where a In March 2010, Brian Dugan was admitted to reduced sentence was imposed on the basis of the con- Northwestern Memorial Hospital in Chicago to have frmation of anatomical diferences in the defendant’s an fMRI and a series of cognitive, attention and moral brain as well as genetic anomalies that were capable of decision-making tests. In 1983, Dugan had kidnapped, infuencing the behavior of individuals, causing them to raped, and beaten to death 10-year-old Jeanine Nica- carry out heinous behaviors against their fellow humans. rico. One year later, he raped a 27-year-old nurse and drowned her in a quarry. In 1985, he kidnapped, raped, and killed a 7-year-old girl. Te neuroscientist Kent Kie- US criminal trials hl who concluded that Dugan’s brain sufered from an anomaly in his prefrontal cortex performed neurological Te use of bioscience in criminal cases dates back testing. Consequently, Dugan was incapable of control- to the early 1980s when chemical imaging and behavio- ling his impulses, diferentiating right from wrong, and ral genetic evidence began to enter US courtrooms. On understanding the consequences of his actions. Kie- March 30, 1981, John Warnock Hinckley Jr. shot Presi- hl’s report was used to mitigate death penalty charges dent Ronald Regan six times, also critically wounding against Dugan.53 police ofcer Tomas Delahanty, Secret Service agent One recent study reported over 1585 cases between Timothy McCarthy, and Reagan’s press secretary, James 2012 and 2015 in which neuroscience and behavio- Brady. When he was put on trial for his crimes, Hinck- ral genetics were reported in a judicial decision in the ley was found not guilty by reason of insanity and con- US criminal justice system. Te data collected show an fned to St. Elizabeth’s mental hospital in DC. He was increasing use of neurological testing by the courts, with released in September 2016. Te Hinckley case revo- over 300 decisions in 2012 alone. Neurobiological evi- lutionized the US criminal justice system introducing dence has been introduced in 5-6% of all murder trials CAT scans as a useful tool for proving mental incapac- and the most common type of brain scanning is MRI or ity. Te psychiatric David Bear explained to the jury CAT, rather than fMRI or SPECT. Te study then high- that the defendant’s brain scans indicated that Hinck- lights the use of neuropsychological testing and neuro- ley’s sulci were wider than average, a feature that he had imaging in pretrial proceedings as an improvement in noticed in patients sufering from schizophrenia. Te the evaluation of subjective competency.54 CAT scans bolstered Bear’s diagnosis of mental insanity 78 Vincenzo Lusa, Annarita Franza

Italian criminal trials - Te Bayout case an’s brain where abnormally dense gray matter was found in the anterior cingulate. As a matter of fact, the In 2007, a forty-year-old Algerian, Bayout was test showed that the volume of gray matter mentioned involved in a quarrel in Udine, which erupted into a above, in the anterior cingulate gyrus, turned out to fght as the defendant usually wore eye make-up for be diferent from that in a control group of ten healthy religious reasons and for having been insulted with women. It is noted that the previously mentioned cere- racial slurs. Tis led to his stabbing to death the indi- bellar anatomical region is intended to inhibit automatic vidual who had provoked him. Sentenced to nine years and instinctive behaviors. Moreover, in critical situa- in prison, the Court of Appeal made arrangements tions, it regulates aggressive reactions as well as mendac- for a genetic test in which anomalies were detected in ity (Como (Italy) Court, judgment 05/05/2011 no. 536).55 fve of the genes linked to violent behavior, including a gene polymorphism (MAO-A). Te court of justice stated that being a carrier of the low-activity allele for RESULTS AND DISCUSSION the MAOA gene (MAOA-L) would make Bayout more inclined to manifesting impulsive, aggressive behav- If we further examine the risks listed by NASA for ior when provoked or socially excluded (from judgment the space voyages completed so far, it seems incontro- no. 5-09/18/2009). Namely, he was afected by a kind of vertible to state that all kinds of possible emergencies “genetic vulnerability” and the presence in the defend- can indeed take place in space. Some of these contin- ant’s chromosomal inheritance of certain genes made gencies are known, while others are currently unknown. him “particularly reactive in terms of aggressivity in Indeed, the high degree of stress to which the astro- stressful situations, and even more so if the individual nauts could fnd themselves subjected could promote had spent his childhood in a disadvantageous, fam- the gradual development of uncontainable stress in ily environment” (from sentence no. 5-09/18/2009). For individuals whose genetic makeup contains particular these reasons, the Trieste Court of Justice reduced the polymorphisms or structural abnormalities in the brain sentence by one-third (Trieste (Italy) Court of Assizes that could lead an antisocial behavior being manifested. Appeal no. 5 09/18/2009). Te protocol presented here is based on forensic neuroradiology and will have to be implemented together with the psychological and aptitude tests that are usu- Te Albertani case ally used in astronaut selection. Terefore it should consider, for maximum crew safety the genetic profle and The judge for preliminary investigations at the brain anatomy of the candidates and in particular: 1) Como Criminal Court sentenced to twenty years in an examination of those brain structures involved in prison a young woman accused and found guilty of kill- controlling aggressive impulses and possible anatomical ing her forty-year-old sister. Besides also strangling her malformations including the telencephalon and cerebral mother to death, she had tried to destroy the corpse cortex on which the limbic system is present; 2) a study by burning it. Simultaneous to the sentence being proof some alleles that enable understanding the relation- nounced, a partial mental defect of the killer came to ship between genetics and crime. In fact, the protocol the judge’s attention because of the results of specifc should be based on examining those brain structures neuroscientifc tests intended to ascertain whether the involved in controlling aggressive impulses and any ana- woman presented those alleles signifcantly associated tomical defects inherent in the telencephalon in which “with an increased risk of impulsive, aggressive and vio- the limbic system is present. Examinations must also be lent behavior” (from judgment 05/05/2011 no. 536). Dur- performed on the hypothalamic-pituitary-adrenal axis, ing the trial, the judge established the presence, through (involved in the control and adaptation to stress) as well biological testing, of some unfavorable alleles present as on the connections between the limbic system (the in the defendant’s genetic inheritance, such as the low- seat of emotions) and the prefrontal cortex (control over activity MAO-A allele, SCL6A4 (STin2 polymorphism) impulses including aggressive ones) by using chemical and COMT (rs4680 polymorphism), with the sentence imaging such as functional neuroimaging techniques later essentially being reduced. High-resolution brain- that study brain function based primarily on measuring imaging techniques (Voxel-based morphometry), were blood fow (fMRI), (SPECT), and glucose metabolism also used in the Como case, which proved how it was (PET) in diferent areas of the brain. Voxel-based mor- possible to make a defnite correlation between anoma- phometry (VBM) shows anatomical connections in the lies in certain sensitive areas of the brain and the young brain as well as the density of gray matter composed of murderer’s antisocial actions, particularly, in the wom- Visualizing Chemistry. Te Application of Chemical Imaging to Address Scientifc Challenges in Space Research 79 neurons and of white matter that instead forms axons. 4. P. Wu, J. Morie, P. Wall, T. Ott, K. Binsted, Procedia Using this technique, anomalous structures in the brain Eng. 2016, DOI 10.1016/j.proeng.2016.08.132 can be highlighted – something that proved valuable in 5. A.I. Skoog, I.P. Abramov, A.Y. Stoklistky, M.N. Doo- the Como court case discussed here. dnik, Acta Astronaut. 2002, DOI 10.1016/S0094- Te protocol then focuses on examining some of the 5765(02)00092-9 biomarkers genetically predictive of criminal behavior. 6. J.I. Trombka, J. Schweitzer, C. Selavka et al., Forensic For example, regarding the polymorphisms mentioned Sci Int. 2002, DOI 10.1016/S0379-0738(02)00079-8 here, the aspiring astronaut will have to have evaluated 7. V.S. Kokhan, M.I. Matveeva, A. Mikhametov, A.S. and monitored the monoaminergic system to determine Shtemberg, Neurosci Biobehav Rev. 2014, DOI the presence of the shorter MAO-A variant (on the X 10.1016/j.neubiorev.2016.10.006 chromosome) in the polymorphic version of MAOA-L 8. N. Kanas, Acta Astronaut. 2011, DOI 10.1016/j.act- as well as the catechol O-methyltransferase (COMT) to aastro.2010.04.012. confrm the Val158met polymorphism in the COMT. 9. K. Marshall-Bowman, M.R. Barrat, C.R. Gibso, Acta Regarding the serotonergic system, the SCL6A4 gene Astronaut. 2013, DOI 10.1016/j.actaastro.2013.01.014 must be monitored as it is able to reduce aggressive 10. A. Demertzi, Van Ombergen, E. Tomilovskaya, behavior in humans that codes for the serotonin trans- B. Jeurissen et al., Brain Struct Funct. 2016, DOI porter (SERT), a key modulator of serotonergic trans- 10.1007/s00429-015-1054-3 mission. Te possible detection of reduced serotonergic 11. I.A. Brazil, J.D.M. Van Dongen, J.H.R. Maes, R.B. activity in the brain may result in increased impulsive Mars, A.R. Baskin-Sommers, Neurosci Biobehav R. and aggressive behaviors via the “s” polymorphism of 2016, DOI 10.1016/j.neubiorev.2016.10.010 the SCL6A4 gene, whose existence involves a greater ina- 12. S. McSwiggan, B. Elger, P.S. Appelbaum, Int J Law bility to adapt in unfavorable environmental conditions. Psychiat. 2017, DOI 10.1016/j.ijlp.2016.09.005. Te dopaminergic system will also be subject to care- 13. Y.W. Chang, J.S. Chern, Acta Astronaut. 2016, DOI ful study as, for example, the SLC6A3 gene (dopamine 10.1016/j.actaastro.2016.06.008. transporter) has been linked to extremely violent impul- 14. Institute of Medicine and National Research Coun- sive-aggressive and antisocial behaviors in an individual cil, A risk reduction strategy for human exploration of whose SLC6A3 gene is found to have the 10 allele. space: A review of NASA’s Bioastronautics Roadmap. Te National Academies Press Washington DC, 2006. 15. NASA, Human research program requirements docu- CONCLUSIONS ment, National Aeronautics and Space Administra- tion Johnson Space Center, Huston, 2013, Revision F. Te ability to image the interior of the human body 16. NASA, NASA space fight human system standard, with techniques with chemical imaging has revolution- Vol. I: Crew Health, NASA-STD-3001, National Aer- ized medical diagnosis and treatment. Clearly it appears onautics and Space Administration Johnson Space evident that chemical imaging, forensic neuroradiol- Center, Houston, 2007. ogy and behavioral genetics are able to produce inno- 17. NASA, Astronaut candidate program, National Aero- vations and undoubted benefts to the delicate task of nautics and Space Administration Johnson Space understanding an individual’s true personality. In fact, Center, Houston, 2013. this research was able to show the validity of the appli- 18. K.J. Slack, T.J. Williams, J.S. Schneiderman, A.M. cation of chemical imaging to such an unusual forensic Whitmire, J.J. Picano, Evidence report: Risk of Adverse research feld as that of space exploration. Cognitive or Behavioral Conditions and Psychiatric Disorders, National Aeronautics and Space Adminis- tration Johnson Space Center, Houston, 2016. REFERENCES 19. E. Flynn-Evans, K. Gregory, L. Arsintescu, A. Whit- mire, Evidence report: Risk of Performance Decre- 1. S.J. Dick, NASA’s frst 50 years: historical perspectives. ments and Adverse Health Outcomes Resulting from U.S. Government Printing Ofce, Washington, 2009. Sleep Loss, Circadian Desynchronization, and Work 2. J. Kahn, C.T. Liverman, M.A. McCoy, Health stand- Overload, National Aeronautics and Space Admin- ards for long duration and exploration space fight, istration Lyndon B. Johnson Space Center, Houston National Academies Press, Washington, 2014. 2016. 3. N. Kanas, Acta Astronaut. 2014, DOI 10.1016/j.act- 20. L. Landon Blackwell, W.B. Vessey, J.D. Barrett, Evi- aastro.2014.06.026 dence report: Risk of Performance and Behavioral 80 Vincenzo Lusa, Annarita Franza

Health Decrements Due to Inadequate Coopera- Aspects, Lippincott-Raven, Philadelphia, 1999. tion, Coordination, Communication, and Psychoso- 40. R.L. Van Heertum, R.S. Tikofsky, M. Ichise, Function- cial Adaptation within a Team, National Aeronautics al Cerebral SPECT and PET Imaging, Lippincott Wil- and Space Administration Lyndon B. Johnson Space liams & Wilkins, Philadelphia, 2009. Center, Houston, 2016. 41. P.J. Uhlhaas, P. Liddie, D.E.J. Linden et al., Biol Psy- 21. S.P. Chappell, A.F. Abercromby, M.L. Gernhardt, Acta chiatry Cogn Neurosci Neuroimaging 2017, DOI Astronaut. 2013, DOI 10.2514/6.2013-3508. 10.1016/j.bpsc.2017.01.005 22. C. Strewe, B.E. Crucian, C.F. Sams, et al., Brain Behav 42. A.J. Reiss, K.A. Miczek, J.A. Roth, Understanding and Immun. 2015; 10.1016/j.bbi.2015.06.017. Preventing Violence: Volume 2. Biobehavioral infu- 23. J.W. Norbury, W. Schimmerling, T.C. Slaba, et al., Life ences. Panel on the Understanding and Control of Vio- Sci Space Res. 2016, DOI 10.1016/j.lssr.2016.02.001. lent Behavior, Committee on Law and Justice, National 24. L.A. Palinkas, P. Suedfeld P, Lancet. 2008, DOI Research Council, National Academy Press, Washing- 10.1016/S0140-6736(07)61056-3 ton DC, 1994, pp. 59-172. 25. Institute of Medicine, Safe passage: Astronauts care 43. H. Soderstrom, M. Tullberg, C. Wikkelsö et al., Psy- for exploration missions, National Academy Press, chiatry Res. 2000, 98, 29-41. Washington, DC, 2001. 44. M. Boccardi, R. Ganzola, R. Rossi et al., Hum Brain 26. M. Shapanek, Aviat Space Env Med. 2005,76, B25-30. Mapp. 2010, DOI 10.1002/hbm.20877 27. Institute of Medicine (US) Committee on Creating 45. P. Marchetti, Psicologia e Giustizia, 2012, 12, 1-14. a Vision for Space Medicine During Travel Beyond 46. Y. Yang, A. Raine, K.L. Narr et al., Arch Gen Psychiat. Earth Orbit, Safe Passage: Astronaut Care for Explora- 2009, DOI 10.1001/archgenpsychiatry.2009.110 tion Missions, National Academies Press, Washington 47. K.N. Ochsner, J.S. Beer, E.R. Robertson et al., Neuro- DC, 2001. image. 2005, DOI 10.1016/j.neuroimage.2005.06.069 28. K Cassady, V. Koppelmans, P. Reuter-Lorenz et 48. A. Raine, Y. Yang, Soc Cogn Afect Neurosci. 2006, al., Neuroimage. 2016, DOI 10.1016/j.neuroim- DOI 10.1093/scan/nsl033 age.2016.07.029 49. Raine A, Lencz T, Taylor K, Hellige JB, Bihrle S, Lac- 29. L.A. Palinkas, F. Glogower , M. Dembert et al., Int J asse L et al. Corpus callosum abnormalities in psy- Circumpolar Health. 2004, 63. 157-168. chopathic antisocial individuals. Arch Gen Psychiatry 30. J. Brcic, Acta Astronaut. 2010, DOI 10.1016/j.actaas- 2003;60(1):1134-42. tro.2010.06.044 50. O.D. Jones, J.D. Schall, F.X. Shen, Law and Neuro- 31. L.S. Bishop, R. Kobrick, M. Battler, Acta Astronaut. science, Wolters Kluwer Law & Busines, New York, 2010, DOI 10.1016/j.actaastro.2009.11.008 2014. 32. N. Kanas, D. Manzey, Space Psychology and Psychia- 51. D. Patterson, M.S. Pardo, Philosophical Foundations try, El Segundo California and Springer, Dordrecht, of Law and Neuroscience, Oxford University Press, 2008. New York, 2016, pp. 69-83. 33. G.G. De La Torre, B. Van Baarsen, F. Ferlazzo, et 52. A.R. Beech, A.J. Carter, R.E. Mann, P. Rotshtein, Te al., Acta Astronaut. 2012, DOI 10.1016/j.actaas- Wiley Blackwell Handbook of Forensic Neuroscience, tro.2012.08.013. Wiley-Blackwell, New York, 2018. 34. S. Schneider, V. Brummer, H. Carnahan et al., Neuro- 53. M.J. Saks, N.J. Schweitzer, E. Aharoni, K.A. Kiehl, J image. 2008, DOI 10.1016/j.neuroimage.2008.06.010 Empirical Legal Stud. 2014, DOI 10.1111/jels.12036 35. F.A. Elliot, Arch Neurol. 1992, 49, 595-603. 54. N.A. Farahany, J Law Biosci. 2016, DOI 10.1093/jlb/ 36. J.W. Britton, L.C. Frey, J.L. Hopp et al., Electroenceph- lsv059 alography (EEG): An Introductory Text and Atlas of 55. L. De Cataldo Neurburger L., La prova scientifca nel Normal and Abnormal Findings in Adults, Children, processo penale, Cedam, Padova, 2003, pp. 317-34. and Infants, American Epilepsy Society, Chicago, 2016. 37. A.C. Mamourian, CT Imaging. Practical Physics, Arti- facts and Pitfalls, Oxford University Press, New York, 2013. 38. S. Ulmer, O. Jansen, fMRI. Basics and Clinical Appli- cations, Springer, Berlin, 2013. 39. G.J. Siegel, B.W. Agranof, R.W. Albers et al., Basic Neurochemistry: Molecular, Cellular and Medical Firenze University Press www.fupress.com/substantia

Research Article From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the Citation: A.M. Papini (2018) From morphine to endogenous opioid pep- perfect painkiller tides, e.g., endorphins: the endless quest for the perfect painkiller. Sub- stantia 2(2): 81-91. doi: 10.13128/substantia-63 Anna Maria Papini Copyright: © 2018 A.M. Papini. This is Interdepartmental Laboratory of Peptide and Protein Chemistry and Biology, Department an open access, peer-reviewed article of Chemistry “Ugo Schif”, University of Florence, Via della Lastruccia 13, I-50019 Sesto published by Firenze University Press Fiorentino (Italy) (http://www.fupress.com/substantia) Email: [email protected] and distribuited under the terms of the Creative Commons Attribution License, which permits unrestricted use, distri- Abstract. Opium was known since the Neolithic era and in 5th century wild Papa- bution, and reproduction in any medi- ver use was reported to induce sleep and relieving pain. First active component iso- um, provided the original author and lated from Opium was morphine, the paradigm of a natural product discovered 150 source are credited. years before isolation of endogenous opioid ligands, brain pentapeptide enkephalins. Data Availability Statement: All rel- Since then many endorphin peptides and their mode of action were discovered. Native evant data are within the paper and its endorphins were characterized thanks to the synthetic antagonist naloxone. Supporting Information fles. Keywords. Opium, morphine, peptides, peptidomimetics, analgesics. Competing Interests: The Author(s) declare(s) no confict of interest.

MORPHINE: A PARADIGMATIC EXAMPLE OF A NATURAL PRODUCT MIMICKING ENDOGENOUS MOLECULES

Te story of morphine (Figure 1) and its analogues is a paradigmatic example of the classical pathway to drug development undertaken by those researchers starting from a natural compound. In particular the phases of development and optimisation of morphine-like drugs are exemplifed as follows: 1. Recognition of the pharmacological activity of a plant 2. Extraction and identifcation of the active ingredient 3. Studies on synthesis (partial and total) 4. Structure-activity relationship studies (through synthetic analogues) 5. Development of analogues as drugs to optimize activity and decrease side efects 6. Receptor theories; ‘rational’ synthesis of analogues/structure-based design Morphine was isolated for the frst time from the opium poppy by Frie- drich Wilhelm Adam Sertürner (Figure 2). In a letter to the editor of the Journal Der Pharmacie Für Ärzte Und Apotheker (Vol. 13, 1805), he reported on the isolation of a substance from opium with alkaline character. In 1806, Sertürner moved to Einbeck, Ger-

Substantia. An International Journal of the History of Chemistry 2(2): 81-91, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-63 82 Anna Maria Papini

N N H (R) H (S)

(S) H (R)

(R) (S) (S) (R) (R) (S)

HO O H OH HO O H OH (+)-Morphine (-)-Morphine

Figure 1. Structures of the bioactive (-)-Morphine and of the inactive (+)-Morphine.

many where he was assistant to the tenant of the mag- istrate’s pharmacy. In 1809, he became pharmacist and, Figure 2. Friedrich Wilhelm Adam Sertürner (1783-1841) who since the tenant was already 75 years old, he intended to isolated the substance from opium with alkaline character that he take charge of the pharmacy. However, he was not suc- called “Morphine”, a name with a clear mythological reference: the cessful. During the invasion of Napoleon Bonaparte’s Greek God of dreams Morpheus. troops into Germany, French legislation became the law of the land in those parts, which fell under French government and Sertürner was allowed to open a second use of wild poppy Papaver agreste for inducing sleep and pharmacy. Terefore Sertürner continued his research relieving pain. work on morphine in Einbeck and published the results Te Persian Abu Bakr Muhammad ibn Zakariyya in two papers. In one of these (1817), he reported his al-Razi (865-925 AD; 251-313 AH) also known as Rhazes observations on the use of a drug in humans that he because of the place where he was born and died, i.e., called for the frst time “morphine”, name with a clear Rayy, near Teheran. He studied medicine in Baghdad mythological reference. Te French chemist Gay-Lussac and became one of the greatest physicians of the medi- was interested in Sertürner’s publication and ordered a eval period, writing over 200 works; half of them on French translation, which earned Sertürner the scientifc medicine, but others on topics that included philosophy, 5 breakthrough of morphine. He is recognized the pioneer theology, mathematics, astronomy and alchemy. He in alkaloid research, and for that he received a doctor made use of opium in anesthesia and in “In the Absence degree from the university of Jena in 1817 when he pub- of a Physician” (a home medical manual directed toward lished the isolation of pure morphine from opium afer ordinary citizens for self-treatment), he recommended at least thirteen years of research and a nearly disastrous the use of opium for treatment of melancholy. All the trial on himself and three teenagers, leading to pain in leading physicians of Baghdad used opium that was con- the region of the stomach, exhaustion, and severe narco- sidered particularly efective for diseases of the intestines sis that came close to fainting as described in details by and of the eyes, but it also featured in a number of rem- Sertürner.1-4 edies to treat gout and painful joints. Al-Razi gave recipes for gout and the joints based on ointments that were applied to the painful areas with a damp piece of paper HISTORICAL BACKGROUND or cloth to keep the medication moist. A good paste that al-Razi described contained equal parts of opium and 6 Te poppy and possibly its bioactivity was known liquid storax (Liquidambar orientalis). Te renowned since the Neolithic era, since seeds were found in tombs Andulasian opthamologic surgeon Abu al-Qasim al- dating to 4200 BC. It was certainly cultivated in Meso- Zahrawi (“Abulcasis”, 936–1013 AD) relied on opium potamia, Persia, India and China and widely used as as a surgical anaesthetics and wrote a treatise, al-Tasrif, sleeping or sedative remedies, but also used in religious that infuenced medical thought well into the 16th cen- and spiritual rituals. tury. Te fve-volume De Materia Medica written by Ped- In 1527 the Swiss physician, alchemist and astrolo- anius Dioscorides, remained in use from the 1st to the ger Philippus Aureolus Teophrastus Bombastus von 16th centuries, described opium and the wide range of Hohenheim (1493-1541) who called himself Paracelsus, its uses. In the 5th century Pseudo-Apuleius refers to the introduced to Western medicine Laudanum or opium tincture returning from Arabia with a famous sword, From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the perfect painkiller 83 within the ball of which he kept “Stones of Immortality” Laudanum became the basis of many popular pat- composed of opium thebaicum, citrus juice, and “quin- ented medicines of the 19th century (Figure 4). tessence of gold”. Te name Laudanum was invented by Te English physician Tomas Dover (1660-1742) Paracelsus from the latin “laudare” or was a corrupted was the frst to market in England a powder, the Dover’s form of ladanum (from the Persian ladan), a resinous powder, a preparation of opium and ipecacuanha, the juice or gum obtained from various kinds of the Cis- later was added for its emetic properties to limit its use.8 tus shrub (by M. Ray, Editor of Encyclopaedia Britan- Its recreational use was therefore widespread, Godfrey’s nica, 2017). Te term is used now to describe the alco- Cordial was sold freely,9 and opium was freely imported holic tincture of opium, a 10% solution of opium powder from India, like tea or tobacco. dissolved in high-proof distilled spirits. It was used as In the 17th century in China the use of smoking an analgesic substance and Paracelsus understood that opium was widely spread, also because of the prohibition opium was more soluble in alcohol and reported the frst of tobacco (1644); consumption was very high and opi- evidence of dependence. Laudanum was a major part of um was imported from India via Canton by English and the pharmacopeia into the 20th century. It was a com- American merchants. Te blockade of the importation mon drug of abuse during the Victorian era. Paracelsus by Chinese authorities caused the Opium War (1839- considered himself an alchemist and his ideas were not 1842). Interestingly in 1841, the US president William always well accepted by the medical community. Howev- Henry Harrison was treated with laudanum.10 Moreover, er he was the frst to introduce chemistry into medicine in the American Civil War, the Union Army used 2.8 in the 16th century. Most of his work was published only million ounces of opium tincture and powder and about afer his death (Figure 3) and Peder Sorensen in 1571 in 500,000 opium pills.11 “Idea medicinæ philosophicae” started emphasizing Par- acelsus’s pioneering work in Chemical Medicine.7

Figure 3. Cover of the Labyrinthus Medicorum Errantium by D. Figure 4. Label of Laudanum bottle prepared by Chas. Hooper & Teophrasti Paracelsi. Sons, Chemists and Druggists, London. 84 Anna Maria Papini

OPIOIDS AND OPIATES

Opium is the latex or rubber obtained by etching the immature capsules of Papaver somniferum (Figure 5). It contains various alkaloids with analgesic action, of which the most relevant one is morphine. The term opiate (widely used until the 1980s) describes any natural or synthetic agent derived from morphine. In 1833 MacFarlane prepared morphine on a commercial scale,12 and in 1853 injectable morphine comes into use during the American Civil War.13 Noteworthy is that in 1898, Bayer registered diamorphine (diacetylmor- phine) the name of heroin in Germany as an antitus- sive (cough suppressing) drug.14 Te Harrison Narcotics Act, which was passed in 1914 and took efect in 1915 marked the beginning of federal narcotics control in USA. Tis act aimed to control each phase of production and distribution of opium, morphine, heroin and any new derivatives that could have similar biological activity. Te frst drug prohibition federal law in USA was the Smoking Opium Exclusion Act. It passed in 1909 and prohibited the importation of opium prepared for smoking in the United States.15

FROM MORPHINE TO CODEINE AND TO SEMI- SYNTHETIC ANALOGUES VIA THEBAINE

Morphine structures Figure 5. Papaver somniferum. Morphine is an opium phenantrenic alkaloid with 5 stereogenic centers (*), and therefore with the theo- almonds, cafeine, etc.19 He reported that he was com- retical possibility of presenting 25 = 32 stereoisomers. missioned by the “Société de Pharmacie” to examine Practically, geometric restrictions limit the possibilities the procedure to extract morphine that was proposed to 16 stereoisomers. Te natural bioactive enantiomer is by William Gregory (1803-1838) in Edinburgh and that (-)-Morphine (5R, 6S, 9R, 13S, 14R) (Fig. 1). Isomorphine during his routine work he discovered codeine as a pow- is the epimer in which the absolute confguration at C-6 der crystallized afer evaporation of the mother liquor is R (hydroxyl in position b). Te synthetic enantiomer lef afer treatment with KOH and washing with water. of (-)-Morphine, the (+)-Morphine (Fig. 1) has about Terefore, for the frst time he discovered that morphine 10,000 times less afnity than the natural (-)-Morphine was not the only active ingredient in opium. He named and possesses no functional efcacy when tested at con- the new ingredient codeine and M. Kunckel demonstrat- centrations up to 100 fold the efective dose of natural ed its strong action on the spinal cord and that it did not morphine. paralyze the back parts as morphine did.20 In 1925 Robinson and Gulland determined for the Tebaine (Figure 7) is a minor constituent of opium frst time its structure16 and only more than 25 years lat- similar to morphine and codeine but with a weak anal- er, i.e. in 1952, Gates performed its frst chemical synthe- gesic action. Its signifcance comes essentially from its sis.17 It took additional 20 years to univocally determine industrial use as the starting material to produce semi- morphine X-ray structure.18 synthetic drugs such as codeine but also the opiate In 1832 Codeine (Figure 6) was isolated from opium antagonist naloxone (see below). and characterized by Pierre-Jean Robiquet, a French Codeine is none other than methyl-morphine. pharmacist who discovered other important natu- Codeine itself is oxidized into codeinone, and the ral substances such as asparagine, amygdaline in bitter methyl ether of the enol form of codeinone is thebaine. From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the perfect painkiller 85

16 N 10 9 H 1 8 11 15 H 14 12 2 7 13 5 6 3 4 HO O H OR

Figure 8. Heroin. Adapted from Ref. 21.

Figure 6. R = H: Morphine; R = CH3: Codeine. Adapted from Ref. 21.

OCH3 H N H C 3 O

OCH3

Figure 7. Tebaine or paramorphine (0.3%) weak analgesic action. Adapted from Ref. 21.

It was not until 1927 that a compound identifed as the dimethyl ether of morphine was fnally isolated from the mixture of the products of hydrogenation of opium.22 To conclude this enumeration of the most striking chemical interrelations between morphine, codeine and thebaine, it should be recalled that, as Knorr demonstrated in 1909,23 treatment with acids transforms thebaine into codeinone. In 1874 heroin was prepared as first example of a semi-synthetic opioid by the English chemist and physicist, C.R.A. Wright, at St. Mary’s Hospital Medi- cal School in London.24 Wright synthesized heroin (dia- cetylmorphine) afer mixing and simmering morphine with acetic anhydride (Figure 8). Heroin displayed 5-fold the analgesic activity of morphine. Figure 9. Advertisement for Bayer Heroin: the sedative for coughs. First tests of heroin were conducted on dogs and rabbits showed severe side-efects and C. R. A. Wright stopped the experiments. However, in 1897 heroin was rediscovered by Felix Hoffmann at the Bayer pharmaceutical company in Elberfeld (Germany) acetylat- ing morphine with the objective of producing codeine. Terefore heroin, the same compound discovered by Wright was not patentable. Before the extreme addictive- ness of heroin was recognized, from 1898 to 1910 heroin was marketed by Bayer as a non-addictive morphine substitute and cough medicine for children,25 to prepare Figure 10. Structure-Activity Relationship: modifcations in the patients for anesthesia, and to control certain mental morphine structure and analgesic efect. Morphine = 100 as a refer- disorders (Figure 9). ence. A range of synthetic opioids such as methadone (1937), pethidine (1939), fentanyl (late 1950s), and derivatives thereof have been introduced, and were targeted for certain specialized applications. 86 Anna Maria Papini

Morphine pharmacological profle

Morphine is the most studied molecule of natural origin in the last two hundred years, with the aim of discovering a central analgesic orally active molecule, free of side efects and not addictive. Te main efects of morphine are: analgesia, euphoria Figure 11. Cyclazocine. and dysphoria (psychological distress), sedation, respiratory depression (the frst cause of death due to morphine overdose), depression of cough refexes, nausea and vomit, debated in a symposium sponsored by the National physical and psychological dependence, miosis (constric- Institute of Mental Health and the Department of Psy- tion of the pupil), constipation (reduction of intestinal chiatry, New York Medical College. Many molecules motility), spasms of the biliary tract, stimuli and difcul- were tested and Cyclazocine apperared to be the most ties in urination, stimulation of histamine release with promising one (Figure 11). consequent vasodilation, bronchial constriction, redness Cyclazocine was found to be a clinically efective and itching, efects on the endocrine system (decreased and protracted opiate antagonist whose efect lasted for libido, impotence, amenorrhea) and immunosuppression. at least 24h. In addition it exhibited unpleasant initial Morphine can be administered orally (the analgesic side-efects including dizziness, headaches and halluci- potency is reduced to about 5-30% of that obtained by nations that were disproportionately intensifed as the parenteral administration); subcutaneously and intra- dose was increased and reappeared when it was discon- muscolarly absorption is efective with the inconven- tinued. It was concluded by the review of the clinical ience of tissue irritation; intravenous administration via data that an ideal antagonist would be one exhibiting slow infusion is preferred for better analgesic coverage antagonistic efcacy for weeks or months, without ago- and reduction in the risk of overdose. nistic activity.26 Te antalgic therapy uses drugs belonging to dif- Eforts to avoid these side-efects led clinicians to ferent classes: non-steroidal anti-inflammatory drugs the use of the frst “pure” antagonist: naloxone (Figure (NSAIDs), opioids (morphine-like) and local anesthetics. 12). Originally synthesized in the private laboratory of Local anesthetics (e.g. lidocaine) clearly have a mecha- Mozes Judah Lewenstein and subsequently developed by nism of action external to the Central Nervous System Endo Laboratories, Garden City, Naloxone has no phar- (CNS), as they afect transient pharmacological block- macological properties of its own but it abolishes or pre- ade of nerve conduction from peripheral receptor sites. vents the hallucinations, euphoria, respiratory depres- Te assumptions underlying the historical distinction sion, nausea, convulsions and other efects produced by between NSAIDs, endowed with peripheral activity, and narcotics. It can also abolish these efects when they are opioids, active at the CNS, no longer seem to be justi- produced by other antagonists. Naloxone is synthesized fed. In fact, the central efects of some NSAIDs have from thebaine, which explains its high cost.27 been demonstrated, as well as the existence of peripheral Naloxone is a specifc opiate antagonist that has opioid receptors. no residual agonist activity. It is a competitive antagonist at the diferent receptors: μ, δ, and k opioid receptors above described. It was introduced in 1973 and is FROM AGONIST TO ANTAGONIST ACTIVITY: ON used to inhibit the efect of narcotics on the CNS (see THE WAY TO THE DISCOVERY OF ENDORPHINS below). AND OPIOID RECEPTORS

Observing the relationship between structure and analgesic activity not only in diferent opium compo- OH H X nents but also in the semi-synthetic analogues derived N from morphine (Fig. 9), it is reasonable to think that a R natural plant originated product such as morphine could O be further optimized for interaction with a mammalian opiate receptor (for a long time unknown) for which it is OH not the natural ligand. Interestingly on June 4, 1970 the use of narcotic Figure 12. Morphine agonist Oximorphone and antagonist Nalox- antagonists in the treatment of heroin addiction was one. Oximorphone (X = CO, R = CH3) agonist. Naloxone (X = CO, R = CH2-CH=CH2) antagonist used in opiates overdose. From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the perfect painkiller 87

Narcan, an injectable form of naloxone is currently used to fght the opioid epidemic and substance abuse to reverse drug overdose and addiction.28

FROM MORPHINE ANTAGONIST NALOXONE TO ENDOGENOUS OPIOID PEPTIDES: THE DISCOVERY OF ENDORPHINS Figure 13. Similarities of the pharmacophoric feature in bold of morphine (A) and [Met5]enkephalin (B). Afer the advent of the pioneering work of Robert B. Merrifeld,29 who introduced the solid-phase synthesis of peptides, the easiest pathway to develop drugs acting Interestingly, only one single C-terminal amino acid at peptidergic receptors is based on characterization of residue is the diference between [Met5]enkephalin and endogenous bioactive peptides that can be synthesized [Leu5]enkephalin. Both peptides induce in vivo a deep and subsequently structurally stabilized to increase in analgesia in rat (completely antagonized by naloxone), vivo half-life, limiting rapid metabolism and excretion but the activity is short-lived, because of a fast degrada- and fne tuning their biological properties. tion by blood and cerebral peptidases. [Met5]enkephalin However, this general trend was not observed in has about 30% of the morphine potency and is about the case of morphine, since evidences of the existence 3-fold more potent than [Leu5]enkephalin. of endogenous substances with morphine-like activ- Te N-terminal tyrosine residue was found to be the ity (endorphins) were obtained only in the 70’s, afer main pharmacophoric determinant of opioid peptides, decades of use of morphine and its derivatives for both a structural feature shared also by morphine and ana- recreation and therapeutic applications, extensively logues (Figure 13). In fact, this portion is strictly main- described above. Of particular significance was the tained by all the brain opioid peptides discovered subse- observation that naloxone was able to antagonize the quently (Figure 14). analgesia induced by electrical stimulation of specifc Te brain opioid peptides discovered in the ‘70s led areas of the brain (gray periaqueductal area). Te emerg- to the implementation of a new nomenclature. In fact, ing hypothesis suggests that under stress conditions the endogenous peptides were initially considered not (electrical stimuli), the release of endogenous substances related to morphine from a structural point of view, but with an activity profle similar to that of morphine is their pharmacological actions are similar to those of activated. Tis is the paradigmatic example that exog- morphine, as they are ligands of the same receptors. enous natural products are able to mimic the activity of Te term opioid has over the years been used to endogenous molecules. indicate a substance that is pharmacologically similar In 1975, Hans W. Kosterlitz and his former student to opium or to morphine, both of an endogenous nature John Hughes discovered Enkephalins, natural ligands and of a synthetic or semi-synthetic nature. Enkephalins for opiate receptors that were characterized as the pentapeptides H–Tyr–Gly–Gly–Phe–Met–OH and H– Tyr–Gly–Gly–Phe–Leu–OH (Figure 13). Te structure was elucidated by the determination of the amino acid 5 sequence of natural enkephalins by the dansyl–Edman [Leu ]enkephalin Tyr-Gly-Gly-Phe-Leu [Met5]enkephalin Tyr-Gly-Gly-Phe-Met procedure and mass spectrometry and followed by syn- Dinorphin A(1-17) Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg- thesis and demonstrating the complete chemical and Pro-Lys-Leu-Lys-Trp-Asp-Asn-Gln biological equivalence of the natural and synthetic pep- Dinorphin B(1-8) Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile tides. Tese morphine-like peptides that can be antago- Dinorphin (1-13) Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Gln-Phe- nized by opiate antagonists, such as naloxone, are natu- Lys-Val-Val-Tr rally occurring substances in the brain, which afect how α-Neoendorphin Tyr-Gly-Gly-Phe-Leu-Arg-Lys-Tyr-Pro- 30 Lys we feel pleasure and help fght pain. Independently β-Neoendorphin Tyr-Gly-Gly-Phe-Leu-Arg-Lys-Tyr-Pro Solomon H. Snyder identifed the same two peptides in βh-Endorphin Tyr-Gly-Gly-Phe-Met-Tr-Ser-Glu-Lys- bovine brain.31-33 Kosterlitz, Hughes and Snyder shared Ser-Gln-Tr-Pro-Leu-Val-Tr-Leu- Phe- the prestigious Albert Lasker Prize in 1978 for this Lys-Asn-Ala-Ile-Ile-Lys-Asn-Ala-Tyr- research that paved the way for the development of new Lys-Lys-Gly-Gln kinds of nonaddictive analgesic. Figure 14. Endogenous Opioid Peptides: Endorphins.34 88 Anna Maria Papini and other brain peptides discovered later are collectively such a selective agonist by modifcation of morphine known as endorphins (Figure 13). failed, simply because each receptor subtype mediates Endorphins are among the brain chemicals known many diferent biological efects. as neurotransmitters, which are molecules inducing electrical signals from one neuron to the subsequent within the nervous system. At least 20 types of endorphins have κ Opioid receptors been discovered in humans. Endorphins can be found in Te hypothesis that suggested their existence was the pituitary gland, in other parts of the brain, or dis- 36 tributed throughout the nervous system. Stress and pain advanced by Martin, and was based on the diferent are the two most common factors leading to the release efects produced by Morphine and some structurally- of endorphins. Endorphins bind and interact with the related analogues (for example, ketociclazocine). Tey opiate receptors in the brain leading to nociception, a have been so named to highlight the ketonic nature of perception of analegesia. As such they act similarly to the compounds that activate them selectively (in par- drugs such as morphine and codeine. In contrast to the ticular EKC, ethyl ketociclazocine). opiate drugs, however, activation of the opiate receptors Tey play an important role in receiving and pro- by the body’s endorphins does not lead to addiction or cessing the primary aferent pain information. In the tolerance. brain they integrate the ascending pain information and inhibit the painful sensations that descend to the spinal cord. Also located in the limbic system and in the cer- THE CHARACTERIZATION OF OPIOID RECEPTORS ebral cortex, they are involved in afective and emotional states and in the awareness of analgesia. Identifcation in the 70’s of endorphins and in par- Function of receptor κ. They exert a modulation ticular enkephalins, as the endogenous substances with role of the processes of: analgesia, diuresis, hypother- morphine-like activity enabled the search and character- mia, neuroendocrine secretions, feeding (activation of ization of the opioid receptors. Terefore it was possible k receptors causes an increase in appetite, inhibited by to classify the opioid receptors into three types, called μ, nor-BNI, a selective k antagonist. δ, κ.35 Te genes of these receptors were cloned and the κ-Agonists. In various animal models, the κ-agonists relative transcripts showed similarities of more than 60% cause sedation at lower concentrations than the of the nucleotide sequence. μ-agonists. Te increase in diuresis, induced in a charac- All opioid receptors belong to the superfamily of G teristic way by the κ-agonists, is consequent to the inhi- protein coupled receptors (GPCR), whose α-subunits are bition of the release of VP from the neurohypophysis. of the Gi/0 type. Using these transductional couplings, Te withdrawal symptoms from κ-agonists are dif- opioid receptors control the activity of efectors such as ferent and less severe than those that occur with ago- adenylate cyclase (inhibition) and some ion channels nists μ. A further advantage of their use is linked to the (Ca2+ and K+). absence of respiratory depression and constipation. A disadvantage, found when κ-agonists are administered, is the occurrence of dysphoric and psychotomi- μ Opioid receptors metic efects in humans. Tis is due to a lack of selectivity, as many of these substances also interact with Te μ receptors are the most widespread and abun- δ-receptors. dant and mediate most of the opioid analgesic efects. Te use of κ-agonists with arylacetamide structure Morphine and naloxone have a weak μ selectivity. How- induces neuro-protection from cerebral ischemic damages. ever, μ opioid receptors are located also outside the The endogenous ligands of the κ-receptor belong CNS, in numerous intramural nerve plexes (gastro- to the dynorphin and neoendorphin families. Te frst intestinal tract, biliary tract, urogenital pathways, circu- peptide to be isolated from the pig’s pituitary gland latory and respiratory systems). Accordingly, it was not was Dinorphin – (1-13) in 1979;37 in 1982 the Dynor- possible to confrm the elegant hypothesis, that a single phin-(1-17) was isolated,38 and the neoendorphin was opioid receptor subtype could be the unique mediator frst isolated in 1979 from the porcine hypothalamus but of “pure” analgesic activity. Based on this hypothesis, a an incorrect sequence was assigned.39 Te sequence was selective agonist of this receptor subtype would be the then assigned corrected in 1981 by the same research “perfect” analgesic, devoid of side-efects, induced by group.40-41 activation of diferent receptor subtypes by non-selective compounds. In fact, several attempts to develop From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the perfect painkiller 89

δ Opioid receptors

Te distribution of δ opioid receptors (studied by autoradiographic techniques with tritiated and iodinat- ed radioligands) at the CNS level is pre-eminent in the more evolutionary younger cerebral structures (olfactory bulb, caudate nucleus, neopallio, putamen), while it is relatively poor in midbrain and in the medulla oblon- gata. Following the observation of their almost total Figure 15. General scheme of peptidergic synapses.43 absence, for example in reptiles and birds, this receptor subtype would have developed fully later than the other opioid receptors. Function of δ Receptor. δ Opioid receptors play a role in regulating the processes of analgesia, motor coordination, intestinal motility, smell, respiration, cognitive function, emotional state.

OPIOID-RECEPTOR-LIKE

In addition to the three classic opioid receptors (μ, δ, κ), a new opioid receptor named Opioid-Receptor- Figure 16. Peptidomimetic lead compound concomitantly acting as 44 Like (ORL) was discovered in 1994.42 It is also coupled μ-opioid receptor agonist and δ-opioid receptor antagonist. to G proteins and has a high sequence homology (> 60%) compared to μ, δ, κ. However, the typical opioid ligands (peptides and non-peptides) do not bind to ORL. It is are excited. (6) Endogenous opioid peptides bind to the present in all brain regions and in the spinal cord. It is postsynaptic receptor, activating the inhibitory G pro- located in the intestine, in the vas deferens, in the liver tein (Gi) that induces inactivation of adenylyl cyclase, and in the spleen. It was not found in skeletal muscles, in inhibiting release of cAMP and (7) infuence the infux the esophagus, in the kidneys and in the adrenal glands. of potassium ions through the cell membrane. Te over- An endogenous agonist characterized to be an epta- all efect is hyperpolarization of postsynaptic neuron decapeptide structure (see comparison with Dinorphin and inhibition of cell excitation. (8) Exogenous opioids A) called Orphanin-FQ (OFQ) or Nociceptin (NC) has (Op) such as morphine bind to the opioid receptors and been identifed. simulate the action of (E). (9) Opioid antagonists such OFQ (or NC) is generated by pro-Orphanin-FQ (or as naloxone (Nx) bind to receptors and competitively pro-Nociceptin). Te pharmacological profle of NC has inhibit the action of (E) and (Op). (10) Te action of (E) not yet been fully defned, but NC has analgesic activity. is interrupted by a membrane-bound peptidase, which OFQ: Phe-Gly-Gly-Phe-Thr-Gly-Ala-Arg-Lys-Ser- hydrolyzes the peptide bond Gly3-Tyr4 in enkephalin Ala-Arg-Lys-Leu-Ala-Asn-Gln and leads to its inactivation. DinA: Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg-Pro- Lys-Leu-Lys-Trp-Asp-Asn-Gln TOWARD THE PERFECT PAINKILLER: DUAL ACTIVITY PEPTIDOMIMETICS SYNAPSIS AND ENDORPHINS RELEASE Pharmacological studies suggest that the δ-opioid receptor plays a key role in modulating some side efects As depitected in the general scheme reported in Fig- associated with opioids including analgesic tolerance ure 15, pro-opioid proteins are synthesized in the cell and physical dependence. nucleus (1) and are transported by microtubular system In fact coadministration of δ-opioid receptor antag- (2) to the nerve terminal. (3) Te active endogenous opi- onist with morphine attenuate analgesic tolerance, phys- oids (E) are released from the pro-opioid proteins by ical dependence and drug-seeking behavior. Accordingly the “process” proteins that are specifc proteinases, (4) recent studies aimed to develop molecules concomitantly they are transported and stored in the presynaptic vesi- acting as μ-opioid receptor agonist and δ-opioid receptor cles, and (5) are released when the presynaptic neurons 90 Anna Maria Papini antagonist. In particular recently Mosberg et al. reported 10. J. McHugh, P. A. Mackowiak, Death in the White a family of peptidomimetics (Figure 16) producing long- House: President William Henry Harrison’s Atypi- lasting and dose dependent antinociception in mice afer cal Pneumonia. Clinical infectious diseases 2014, 59, peripheral administration.44 990-995. Te bioavailable molecules recently described are 11. H. Duthel. Illegal Drug Trade: Te War on Drugs, promising leads in the search for future drug candidates CreateSpace Independent Publishing Platform, Scotts endowed with dual activity as μ-receptor agonist/δ- Valley (CA), 2011, p. 1-450. receptor antagonist. Tese novel peptidomimetics can 12. G. L. Patrick. An Introduction to medicinal chemis- represent the long sought painkillers. try, Oxford University Press, Oxford, 2001, p. 1-620. 13. G. R. Hanson, P. J. Venturelli, A. E. Fleckenstein. Drugs and Society, 13th Ed., Jones & Bartlett Learn- ACKNOWLEDGMENTS ing, Burlington (MA), 2018, p. 1-693. 14. D. F. Musto, P. Korsmeyer, T. W. Maulucci. One hun- Fulvio Gualtieri, Emeritus Professor of Medicinal dred years of Heroin, Auburn House, Westport (CT), Chemistry, inspired this review that does not have the 2002, p. 1-252. ambition to be exhaustive. Moreover Paolo Rovero and 15. Drugs in American Society. An encyclopedia of his- Mario Chelli are gratefully acknowledged for fruitful tory, politics, culture, and the law (Eds.: N. E. Mari- discussion. on, W. M. Oliver), ABC-Clio, 2014, p. 1-1163. 16. J. M. Gulland, R. Robinson. Constitution of codeine and thebaine. Memoirs of the Literary and Philo- REFERENCES sophical Society of Manchester 1925, 69, 79–86. 17. M. Gates, G. Tschudi. Te synthesis of morphine. 1. F. Sertürner, Journal der Pharmacie fuer Aerzte und Journal of the American Chemical Society 1952, 74, Apotheker 1805, 13, 229-243. 1109–1110. 2. F. Sertürner, Darstellung der reinen Mohnsäure (Opi- 18. L. Gilbert. Te crystal and molecular structure of umsäure) nebst einer Chemischen Untersuchung morphine hydrochloride trihydrate. Acta Crystallo- des Opiums mit vorzüglicher Hinsicht auf einen- graphica 1973, B29, 1630-1635. darin neu entdeckten Stof und die dahin gehörigen 19. C. Warolin. Pierre-Jean Robiquet (Rennes, 14 janvi- Bemerkungen. Journal der Pharmacie fuer Aerzte er 1780-Paris, 29 Avril 1840). Revue d’Histoire de la und Apotheker 1806, 14, 47-93. Pharmacie 1999, 321, 97-110 article in French trans- 3. F. Sertürner, Ueber das Morphium, eine neue lated by J. Wisniak. Educación Química 2013, 24 salzfähige Grundlage, und die Mekonsäure, als (E1), 139-149. Hauptbestandtheile des Opiums. Annalen der Physik 20. P.-J. Robiquet. Nouvelles Observations sur les Princi - 1817, 55, 56-89. paux Produits de l’Opium, Annales Chimie Physique 4. I. Jurna, Sertürner and morphine - a historical 1832, 51, 225-267. vignette. Schmerz 2003, 17, 280-283. 21. J.H. Schrittwieser, V. Resch. Te role of biocataly- 5. S. Tibi, Al-Razi and Islamic medicine in the 9th cen- sis in the asymmetric synthesis of alkaloids, RSC tury. Journal of the Royal Society of Medicine 2006, Advances 2013, 3, 17602-17632. 99, 206–207. 22. C. Schöpf. Die Konstitution der Morphiumalkaloide. 6. S. Tibi, Te Medicinal Use Of Opium In Ninth-Cen- Justus Liebig Annalen Der Chemie 1927, 452, 211- tury Baghdad, Brill, Leiden, 2005. 267. 7. A. G. Debus, Paracelsus and the medical revolution 23. L. Knorr. Zur Kenntniss des Morphins. VII. Mit- of the Renaissance: A 500th Anniversary Celebration, theilung (mit Heinrich Hörlein): Ueberführung des NIH U.S. National Library of Medicine, Bethesda, Thebaïns in Codeïnon und Codeïn. Berichte der 2011. deutschen chemischen Gesellschaf 1906, 39, 1409- 8. D. N. Phear, Tomas Dover 1662-1742; physician, 1414. privateering captain, and inventor of Dover’s powder, 24. C. R. A. Wright. On the Action of Organic Acids and Journal of the History of Medicine and Allied Scienc- Teir Anhydrides on the Natural Alkaloids. Journal es 1954, 9, 139-156. of the Chemical Society (London) 1874, 27, 1031- 9. T. E. Jordan, Te keys of paradise: Godfrey’s Cordial 1043. and children in Victorian Britain. Journal of the Roy- 25. Bayer Heroin Hydrochloride Advertisement, Ameri- al Society of Health 1987, 107, 19-22. can Journal of Physiology 1901, 7. From morphine to endogenous opioid peptides, e.g., endorphins: the endless quest for the perfect painkiller 91

26. M. Fink. Narcotic antagonists in opiate dependence. 37. A. Goldstein, S. Tachibana, L. I. Lowney, M. Hunka- Science 1970, 169, 1005-1006. piller, L. Hood. Dynorphin-(1-13), an extraordinarily 27. T. H. Maugh II. Narcotic Antagonists: Te Search potent opioid peptide. Proceedings of the National Accelerates. Science 1972, 177, 249-250. Academy of Sciences of the United States of America 28. M. Salaun. Te New York Times, Oct 26, 2017. 1979, 76, 6666-6670. 29. R. B. Merrifeld. Solid Phase Peptide Synthesis. I. Te 38. A. Chavkin, I. F. James, A. Goldstein. A. Dynorphin Synthesis of a Tetrapeptide. Journal of the American is a specifc endogenous ligand of the k opiate recep- Chemical Society 1963, 85, 2149–2154. tor. Science 1982, 215, 413-415. 30. J. Hughes, T. W. Smith, H. W. Kosterlitz, L. A. 39. K. Kangawa, H. Matsuo, M. Igarashi. Biochemical Fothergill, B. A. Morgan, H. R. Morris. Identifca- and Biophysical Research Communications 1979, 86, tion of two related pentapeptides from the brain with 153-160. potent opiate agonist activity. Nature 1975, 258, 577– 40. K. Kangawa, N. Minamino, M. Chino, S. Sakakibara, 579. H. Matsuo. Biochemical and Biophysical Research 31. R. Simantov, S. H. Snyder. Rabi Isolation and struc- Communications 1981, 99, 871-878. ture identification of a morphine-like peptide 41. N. Minamino, K. Kangawa, M. Chino, S. Sakakibara, “enkephalin” in bovine brain. Life Sciences 1976, 18, H. Matsuo. Biochemical and Biophysical Research 781-787. Communications 1981, 99, 864-870. 32. R. Simantov, S. H. Snyder. Morphine-like peptides 42. C. Mollereau, M. Parmentier, P. Mailleux, J. L. But- in mammalian brain: isolation, structure elucida- our, C. Moisand, P. Chalon, D. Caput, G. Vassart, J. tion, and interactions with the opiate receptor. Pro- C. Meunier. ORL1, a novel member of the opioid ceedings of the National Academy of Sciences of the receptor family. Cloning, functional expression and United States of America 1976, 73, 2515-2519. localization. FEBS Letters 1994, 341, 33-38. 33. R. Simantov, S. H. Snyder. Elevated levels of 43. V. S. Seybold. Te Role of Peptides in Central Sensi - enkephalin in morphine-dependent rats. Nature tization. In: Sensory Nerves, Vol. 194 (Eds.: B. J. Can- 1976, 262, 505–507. ning, D. Spina), Handbook of Experimental Pharma- 34. R. J. Bodnar. Endogenous Opiates and Behavior: cology book series, Springer Nature Switzerland AG, 2016. Peptides 2018, 101, 167-212. Basel, 2009, p. 451-491. 35. C. Stein Opioid Receptors. Annu Rev Med. 2016, 67, 44. A. F. Nastase, N. W. Griggs, J. P. Anand, T. J. Fernan- 433-51. dez, A. A. Harland, T. J. Trask, E. M. Jutkiewicz, J. 36. W. R. Martin, C. G. Eades, J. A. Tompson, R. E. R. Traynor, H. I. Mosberg. Synthesis and Pharmaco- Huppler, P. E. Gilbert. Te efects of morphine- and logical Evaluation of Novel C-8 Substituted Tetrahy- nalorphine-like drugs in the nondependent and mor- droquinolines as Balanced-Afnity Mu/Delta Opioid phine-dependent chronic spinal dog. Journal of Phar- Ligands for the Treatment of Pain. ACS Chemical macology and Experimental Terapeutics 1976, 197, Neuroscience 2018, 9, 1840-1848. 517-532.

Firenze University Press www.fupress.com/substantia

Historical Article Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Citation: E. Kenndler, N.M. Maier (2018) Gas Chromatography and Perspective Analysis of Binding Media of Museum Objects: A Historical Perspective. Sub- stantia 2(2): 93-118. doi: 10.13128/substantia-64 Ernst Kenndler1, Norbert M. Maier2 Copyright: © 2018 E. Kenndler, N.M. 1 Institute for Analytical Chemistry, Faculty of Chemistry, University of Vienna, A 1090 Maier. This is an open access, peer- Vienna, Austria. Corresponding author reviewed article published by Firenze 2 Department of Chemistry, A.I. Virtasen aukio 1 (P.O. BOX 55), FI-00014 University of University Press (http://www.fupress. Helsinki, Finland com/substantia) and distribuited under E-mail: [email protected]; [email protected] the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and Abstract. Tis contribution covers the major historic milestones of the evolution of gas reproduction in any medium, provided chromatography (GC) from its beginnings to its current status as one of the most pow- the original author and source are credited. erful analytical separation techniques, and demonstrates simultaneously how this technique has enabled and continuously improved the analysis of organic binding media Data Availability Statement: All rel- in objects of cultural heritage. Afer an introduction into the basics of chromatogra- evant data are within the paper and its phy, the development of GC is traced from its emergence in the late 1800s as a mere Supporting Information fles. preparative technique through a period of relative stagnation into the mid of the 20th century. Ten, the 1950s are covered by highlighting the major advances in theory and Competing Interests: The Author(s) declare(s) no confict of interest. technology within this decade, all of which contributed to frmly consolidate the status of GC as a modern analytical separation technique. From there the maturing of GC is followed through the 1960s up to the present days, a period being marked by the transition from packed to capillary columns; the essential adaptation of injection and detection devices; the replacement of glass by fused silica as column material; major progresses in stationary phase chemistry; and, fnally, the advent of the hyphenation of GC with mass spectrometric detection devices. Troughout this survey, examples of applications of contemporary GC techniques to binding media analysis are discussed to provide an illustrative historic record of the continuous improvements achieved. Te account will be closed with critical refections on GC’s current relevance to and future role in the analysis of binding media in objects of cultural heritage.

Keywords. Natural organic binding media, gas chromatography, history, cultural heritage, museum objects.

1. INTRODUCTION which we shall use in the following the term “museum objects”) can con- Knowledge of both the genuine sist of nearly innumerably inorganic techniques and materials employed and organic materials. Nowadays, the in the creation of objects of cultural reliable identifcation of the chemical heritage are of crucial importance nature and source of these materials for their scientifc and artistic analy- is of eminent importance to guide sis. Objects of cultural heritage (for the development of object-appro-

Substantia. An International Journal of the History of Chemistry 2(2): 93-118, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-64 94 Ernst Kenndler, Norbert M. Maier priate conservation and restoration techniques. In addi- stationary phase; either a solid or a liquid) is kept immo- tion, knowledge on the constituting materials in museum bile whereas the second phase (the mobile phase; either objects may provide a host of other important scientifc a gas or a liquid) is forced to fow continuously through insights, such as a historic record of the social and eco- the separation system. Afer introduction of the sample nomic conditions at the time the respective objects were into the system, the contained analytes are distributed fabricated, and information on the contemporary status between the mobile and the stationary phase according of crafsmanship, and intercultural exchange and technol- to their relative afnities, leading to their physical sepa- ogy transfer. Elucidation of the geographic provenience ration. of materials integrated in museum objects may also help Chromatographic methods can be categorized by to trace both ancient trade relationships and trade routes. several criteria, e.g. (1) according to the geometry of the And most importantly, a detailed profle of the consti- system; (2) the physical state of the phases and the oper- tuting materials in museum objects may provide valu- ative interphase distribution mechanisms; and (3) by the able evidence concerning the period of production and mode of sample introduction. geographic origin, and thus for reliable establishment of Concerning geometry, a chromatographic system authenticity of a given object of cultural heritage. may exhibit a (quasi-)two-dimensional format, such as Te present contribution will focus on one class of in thin layer and paper chromatography; however, we these materials, viz. natural organic binding media (in will not discuss these methods in the following. In con- the following termed binding media or binders).[1, 2] Te trast, in column chromatography, the stationary phase is determination of these materials in museum objects has situated within a tube, generally referred to as column, a long tradition, and many analytical approaches have through which the fow of the mobile phase is directed. been applied to this purpose. Tese methods range from With regard to the physical state of the phases, chro- visual examination over microchemical tests to the cur- matographic methods are classifed either as gas chro- rent state-of-the-art spectrometric and separation meth- matography (GC) or as liquid chromatography (LC), ods, such as liquid and gas chromatography, ofen used depending on the state of the mobile phase. GC and LC in conjunction with powerful mass-sensitive detection may further be categorized based on the physical state devices. Gas chromatography (GC), probably the most of the employed stationary phase as gas solid (GSC) and frequently employed analytical technique for the iden- gas liquid (GLC), and as liquid liquid (LLC) and liquid tifcation of binding media in museum objects, will be solid (LSC) chromatography. Note that in the present the central subject of this contribution. Specifcally, the contribution the method is termed partition chromatog- intent of this account is twofold: our frst objective is to raphy in case that the analytes are distributed between provide an overview on the historical development of a liquid or a gaseous mobile phase, respectively, and a GC from its humble beginnings to its current mature liquid stationary phase; distribution is based on absorp- status, and to pay credit to those scientists who through tion in the two phases. In adsorption chromatography, in their ingenious contributions have advanced GC to one contrast, the stationary phase is a solid surface (e.g. as in of the most powerful of the current analytical separa- ion exchange chromatography). tion techniques. In addition, and as our second goal, we Te mode of sample introduction may involve either wish to demonstrate how the continuous technological the injection of a sample amount being small relative advances achieved in GC methodology over the last fve to the volume of the system, or, alternatively the con- decades have made possible addressing the formidable tinuous introduction of sample solution. For the former challenges associated with binding media analysis in mode - the elution mode - the sample is introduced as museum objects. a narrow plug at the inlet of the column into the fow- Before advancing to the discussion of the historic ing mobile phase. During the transport of the sample milestones in the synergistic evolution of GC method- through the column, the contained analytes are separat- ology and concurrent improvements in binding media ed into individual sample zones with the mobile phase analysis, we fnd it benefcial to provide a brief section in between. It should be mentioned that for almost all clarifying some fundamental terminology and aspects of analytical applications, elution mode chromatography chromatography. is employed, and the term chromatography (that will be ofen used in the present contribution) is currently a common synonym of elution mode column partition 2. METHODOLOGY OF CHROMATOGRAPHY chromatography. In the latter mode - the frontal analysis mode - the A chromatographic system consists of two immisci- sample is either continuously fed as the mobile phase ble phases. Under operational conditions, one phase (the Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 95 into the column, or may be continuously introduced dissolved in the mobile phase. In this mode only the frst eluting zone contains pure analyte, with the subsequently emerging zones containing mixture of analytes, the composition of which being determined by the relative afnity of the analytes towards the stationary phase. Tis mode of chromatography is normally not used for analytical, but rather for preparative purposes. It shall be mentioned that other modes of chromatography do exist, such as displacement chromatography or chroma- thermography, but these are irrelevant for the current topic and therefore are not further discussed here. Elution mode GLC is currently the sole gas chromatographic method employed for the analysis of binding media in museum objects. Historically, GLC has emerged from a number of precursor techniques, which will be outlined in some detail in the following account on the invention and evolution of chromatographic techniques relevant to binding media analysis. However, prior to these discussions, a brief overview of the most important classes of organic compounds used as binding Figure 1. Representative constituents of natural organic binding media shall be given. media. (i) Waxes: Triacontanyl palmitate (in bees wax). (ii) Diter- penoic resins: Larixyl acetate (in Venetian turpentine). (iii) Dry- ing oils: Linolenic acid (9,12,15 octadecatrienoic acid (in linseed 3. NATURAL ORGANIC BINDING MEDIA IN OBJECTS oil). (iv) Animal glues: Hydroxyproline (in collagens). (v) Plant OF THE CULTURAL HERITAGE gums: Glucuronic acid (in gum Arabic). (vi) Bituminous materials: Hopane in asphaltenes. Remarkably, despite the fact that Nature provides a sheer unlimited repertoire of organic compounds, only a few classes have been applied as binding media Chemically, oils and fats are triglycerides of long- in museum objects. In accordance with their chemical chain fatty acids. Drying oils, employed as binders of the nature, binding media encountered in art objects can be pigments for oil paintings in Western art, contain a high categorized into six classes (see e.g. refs. [2, 3]), viz. waxes, proportion of unsaturated fatty acids. Te commonly resins, oils and fats, animal glues, plant gums, and bitu- used linseed oil mainly consists of the C18 fatty acids minous material. Representative compounds for each of oleic (C18:1), linoleic (C18:2) and linolenic (C18:3) acids these classes are depicted in Figure 1. (the sufx 18:2 denotes the presence of two C=C double The main constituents of natural waxes are long bonds in a fatty acid containing 18 C-atoms). Linoleic chain n-alkanes (and of beeswax also esters of long-chain and linolenic acids possess isolated double bonds, in fatty acids and alcohols, and smaller amounts of free fat- contrast to eleostearic acid, a C18:3 acid with three conju- ty acids). Waxes are found in museum objects amongst gated C=C double bonds, a main constituent of tung oil, others as varnishes and coatings, or as matrix compo- which was applied in some objects as substitute for lin- th [4, 5] nents of wax models, and have been widely used in antiq- seed oil in the frst half of the 20 century. Te dry- uity as binders for pigments in Encaustic painting. ing process is a radical-induced oxidative polymerization Natural resins are products secreted from woody of the unsaturated fatty acids, leading to the formation plants, mainly consisting of complex mixtures of diter- of a three-dimensional cross-linked network. Tis pro- penoids or triterpenoids (compounds with either 20 or cess is accompanied by cleavage of the double bond and 30 C-atoms in their molecules), preferentially of cyclic formation of short chain C7, C8, C9 dicarboxylic acids, structures; these compounds ofen contain C=C double which are generally detectable in aged dried oil. bonds, and bear hydroxy and carboxylic groups. It is Egg, casein and collagens are the main animal glues worth noting that a given plant produces either diter- used; they are proteins and consist of peptide chains penoid or triterpenoid resins, but not both. In museum containing essentially all of the twenty natural amino objects, resins have been generally employed as varnish- acids. However, the amino acid hydroxyproline is a spe- es, for coatings, additives, and consolidants. cifc constituent of collagens, formed by post transla- 96 Ernst Kenndler, Norbert M. Maier tional modifcation of proline, and is a unique marker is key for an informed selection of favorable operation- of this type of glue. Egg was widely used as a binder of al variables and parameters. While we wish to restrain pigments in tempera painting, the dominating painting from a detailed treatment of chromatographic theory, technique used prior to the invention of oil painting in we consider it benefcial to familiarize interested read- Western art in the early 15th century. ers with some fundamental relationships. We hope these Plant gums are polysaccharides, composed of a will aid the understanding of crucial milestones that range of monosaccharides and uronic acids (typically marked the development of GC from its beginning as a glucuronic and galacturonic acids). Frequently employed preparative technique to its present status as one of the plant gums in binding media are gum Arabic, gum tra- most powerful methods for trace analysis. gacanth and cherry gum. Occasionally, starch has also been applied. Bituminous material is a very complex mixture of 4.1 Zone Migration: Retention Time, Retention Factor and high molecular mass compounds, and is a generic term Separation Selectivity for two classes of substances, amongst bitumen and The migration velocity, u , of a given analyte, i, asphalt are natural materials, and tars and pitches are i through a chromatographic column with the mobile technical products. Pitches are directly resulting from phase fow velocity, v, is determined by the analyte´s pyrolysis of wood or resin, while tars represent the degree of distribution between the stationary (denoted products of a subsequent distillation. However, these by subscript s) and mobile phase (denoted by subscript compounds are rarely encountered as binders in muse- m). In the steady state, the fraction of the analyte in the um objects, and will not be treated here in more detail. mobile phase is equal to the ratio of mole number n Readers interested in an in-depth treatise on organic i,m to the total mole number (n +n ) being n /(n +n ), binding media and coatings are directed to ref. [2]. i,m i,s i,m i,m i,s which can be also expressed as 1/[1+(n /n )]; the ratio As an essential prerequisite, GC analysis requires i,s i,m n /n is the mass distribution coefcient. In partition the analytes of interest (except for pyrolysis GC, see i,s i,m chromatography like GLC, the respective mole numbers below) to be sufciently volatile and thermally stable to are equal to n =c.V, the product of corresponding con- avoid decomposition at the typically employed elevated centrations, c, and volumes, V. Te analyte concentration temperatures. Terefore, GC is not directly applicable to ratio3 between stationary and mobile phase is the parti- the majority of the common binding media. However, tion coefcient, K =c /c , and the ratio of the volumes this limitation can be conveniently overcome by chemi- i i,s i,m of stationary and mobile phase is named phase ratio, cal transformation of these materials into more volatile q=Vs/Vm. Combining these expression allows formula- compounds, e.g. by de-polymerization via acid-catalyzed tion of the fraction of the analyte in the mobile phase by hydrolysis and appropriate derivatization of the emerging low-molecular mass constituents. ni,m/(ni,m+ni,s)=1/(1+Kiq)=1/(1+ ki) (1)

4. A BRIEF REVIEW OF THE GENERAL PRINCIPLES We defne ki, the retention factor, as ki=Kiq; it is AND THE TERMINOLOGY OF CHROMATOGRAPHY identical with the mass distribution coefcient, and is one of the most important parameters for the descrip- Two main parameters determine the separabil- tion of any chromatographic process. ity of analytes in chromatography, namely the retention Since the fraction 1/(1+ki) of the analyte in the factor1 and the plate number2. Te former parameter mobile phase (see Equation 1) migrates with velocity v, refects the velocity by which a given analyte zone moves and assuming that the rate of the mass exchange of the through the column, while the latter is a measure of the analyte between the two phases by distribution is fast, it continuous broadening the zone underlies upon migra- follows that the entire analyte zone moves through the tion through the chromatographic system. Certainly, a column with velocity ui, which can be expressed as given combined knowledge of the properties of the chromato- in Equation 2a by graphic system and the analytes, and their interactions

3 Note that in partition chromatography the concentrations in both 1 We prefer to use the more explicatory term retention factor rather than phases are defned by moles per volume. In adsorption chromatography, mass distribution ratio, as proposed by IUPAC, or capacity factor, the in contrast, the concentration of the analyte at the stationary solid sur- more common term in older literature. face with area A is given by moles per area [mol/A]. In this case the 2 Initially, in the classical literature of zone dispersion (see Chapters 4.2. distribution coefcient K is not dimensionless, but has the dimension and 5.5) the term plate number is named “number of theoretical plates”. of a length. Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 97

Under the premise that the chromatographic col- a) ui = v (1 + ki ) = v (1+ Kiq) umn is operated at constant temperature and the sample b) tRi = L ui = (L v)(1+ ki ) = tR0 (1+ ki ) (2) is introduced as an (infnitely) narrow plug into the column, the recorded concentration distribution of the ana- c) ki = (tRi −tR0 ) tR0 lyte forms the typical Gaussian curve (usually referred to as a peak) due to various dispersion processes. Te From Equation 2a it can be concluded that in GLC width of the peak is expressed by its standard devia- 2 the migration velocity, ui, of analyte, i, depends on three tion, σ. During migration, the peak variance, σ x in the parameters, namely (i) on the mobile phase fow velocity length scale increases directly proportional to the migra- s m 2 v, and (ii) on the phase ratio q=V /V , with both of these tion distance, x, according to σ x = H x Te proportion- parameters being equal, i.e. unspecifc, for all analytes; ally factor, H, is a characteristic parameter for the dis- and (iii) on the partition coefcient, Ki. Tis partition persion property of the chromatographic column, i.e. coefcient is an analyte-specifc quantity, refecting the for its efciency. For historical reasons, H is referred to distinct interactions a given analyte undergoes with the as height equivalent of a theoretical plate, or (theoretical) stationary and mobile phase, respectively. Diferences plate height and has the dimension of a length. Note that in K, or more specifcally in the retention factors k are the efciency of a column must not be confused with its mandatory to achieve analyte separation. Te degree of ability to separate compounds, efciency (expressed by a separation depends on the ratio of the retention factors, fgure) is a property that is strictly related to zone broad- the so-called selectivity coefcient rji=kj/ki (with kj≥ki), ening. which is a measure for the separation selectivity of the In column chromatographic practice, however, the system for a given pair of analytes, i and j. zone width is not measured in the length domain at a For column chromatography, the retention or certain time (as it is done in planar chromatography). residence time, tRi, is the time the analyte needs to Rather, all components are permitted to traverse the migrate with its velocity ui through the column with entire column length L and are registered at the outlet of length, L; it is given by Equation 2b. Te void or dead the column as a function of time. For convenience, the time, tRo is the time the mobile phase requires to fow resulting peak widths are measured in the time domain, through the column over length, L. Experimentally, e.g. by the time-based standard deviation, σt,i. Both retention times are measured at the maximum of cor- standard deviations, σt,i and σx,i (since at x=L, σx,i can be responding analyte concentration profiles (usually written as σL,i), are related to the migration velocity, ui, Gaussian) upon elution from the columns. Note that which is L/tR,i by the retention time tRi of the analyte depends on the same parameters as the migration velocity, viz. on σ = H L the unspecifc mobile phase velocity (and the column L,i 2 2 (3) b) σ = t × H L length), and on the analyte-specifc retention factor, ki t ,i R,i again emphasizing the importance of this parameter. Te retention factor ki can simply be calculated from Te ratio L/H=N is referred to as the number of the measured retention time tRi and the dead time tRo (theoretical) plates or (theoretical) plate number, N, and according to Equation 2c. is a measure for the efciency of a given column with length L and plate height, H. Te plate number, N, can be conveniently calculated from the time-based peak 4.2 Zone Broadening: Plate Height, Plate Number and Sep- width and the corresponding retention time according aration Efciency to Equation 4a as It is important to recognize that diferent migration = 2 σ 2 velocities of a pair of analytes in the chromatographic a) N tR,i t ,i (4) column are an essential but not a sufcient criterion for b) σ t ,i = tR,i N their successful separation. Tis is caused by the fact that the migrating analytes are continuously diluted by the mobile phase, i.e. they become dispersed within a From the rearranged Equation 4b it can be seen that larger volume and, as a consequence, their zones become σt,i increases directly proportional to the retention time broader; thus, neighboring zones of a given pair of ana- tR,i (the causes for this increase will be discussed in more lytes might overlap even if they possess diferent reten- detail in Chapters 5.5 and 7). tion factors. 98 Ernst Kenndler, Norbert M. Maier

4.3 Te Chromatographic Resolution: the Quantity for the (e.g., very volatile compounds in thin flm capillaries, Degree of Separation see below). (ii) Te third term, the efciency term, contains the plate number, expressing the efect of peak As outlined above, the separation of a given pair of broadening on resolution. Since separation depends on analytes, i and j, is governed by two processes, namely the square root of N, realizing a twofold improvement zone migration and dispersion. Certainly, there is a need in resolution requires a fourfold increase in plate num- for a quantity that expresses the degree of separation in ber. Practically, on the one hand, this could be achieved a well-defned fashion considering the combined efect by employing a fourfold longer column, yet improve- of these processes. Tis quantity is the chromatographic ments in resolution would come at the prize of a four- resolution, Rj,i. fold increase in analysis time. On the other hand, ef- It is obvious that for successful separation of a pair ciency can be enhanced by using columns with lower of analytes their retention times must differ (which plate heights; this approach became feasible through the means that the selectivity coefcient rji must be larger invention of open tubular capillary columns (see Chap- than unity). It is, however, not meaningful to measure ter 7). Very signifcant improvements in resolution can this diference (tr,j–tr,i) in absolute time units, because at be realized by changing separation selectivity, which is a given retention time diference narrow peaks may be represented in the frst term of Equation 6 by the selec- well resolved, while broad peaks may still strongly over- tivity coefcient, rji. Evidently, any variation in station- lap. Terefore (tr,j–tr,i) is related to the width of the two ary phase and/or the operation conditions that results peaks, given by their time-based standard deviations, σt,i even in a minute increase of rji will produce pronounced and σt,j. According to IUPAC, the degree of separation, changes in resolution. the chromatographic resolution, is defned for this pair of analytes by 5. A SHORT HISTORY OF GAS CHROMATOGRAPHY (tr ,j −tr ,i ) Rj,i = (5) 2σ + 2 σ Decades before gas liquid chromatography was t ,i t ,j invented, separations were already carried out by liquid chromatography. Te invention of liquid chromatogra- Te resolution as defned in Equation 5 is a dimen- phy is attributed to M. S. Tswett, yet there were several sionless number; baseline separation of two peaks (of earlier studies by other scientists applying essentially equal size) is obtained when the resolution has a value liquid solid chromatographic techniques in the frontal equal to or larger than 1.5. analysis mode for preparative purposes; e.g. in 1893, L. However, Equation 5 is not very practical when reso- Reed separated salts by applying their solution onto a lution needs to be expressed as a function of variations tube flled with kaolin as adsorbent[6]; and D.T. Day pub- in experimental parameters. Transforming this relation- lished frst investigations with LSC in 1897 aiming at the ship considering Equations 2b and 4b provides a practi- separation of colored constituents characteristic for oils cally more useful expression for the chromatographic from diferent sources.[7] resolution being a function of retention factors and plate number 4 by 5.1 Te Invention of Chromatography: Liquid Solid Chro- k − k matography 1 ( j i ) ki 1 ki Rji = N = (rji −1) N (6) 4 ki (1+ ki ) 4 (1+ ki ) In his frst publication from 1903 Tswett, a Russian botanist, described the successful separation of plant [8] From Equation 6 it is evident that the achievable pigments. In his experiments, he applied a chlorophyll chromatographic resolution is impacted by three terms extract in ligroin (i.e. petroleum ether) at the top of a (we chose to ignore the factor ¼): Te middle term, vertically arranged cylindrical glass tube (see Figure 2) the retention or retardation term, plays a minor role; flled with particles of a solid material with adsorptive it is relevant only to analytes with very small retention abilities, and continued applying fresh ligroin. Tswett observed the formation of separated colored rings, which migrated through the tube and broadened during their 4 Variation of the experimental conditions is needed only for a pair of migration. closely migrating analyte, i.e. for analytes with very similar retention. In Tswett coined for this separation technique the term this case k ≈k and N ≈N =N However, note that even if the diference j i j i “chromatographic method“, frst mentioned in 1906in (kj–ki) is close to zero, the ratio rji can be signifcantly larger than unit. Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 99

J.P. Martin and R.L.M. Synge were jointly awarded the Nobel Prize in Chemistry 1952 “… for their invention of partition chromatography”. Chromatography employing liquid stationary phases allowed exploiting a much greater variety of solute interactions for tuning selectivity than LSC, and ultimately promoted LLC to one of the most important contemporary analytical separation methods. Te practical utility of LLC was further potentiated by reducing the particle sizes of the packed bed down to the micrometer range and thus producing columns providing vastly enhanced column efciencies. From this efort, High Performance Liquid Chromatography Figure 2. Tswett’s device with four packed chromatographic glass (HPLC) emerged, a chromatographic technique that columns. Drawing (1.): three columns flled with adsorbents for the from a general viewpoint is superior to GC for two rea- separation of plant pigments. Te columns had an inner diameter sons: One crucial limitation of GC is that its applicabili- of 2-3 mm, and a length of 2-3 cm. Drawing (5.): Separated zones ty is restricted to the relatively low number of sufcient- of 5 colored plant pigments (Chlorophylls and Xanthophylls) in a ly volatile and thermostable compounds, requirements chromatographic packed column. From ref. [9] with permission. that are certainly irrelevant to LLC. Te second aspect that favors LLC over GLC is that in the latter technique selectivity emerges through interaction with the liquid two publications in a German journal.[9, 10] Interest- stationary phase only, as the mobile phase is an inert ingly, the name of this method persisted until nowadays gas. In contrast, in LLC, the mobile liquid phase ofers (although it is a misnomer, as analytes processed by an additional and highly versatile tool for varying sepa- modern chromatography are not necessarily colored). ration selectivity via specifc solvent-solute interactions. Afer a dormant period, Tswett´s ideas were revisited Taken together, these advantages enormously widen the in 1931 by the biochemist R. Kuhn and his coworkers, general separation ability of LLC as compared to GLC, who successfully used LSC to accomplish the separation and explain the outstanding success of HPLC in mod- of carotins and xantophylls.[11-14] In 1937, three decades ern analytical sciences. afer Tswett’s reports on the chromatographic method, the frst monograph dedicated to chromatography was [15] published. However, due to the limited scope of com- 5.3 Early GC: Gas Solid Chromatography pound classes that could be addressed, the use of LSC remained somewhat limited in the scientifc community. Even at times predating Tswett´s introduction of Te ultimate breakthrough of liquid chromatography as LSC, the adsorption of gases or liquids on solid sur- a powerful separation technique came with the recogni- faces was actively investigated, with the frst reports tion that replacing solid by (solid-supported) liquid sta- emerging at the beginning of the 19th century. In the tionary phases allowed for an enormous extension and early 20th century research was primarily devoted to the signifcant improvements of selectivity profles. Certain- adsorption of gases on solid sorbents pursuing prepara- ly, this notion marked the birth of modern liquid liquid tive applications. Specifc areas of interest were, e.g. the partition chromatography (LLC). purifcation or recovery of constituents of vapors, or the improvement of the efectivity of gas masks. Essentially all of these investigations were carried out by frontal 5.2 Liquid Liquid Chromatography analysis mode.[18-20] In 1930s P. Schufan introduced separation tech- Te innovative concept of liquid liquid partition niques for which he coined the general term “adsorp- chromatography was published by Martin and Synge tion analysis”, and which may well have been the frst in 1941, including a model to express the efciency of successful demonstration of gas solid chromatography a column[16]; in the same journal issue the separation for analytical purposes, but still at a micro-preparative of N-acetylated amino acids in a column with water as scale. He applied this method for gas analysis in the liquid stationary phase (absorbed in silica gel) and chlo- technical area, and separated and quantifed gases such roform with 0.5 to 1.0% n-butanol as mobile phase was as low-boiling hydrocarbons, carbon monoxide and described.[17] In appreciation of their pioneering work, hydrogen. 100 Ernst Kenndler, Norbert M. Maier

In early 1930s, Tswett’s introduction of elution mode 5.4 Te Innovation: Gas Liquid Chromatography LSC returned from oblivion and was gradually adopted for GSC.[14] G. Hesse et al. described GSC separa- A curiosity in the history of partition GC is the tion using a carrier gas as mobile phase rather than the frst traceable separation apparently based on gas liquid sample mixture as in frontal analysis. Remarkably, in chromatography and described as early as in 1512, in 1943 the frst separation by gas liquid chromatography the period between the Late Middle Ages and the early appears to have been carried out by G. Damköhler and modern age, by Hieronymus Brunschwig (ca. 1450 - ca. H. Teile.[21] Specifcally, they achieved separations of 1512), in his book “Liber de arte Distillandi de Composi- methanol from ethanol, and benzene from cyclohexane, tis. Das buch der waren kunst zu distilieren die Compos- employing tubes flled with grained fred clay as solid ita”[27] (the title page of this book is shown in Figure 3). support and glycerol as stationary liquid phase (it should Brunschwig, a German surgeon and botanist, be mentioned that their primary intention for the addi- describes a procedure in which the vapor from a mix- tion of glycerol was to deactivate the solid surface) and ture of alcohol and water was forced through a sponge hydrogen or nitrogen as mobile phase. Unfortunately, moistened with olive oil, and was leading to the recovery their contribution, afer having been published in a less of a small quantity of pure alcohol. Expressed in modern renowned journal, did not fnd any resonance in the terminology, this technique represents a separation pro- scientifc community. In addition, as both authors were cess based on frontal GLC, with the oil acting as a liq- staf scientists employed at an institution 5 devoted to uid stationary phase, the sponge as a porous supporting support the German war efort, their interest may have material, and the alcohol vapor as mobile phase.[28, 29] been redirected to issues more pressing than further research into this method. Te decisive step towards establishing GSC as a useful microscale separation technique was taken by E. Cre- mer and her coworkers, who constructed the frst fully operational analytical gas chromatograph operating in elution mode, including a sufciently sensitive home- made thermal conductivity cell as a detector. Interest- ingly, Cremer described the results in a manuscript that was accepted for publication in 1944, but which failed to appear in print due to the chaotic conditions at the end of World War II. Cremer’s results were published years later, between 1949 and 1952[22-24], and even then they remained largely ignored. Anyhow, the application range of GSC was found to be rather limited, though A.V. Kiselev et al.[25] and others invested considerable eforts in modifying the adsorbents to enhance the variety of available interactions towards the volatile analytes. Nevertheless, it turned out that even with these modifcations the application range of solid stationary phases remained restricted; and mainly suitable for the separation of low-polarity compounds. Before closing the present section, we would like to refer readers interested in the historic development of GC to a recent review[26], covering most of the early achievements up of the 1950s. Specifcally, this article also gives due credit to a range of exceptional scientists from the former USSR for their contributions on this topic.

Figure 3. Title page of the book Liber de arte Distillandi de Com- positis. Das buch der waren kunst zu distilieren die Composita by 5 At the “Institut für Motorenforschung der Luffahrtforschungsanstalt Hieronymus Brunschwig, published in 1512, describing a kind of Hermann Göring”. gas liquid chromatography.[27] Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 101

0 activity coefcient, γ i , of the analyte at infnite dilu- 0 0 tion in the liquid phase. i.e. Ki = prop 1 ( pi γ i ). We do not give here the derivation of the partition coefcient, but just mention that it can be conveniently obtained by considering Henry’s law for non-ideal binary liquid mixtures and Dalton’s law of ideal gases (see textbooks about GC). As in gas liquid chromatography the selectivity coef- fcient, rji is expressed by Figure 4. Separation of short-chain carboxylic acids by gas liquid chromatography demonstrated in the seminal paper by James and 0 0 Martin in 1952 introducing partition gas chromatography.[30] Sep- kj K j pi γ i rji = = = 0 0 (7) arations were carried out in a packed column in the isothermal ki Ki pj γ j mode. Curve A, experimental results; curve B, derivative of experimental curve. For details see Supplementary Information and ref. [30] [30] . From ref. with permission. and as rji must be larger than unity, it follows from Equation 7 that a given pair of analytes can be readily separated if their products of p0 and γ0 difer. For chemi- However, we will now return to the advancement of cally very similar analytes - with about equal γ0 - this 0 0 gas chromatography as a branch of modern separation can be achieved if the vapor pressures, pi and pj of the science. In the early 1950s James and Martin adopted pure compounds difer at the operational column tem- the concept of liquid liquid chromatography (introduced perature. However, more relevant for the variation of by Martin and Synge in 1941[16, 17]) to gas chromatogra- separation selectivity is the ratio of the activity coef- phy by replacing the solid surface by a liquid as station- 0 0 cients (γ i γ j ) because it refects the specifc intermolec- ary phase, which switched the originally adsorption to ular interactions between the analytes and the liquid sta- partition-based interaction mechanism. In their publi- tionary phase.[32, 33] Tus, by proper selection of station- cations from 1952, James and Martin both suggested a ary phases liquids from a broad range of chemically dis- comprehensive theory for gas liquid partition chroma- tinct compounds, the activity coefcients of given pair tography based on the plate concept (see below), and of analytes can be conveniently adjusted so as to achieve demonstrated experimentally the separation of volatile the level of selectivity required for a given separation. acids and bases.[30, 31] Detection and quantitation were carried out by titration of the eluted analytes with an automatic recording burette. 5.4.2 Isothermal and temperature-programmed mode One of their frst published separation by GLC[30], that of short-chain carboxylic acids on a column packed In GLC, the isothermal elution mode (in which the with solid-supported poly(phenylmethyldimethylsiloxa column is held at constant temperature) is suitable for ne) (with 10% w/w stearic acid added) as stationary and the separation of sample constituents which possess nitrogen as mobile phase at 137°C column temperature retention factors within a reasonably narrow range. At a is shown in Figure 4. Note that all eight acids are com- given constant column temperature, the vapor pressures, pletely resolved. Arguably, this invention of partition gas and the activity coefcients of the sample components, chromatography by using a liquid as stationary phase and therefore their retention factors, all remain essen- was the foundation for all further developments in the tially constant during the chromatographic run. Conse- feld, ultimately making GLC to one of the most useful quently, the selection of an appropriate operational col- analytical separation methods currently available. umn temperature enables the adjustment of the elution of the analytes of interest to an acceptably narrow retention time window. 5.4.1 Separation selectivity in gas liquid chromatography However, isothermal conditions are not favorable for the separation of samples composed of analytes with In practical GLC, the vaporized analyte is distribut- largely difering retention factors. For such mixtures, at ed between the stationary liquid and the mobile gaseous a low column temperature the early eluting analytes can phase with a partition coefcient, Ki, which is inversely be satisfactorily resolved, while those possessing very 0 proportional to the vapor pressure, pi , of the analyte large retention factors will elute at unacceptably long as pure compound at the given temperature, and to the retention times. Moreover, the longer the retention time 102 Ernst Kenndler, Norbert M. Maier the broader the peaks become (see Chapter 4.2), and the limeters and few meters in length. Tey were packed late-eluting wide peaks might even disappear within the with porous solid particles (initially e.g. granules of fre- noise of the baseline of the chromatogram. Selection of brick, later kieselgur, i.e. purifed diatomaceous earth), a high temperature, on the other hand, would be an ef- which were impregnated with the stationary liquid prior cient means to adjust the retention characteristics of the to use. Te early versions of supporting material were late eluting compounds appropriately, yet with the draw- later replaced by well-defned commercially produced back that under these conditions the early eluting com- homogeneous synthetic particles. Packed bed columns pounds would be poorly retained and can thus emerge (also referred to as packed columns) could easily be pre- from the column unresolved (see the middle term of the pared, which favored their general acceptance. resolution equation, Equation 6). Certainly, isothermal As already mentioned, the separability of analytes is GC is not a benefcial method for the analysis of mix- impaired by a number of processes caused by the inevi- tures with such complex compositions. tably broadening of the initially narrow sample zone, Tis general elution problem is valid for all chro- processes that determine the column efciency. In the matographic techniques, i.e. GC and LC. For GLC this frst theoretical approach to describe band broadening, fundamental issue can be conveniently addressed by formulated by the plate theory, the column is considered exploiting the strong temperature dependence of the as being composed of a series of interconnected cells or distribution coefcient and the retention factor, respec- “plates” containing the mobile and the stationary phase. tively. Pioneering investigations of the chromatographic Upon migrating through the chromatographic column, behavior of compounds in adsorption columns in pres- the analyte is assumed to distribute between these two ence of longitudinal temperature gradients were carried phases within each plate element with equilibrium being out by Turkeltaub, Zhukhovitskii, et al.[34, 35]. Te authors reached.[40] Te mobile phase with the fraction of sol- coined for this separation method the term chroma- ute at equilibrium concentration is then transferred thermography. to the next plate downstream, where the same pro- In GLC, the vapor pressure of the pure compound cess takes place again. It is important to recognize that increases exponentially with increasing tempera- the plate theory assumes that at each distribution step ture (according to the Clausius-Clapeyron equation), equilibrium conditions are achieved. Tis requirement and accordingly Ki and ki decrease exponentially with is certainly not fulflled as during the chromatographic increasing temperature. Exploiting these facts, in 1958 S. process in the column the fraction of the solute in the Dal Nogare et al.[36, 37] introduced a method to run sam- mobile phase is continuously transported by the mobile ples consisting of components with large diferences in phase while the fraction in the stationary phase perma- volatility by varying the column temperature, T, as func- nently lags behind. Equilibrium would only be achieved tion of time, t, i.e. by applying a certain gradient dT/dt under the condition of infnitely fast inter-phase mass to the entire column. When employing this temperature transfer, which is an unrealistic proposition. Tis means programming technique, the initial low-temperature con- that analyte distribution actually occurs under non- ditions are adapted to ensure appropriate retention of equilibrium conditions[41, 42], and therefore the kinet- the early eluting analytes, while the fnal high tempera- ics of the mass exchange will additionally contribute to ture conditions are chosen to enable complete elution of the “Height Equivalent of a Teoretical Plate” (HETP; the late eluting sample components. Between these lim- the terminology is adopted from the plate theory). Logi- its, the retention factors of the analytes are continuously cally, this particular contribution will become the more decreases by the action of the well-defned T-gradient, pronounced the higher the migration velocity of the with the consequence that the observed retention times zone is. In contrast, at sufciently low migration veloc- are signifcant shorter than those seen under isother- ity the inter-phase mass exchange will approach equi- mal conditions. Apart from reducing the time of analy- librium conditions. Tis additional contribution to zone sis, temperature programming also causes a pronounced dispersion is taken into account in the rate theory, which compressing of the analyte zones into very sharp peaks. specifcally accounts for the efects of fnite inter-phase [38, 39] mass transfer kinetics. Te rate theory for packed columns, considering the plate height, H, as function of the velocity, v, of the 5.5 GLC with Packed Bed Columns: Low Efciency and mobile phase, was formulated by van Deemter, Zuider- Needs for Selectivity weg and Klinkenberg [43] (for elution chromatography and for isothermal conditions), expanding the theo- Early analytical GC columns were fabricated from ries described by Lapidus and Admunson[44], and by E. metal or glass tubes with inner diameters of several mil- Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 103

Glueckauf[42]. In its simplifed form, H=f(v), is expressed by the three-term function

H=A+B/v+C.v (8)

For GLC with packed columns term A refects the contribution to peak broadening caused by eddy dispersion due to the heterogeneous particle-size distribution of the packing, term B describes the contribution of longitudinal difusion and of the diferent path lengths of the fow lines around the particles, term C accounts for the kinetics of the mass transfer of the solute molecules between the phases. Note that this equation contains objective quantities like retention factor, particle diam- Figure 5. Plate height, H, in dependence on the mobile phase velocity, taken from the original publication of van Deemter, eter, difusion coefcients, but also several empirical fac- Zuiderweg and Klinkenberg from 1956. Measurements were carried tors, i.e. those characterizing packing geometry and tor- out with a packed bed column under isothermal conditions. For tuosity. For more complex conditions the rate theory was details see Supplementary Information and ref. [43]. From ref. [43] refned by J.C. Giddings.[39, 45] with permission. According to the van Deemter equation (Equation 8) H decreases hyperbolically with increasing v considering term B/v, while it increases linearly with v with regard a wide range of polarity, an efort that led to the devel- to term C.v (term A is independent of the velocity). Te opment of several hundred commercially available sta- resulting H vs. v curve, given by the sum of the individ- tionary phases with optimized yet rather narrow selec- ual curves, exhibits a minimum plate height, Hi,min, at a tivity profles. distinct, singular fow velocity; i.e. at this fow velocity the column will produce maximum efciency for a given 5.5.1 Analyte identifcation by GC: Retention or Kováts analyte (note that H depends on the retention factor, i,min Index and therefore the diferent analytes have diferent values of Hi,min even under identical conditions). In the early days of GLC, rather unspecifc detection In their seminal contribution, van Deemter et al. systems (instruments coupled with mass-sensitive detec- put their theory on test by measuring H=f(v) curves tion devices were yet to come) were employed, rendering for a number of analytes and columns, and the results the identifcation of an unknown analyte, y, challenging. obtained for n-butane and i-butane are shown in Figure [43] To address these issues, concepts were developed that 5. Te H vs. v curves show shapes that are in accord- allowed identifcation of unknowns based on their chro- ance with theory, i.e. curves exhibiting a minimum at a matographic behavior by comparing the retention char- singular carrier gas velocity, and distinct Hi,min for the acteristics with those of known reference compounds individual analytes. listed in the literature. The most obvious choice of As a consequence of their inherently low efciency parameter for this purpose, their retention time, is not packed columns produce relative broad peaks, which sufciently robust, as it depends on a number of instru- can easily be rationalized using the general expression mental variables, such as the mobile phase fow veloc- for the plate number N=L/H. Obviously, one reason for ity, the length of the column, the phase ratio, and the the low plate number is the considerably large value of temperature. Te retention factor, ky of the unknown H generally observed for packed columns. Te second analyte, y, being independent of fow velocity and col- reason is the relatively short column length (of about a umn length, is a better choice, yet is still a function of few meters), which is a practical limitation dictated by the phase ratio. However, the variation of the retention the need to keep the inlet pressure of the carrier gas factors with diferent phase ratios can be accounted for workably low. Plate numbers are therefore rarely high- by resorting to relative retention factors, i.e. retention er than a few thousands. Given this limitation in ef- factors that are calibrated using a set of reference com- ciency, considerable research was devoted to improving pounds. separation performance of packed column GLC through [46] [32, 33] For GLC, E. Kováts proposed the homologue optimization of selectivity. Tis goal was met by series of straight-chained alkanes as suitable reference employing a large number of the stationary liquids with compounds, and defined the so-called Retention (or 104 Ernst Kenndler, Norbert M. Maier

Kováts) Indices, IR,n, for all stationary phases (and at all phase, p , relative to that of a highly apolar stationary temperatures) as exactly the hundredfold of the number phase, with the latter being squalane (Sq), a branched of their C-atoms: IR,n=100.n; normal undecane, e.g. has C30 alkane. Initially fve reference compounds (benzene, an Index of 1100. An analyte, y, which elutes between ethanol, ethyl methyl ketone, nitromethane, and pyri- two homologue straight-chained alkanes with carbon dine) were selected by Rohrschneider (the set was later numbers of n and (n+1) is considered to behave like extended to ten reference compounds by McReynolds) a hypothetical alkane with C-number ny, a fractional to represent characteristic types of interaction with the number between n and (n+1). Tis hypothetical number liquid phase (i.e., London dispersion, π– π electron, elec- ny can be calculated from ky, given the linear depend- tron attracting and dipole-dipole interactions, H bond- p/Sq ence of logkn on n of the reference alkanes (this relation- ing capability). Te ΔIR values represent measures for ship is strictly valid under isothermal conditions only); individual intermolecular forces of these reference com- the analyte-specifc Kováts or Retention Index results pounds with the stationary liquid, and are expressed therefore IR,y=100.ny. for practical applications as constants x', y',z',u',s'. Retention indices depend on the stationary liq- Under the assumption that the retention behavior of uid only and thus enable the identifcation of unknown a stationary phase is a manifestation of its intermo- analytes by comparing experimentally measured values lecular interaction forces, the sum of the constants, with those documented in the literature (huge collection Σ = x'+ y'+ z'+u'+ s 'generally known as the Rohrschnei- of IR values for many compounds and stationary phas- der-McReynolds Index, Σ is a specifc measure for the es have been compiled over the years and were readily polarity of a given stationary liquid. available in the literature and databases). Identifcation Tis index can be used to rank the stationary phases of unknown compounds is considerably facilitated by according to their polarity; e.g., Σ is zero for squalane, comparison of indices measured on several diferent sta- 229 for relatively apolar poly(dimethylsiloxane), and tionary phases. If no reference indices are available, the 3682 for highly polar poly(cyanopropylphenylsiloxane). p/ap p ap diference of retention indices, ΔIR = IR − IR , measured In practice, these Indices are particularly helpful for the on a polar (p) and an apolar (ap) phase can be exploit- assessment of the similarity of the polarity for station- ed to gain information about the type of the functional ary phases from diferent commercial sources. Moreover, groups present in the analyte. the Rohrschneider-McReynolds Index and constants can It should be mentioned that this concept was suc- be used to guide the selection of appropriate stationary cessfully applied to the analysis of binding media of phases for the separation of given analytes 6. the paste layer of a shell-inlaid ceremonial shield from the Solomon Islands.[47] Te object originates from the 1st half of the 19th century, and is housed in the Welt- 6. THE INCLUSION OF GAS LIQUID museum (the former Museum of Ethnology) in Vienna, CHROMATOGRAPHY WITH PACKED COLUMNS TO Austria. In the course of the investigation of the compo- THE ANALYSIS OF BINDING MEDIA sition of paste layer two sample constituents were detected by GC, but compound identifcation by MS was ham- From about 1965 the potential of GC for the analy- pered by observation of essentially identical mass spec- sis of binding media of museum objects was started to tra (at least with MS instrumentation available at the be recognized, although this technique was not directly time the study was carried out). However, these two ana- applicable to a number of substance classes of present lytes could be identifed as two isomeric octadecatrienoic interest. Plant gums (polysaccharides) or animal glues p/ap acids by means of their ΔIR . (protein) would rather decompose than evaporate at high temperature, but appropriate procedures for their transformation into GC-conform modifications were 5.5.2 Polarity of stationary phases: Rohrschneider-McReyn- developed or adapted from the literature. However, some olds Index problems specifc to the analysis of museum object still

Polarity is a term employed to describe the chroma- 6 tographic retention characteristics of a stationary phase. Such selections are ofen guided by the well-known rule-of-thumb [48] “similia similibus solvuntur” concept, which may be understood as the Initially used rather intuitively, L. Rohrschneider and three-word essence of the Rohrschneider’s polarity classifcation. It [49] later W.O. McReynolds introduced a concept to cod- appears to have been formulated in analogy to the principle “similia ify the polarity by a number. Tis concept is based on similibus curantur”, attributed to Paracelsus, and “similia similibus curen- p/Sq p Sq tur”, a motto of homoeopathy (for the source of the solubility rule see the Retention Index diferences, ΔIR = IR − IR , of cer- J.H. Hildebrand, R.L. Scott, Te Solubility of Nonelectrolytes, ACS Mono- tain selected reference compounds on a given stationary graph No. 17, Reinhold Publ. Corp., 1950). Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 105 have to be overcome, such as the very limited sample amounts available, the complexity of mixtures of several classes of binders ofen encountered in a single sample, occasionally together with products stemming from degradation and decomposition processes, the large excess of organic and/or inorganic matrix compounds and the presence of contaminants. Much research was devoted to address these issues over the second half of the 1960s.[50- 53] Especially J.S. Mills and R. White carried out a number of systematic investigations concerning GC analysis of the diferent binding media[51, 54-59], and published later a comprehensive account on this topic.[2] First GC analyses of binding media were conducted with packed columns. An example for a chromatogram obtained under isothermal conditions for resinous material present in wax models is shown in Figure 6, top panel. It is worth mentioning in this context that the results of modern analytical eforts can be supported as described in detail in Johann Melchior Cröker’s book “Der wohl anführende Mahler …” which appeared in 1743. In the part attached to this volume entitled “Diesem ist noch beygefüget ein Kunst-Cabinet rarer und geheim gehaltener Erfndungen,...” Cröker disclosed detailed recipes (“…rare and secret inventions,…”) concerning the fabrication of colored wax and recommended addition of Venetian turpentine or, in some cases, Cyprian turpentine. Te chromatograms shown in Figure 6B, top panel, were obtained from an anatomic wax model dating from the 18th century, and in Figure 6C, top panel, from a wax model known as “Christ rejected by the Jews”, created ca. 1579 by Giovanni Bologna. Peaks were attributed to resin acids (actually their methyl ester; the binding media were subjected to methylation prior to analysis Figure 6. Comparison of packed column GC in isothermal with to improve volatility) based on comparison with refer- T-programmed mode. Top panel: Isothermal GC of resinous mate- th ence diterpenoic resins. Note that the detection of larixyl rial in (B) an 18 century anatomic wax model; (C) the wax model “Christ rejected by the Jews” by Giovanni Bologna, ca. 1579[58] acetate points to the presence of Venetian turpentine as (chromatographic conditions as in ref. [54]). Peak at 23.5 min, dehy- additive to the wax. Te same resin was detected by cap- droabietate; 37.5 min, larixyl acetate. For details see Supplementary illary GC in samples taken from an anatomic wax model Information and ref. [58]. From ref. [58] with permission. Bottom pan- belonging to the collection of the History of Medicine in el: T-programmed GC. Samples: Waxes from (A) a surface coating th Vienna, Austria, created in 1786 in the workshops of the of a 15 century Intarsia work; contains bees wax, small proportion of ceresin wax; (B) 18th century Italian wax sketch; contains a famous anatomists F. Fontana (in Pisa) and P. Mascagni [60] mixture of ozokerite with traces of bees wax (esters marked B). For (in Florence). details see Supplementary Information and ref. [59]. From ref. [59] Certainly, the peaks of the chromatograms depict- with permission. ed in Figure 6B and 6C (both top panel) are relatively broad, a feature which is characteristic for packed column GC, and their widths increase signifcantly with examples for such mixtures; e.g. beeswax, which mainly increasing retention time, inherent to isothermal condi- consists of straight-chained hydrocarbons with 21 to 33 tions. As pointed out previously, elution at constant col- carbon atoms, and long chain esters with triacontanyl umn temperature leads to unsatisfactory long times of palmitate (containing 46 carbon atoms, see also Figure GC analysis for samples consisting of components difer- 1) being the most abundant one. It is obvious that for ing widely concerning volatility. Natural waxes are good samples with such a composition application of tempera- 106 Ernst Kenndler, Norbert M. Maier ture programmed GC provides the combined advantages of reduced analysis time and narrow peak shapes. Te benefts of temperature programmed GC are evident in chromatograms given in Figure 6A and 6B (both bottom panel), both of which were obtained on a packed column[59]. Te chromatogram in Figure 6A (bottom panel), was obtained from a sample taken from a surface Figure 7. M. Golay´s one of the frst chromatograms with a cap- coating of an Intarsia work dating from the 15th centu- illary column measured in 1956, separating a mixture of isomeric ry, with beeswax being the main constituent. Te chro- pentanes. Column: 366 cm length, 1.37 mm i.d.; stationary phase matogram in Figure 6B (bottom panel) represents the (coating the inner capillary wall), polyethyleneglycol; isothermal at room temperature. Detection with miniaturized thermal conductiv- compound profle of a sample taken from an Italian wax [61] th ity detector. For details see Supplementary Information and ref. . sketch dating from the 18 century, providing evidence From ref. [61] with permission. for the presence of ozokerite, a naturally occurring wax consisting mainly of long-chain n-alkanes. Golay obtained with capillary columns is shown in Fig- ure 7, providing compelling proof that his predictions 7. THE BREAKTHROUGH TO HIGH EFFICIENCY: concerning the improvements in separation efciencies CAPILLARY COLUMNS were consistent with the physical realities. Golay disclosed his theory and the supporting Te transition of packed bed to capillary GC, an experimental results in 1958[62-64], and thus sparked a event that marks undoubtedly a major milestone in the revolution in terms of further developments in gas chro- evolution of GC, is closely connected with the name of matography. His results were successfully reproduced Marcel J.E. Golay. Golay was a Swiss electrical engineer and confrmed by other leaders in the feld, e.g., by D.H. and mathematician; afer having joined Bell Labora- Desty et al.[65, 66] and by R.P.W. Scott[67], and capillary tories, he worked at the U.S. Signal Corps Engineering columns went on to quickly replace packed bed columns Laboratories for 25 years before afliating with a leading essentially for all but preparative applications. instrument company aiming to develop a multiple slit IR Given the impact Golay´s contributions made on the spectrometer. Remarkably, while Golay had no previous [61] state-of the-art in GC, it appears justifed to review some involvement in chromatography , stimulated by the important aspects of his mathematical description of discussion of colleagues engaged in GC, he took inter- the dispersion phenomena in capillary columns. For an est in the mathematics describing the dispersion pro- open tube of cylindrical geometry with a flm of liquid cesses in the packed columns. Trough a critical analysis deposited at its inner surface and with a gas fow veloc- of the basic assumption of the underlying theory, Golay ity, v, Golay established the following equation for the concluded that the irregular pathways the analytes have dependence of H =f(v) to negotiate upon their passage through the particle i beds are the main source of low efciency of packed col- 2 2 2 2 Dm,i (1 + 6 ki +1 1 ki ) r 2 k d umns. Golay suggested that this deleterious efect may H = + v + i s v (9) i v 2 2 4 D 3 2 D be avoidable by employing straight parallel open tubes (1+ ki ) m,i (1 + ki ) s,i (with a thin flm of the stationary liquid phase coating their inner wall) rather than a packed bed giving rise to Tis equation can be expressed in a simplifed form meandering channels. Within few months, he succeeded as in working out a theory of chromatographic dispersion in open tubular columns, and suggested to test his theoretical predictions by experiment. In addition, to facili- Hi = B v +(Cm +Cs )v = B v +C v (10) tate this eforts, he constructed a low-volume thermal conductivity detector better suited for the capillary col- 7 Te frst term, B/v=2Dm,i/v in Equation 9 stands for umn he used (12 f in length, with 0.055 inner diam- the contribution to peak broadening caused by longitu- eter (i.d.), i.e. 366 cm x 1.37 mm) than the contemporary dinal difusion of the analytes in the mobile phase (Dm,i large-volume devices. One of the frst chromatograms is the difusion coefcient of analyte, i, in the mobile phase). Te middle term in Equation 9 (Cm in Equation 7 Nowadays these columns are generally named capillary columns, 10) expresses the combined contributions of the parabol- although Golay preferred the term open tubular columns, because nar- ic fow profle in the cylindrical tube (with inner radius, row tubes - capillaries - could in principle also be packed with particles. Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 107 r) and the difusion in radial direction which is part of lary columns were rapidly abandoned in favor of glass the mass exchange in the mobile phase. Its contribution capillaries afer D.H. Desty et al.[68] had succeeded in the is weighted by the factor which is solely ki-dependent. development of a device to draw reliable capillaries from Te contribution of the kinetics of mass exchange from a molding blank. the stationary phase is expressed by the third term in In the currently most common column type the Equation 9 (Cs in Equation 10); it depends on the flm capillary wall is directly coated with a thin flm of liq- thickness ds on the difusion coefcient in the liquid uid. However, initially, with glass capillaries the direct phase, Ds,i, and on ki. For columns with very thin flms coating approach was met with difculty due to the this term plays a less pronounced role. low wettability of the solid surface (especially for apo- The simplified form of the Golay equation (see lar liquids) and its undesirable adsorptive properties. Equation 10) resembles that of the van Deemter equa- Adsorption at the glass surface, caused by the presence tion for packed columns, with the exception that it lacks of higher-charged cations, led to a severe distortion of term A (simply because the column contains no pack- the peaks with a distinct tailing, resulting from nonlin- ing) and the C-term does not feature any empirical ear adsorption isotherms of the solutes interacting with parameters. Compared with packed columns, Golay`s the liquid-solid interface. Adsorption-triggered tailing approach predicts higher plate numbers for open tubular occurred especially in the case of polar analytes contain- columns for two reasons: i) the attainable plate heights ing donor/acceptor functionalities. A typical example of H are lower, and ii) the absence of a packing results in adsorption-induced peak distortion on a non-deactivat- lower fow resistance and therefore much longer open ed glass surface reported by Schomburg et al.[69] is given tubes can be employed than is possible with densely in Figure 8. Note that in this particular case, symmetri- packed columns. Consequently, the plate number N=L/H cal peak shapes were obtained afer deactivation of the is larger. Since open tubular columns can be operated glass surface. with lengths of up to 100 m, their plate numbers can To improve the quality of glass surfaces, special reach several hundred thousands, being larger by nearly treatments were necessary prior to conducting the two orders of magnitude as compared to the plate num- coating procedure. Tus, wettability was enhanced by ber achievable with packed columns. increasing the roughness of the surface by acidic etch- From the discussions given above it can be concluding, e.g. using dry gaseous HCl or HF.[70] For the deacti- ed that the mobile phase velocity is an important param- vation of the surface a number of other procedures were eter for adjusting the efciency of a column in practice. advanced, e.g. the deposition of polymer layers, or the It shall be mentioned that, since gases are viscous and exhaustive silylation of surface silanol groups. compressible media, the fow velocity is not constant but is a function of the radial and of the axial position in the column. However, the fact of the diferent axial fow velocity will not be further discussed here, as under the usual operation conditions it is a factor of minor signifcance. If being of interest, pressure-dependent velocities can be calculated and accounted for by appropriate cor- rection factors as described by James and Martin[31] or retrieved from the literature.

7.1 Te Improvement of the Capillary Material

7.1.1 Metal and glass capillaries

Te early capillary columns were fabricated from metal, such as copper, nickel, alumina or, preferentially, from steel, or from synthetic organic polymers, and different methods to implement the stationary phase were developed. However, metal columns sufered from some disadvantages, e.g. considerable activity and the uneven- ness of the inner surface, giving rise to band distortion Figure 8. GC with glass capillary without surface deactivation prior to coating with polyethyleneglycol. Analytes 1-9: alkylamines. Fig- and loss of separation efciency. Terefore, metal capil- ure taken from ref. [69] with permission and modifed. 108 Ernst Kenndler, Norbert M. Maier

Figure 9. Lef panel: Early GC with a glass capillary (isothermal mode) for binding media analysis (drying oil). Samples: fatty acids (as methyl esters). Numbering of acids: 1, palmitic (16:0); 2, suberic; 3, azelaic; 4, stearic (18:0); 5, oleic (18:1); 6, linoleic (18:2); 7, linolenic (18:3). For details see Supplementary Information and ref. [71]. From ref. [71] with permission. Right panel: T-programmed capillary GC. Sample: fatty acids (as methyl esters, afer hydrolysis of the sample). Sample taken from the paste layer of a ceremonial shield from the Solo- mon Islands (collected mid-19th century). Time in min. For numbering of analytes see Legend of lef panel. For details see Supplementary Information and ref. [47]. From ref. [47] with permission.

In Figure 9, lef panel, one of the frst applications with all compounds emerging as narrow peaks (Note of glass capillary columns in the area of binding media that in this case the relatively long overall run time was analysis is shown, namely for the separation of fatty chosen deliberately to accommodate for potentially other acids (converted into their methyl esters prior to analy- unknown components in the sample). sis) from a sample of drying oils. Te glass capillary was Even though glass capillary columns found widely coated with a polar stationary phase and GC was carried use in GC due to their high efciency, some problematic out in the isothermal mode. All relevant analytes are properties remained. One of these - but not the most separated. Yet, the peak widths are generally broad, and relevant - was their poor mechanical stability, which the analytes migrate over a relatively wide retention time complicated their handling and installation, especially window of about 10 min. in context with the coupling to mass spectrometers. In contrast, in Figure 9, right panel, a temperature- However, much more problematic was the fact that the programmed analysis of the same group of analytes is coated glass surfaces, despite of careful surface pretreat- shown, originating from the paste layer of a ceremo- ment, ofen retained a certain level of active adsorptive nial shield collected mid-19th century at the Solomon sites. In eforts to address these issues, rather sophisti- Islands (mentioned in Chapter 5.5.1)[47], demonstrating cated protocols were developed for deactivation, coat- the inherent advantages of this elution mode. Specif- ing and stabilization of the stationary phases, involving cally, here the fatty acid profle (also as methyl esters) the cross-linking between polymer chains of the liquid elutes within a retention time window of about 4 min, stationary phase to obtain immobilized phases, and the Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 109 covalent anchoring of stationary phases onto the capillary surface to create bonded phases. Major contributions in the feld of the stationary phase chemistry in context with glass (and in the next generation of fused silica) capillary columns[61] are connected with the names of L. Blomberg[72], K. Grob[73], M.L. Lee[74], S.R. Lipsky[75], C. Madani[76], V. Pretorius[77], P. Sandra[78] and G. Schomburg[79].

7.1.2 Fused silica capillaries

In the course of their pursuit of more suitable glass materials for capillary GC applications, R. Dandeneau and R.H. Zerenner discovered that fused silica proved the most inert material.[80] Afer their initial demonstration of the superiority of fused silica over glass capil- Figure 10. High-temperature GC. Analysis of intact fatty acid laries for GC applications, this material was intensely mono-, di- and triglycerides. Samples extracted from a sherd recov- investigated and further popularized by S.R. Lipsky ered from an Early Medieval ditch. Alumina cladded fused silica et al.[75] and others. At his time, the technology for the capillary column, immobilized stationary phase, end temperature 350°C. For details see text, Supplementary Information and ref. [81]. fabrication of synthetic fused silica was already well From ref. [81] with permission. established in the feld of fber optics, and could be readily adapted for the production of capillaries with little additional efort. Industrially, fused silica is produced silica capillary columns completely replaced glass-based by hydrolysis of SiCl4 in the gas phase and subsequent columns for most but some specialty application (e.g. for melting of the resulting high purity SiO2. Te emerging chiral separations) for which glass capillaries are still the material contains about 0.1 ppm metal oxides as impu- preferred material. rities only, as opposed to naturally occurring quartz, Especially designed stationary phases featuring for which the content of metal ions is typically higher remarkable thermostability allow analysis of high molec- by two orders of magnitude. Due to this high purity, ular compounds. A typical application of high tem- stationary phases coated or immobilized in fused silica perature GC is shown in Figure 10, for profling lipids capillaries were much more stable thermally, and showed contained in a sherd recovered from an Early Medieval much reduced adsorptivity for polar analytes as com- ditch.[81] In this particular case, GC analysis was con- pared to other capillary materials. Furthermore, residual ducted in a temperature-programmed fashion with an activity due to silanol groups could efectively be sup- upper temperature of 350°C, using a fused silica capil- pressed using silylation procedures established in the lary containing a bonded apolar stationary phase. Ana- past for glass surfaces. lytes were intact mono-, di- and triglycerides, which Most procedures originally developed for the fabri- were directly subjected to analysis without any prior cation of immobilized and bonded phases in glass capil- hydrolysis/derivatization (Note that other sample con- laries (see above) could be directly adopted to the fused stituents featuring free carboxylic and hydroxyl groups silica material. Bonded phases, in many cases polysi- were silylated prior to analysis). loxanes derivatives, possess a number of advantages. Due to their attachment at the surface, they are resistant to extraction with organic solvents, and thus do not 7.2 Te Development of Sample Injectors required for Cap- become detached upon direct injection of samples dis- illary Columns solved in polar solvents and even in water. Furthermore, these columns show, yet at high operating temperatures, The plate heights expressed by the van Deemter negligible bleeding (i.e. less release of lower molecu- equation for packed columns and the Golay equation lar mass constituents of the stationary phase at elevated for capillary columns account only for zone broaden- temperature). Also, they exhibit chemically excellent ing processes occurring within the chromatographic long term stability, and maintain over extended periods columns. In other words, it is assumed that the length of use their chromatographic performance in terms of of the injected sample plug and therefore the standard retention and efciency. Given these advantages fused deviation of the input function approach zero. In prac- 110 Ernst Kenndler, Norbert M. Maier tice, however, the injected zone has a fnite length and mass-overload the column as result of its small phase consequently contributes to the fnal peak width. If, e.g., ratio (the inner diameter of the capillary is about 200- the injected plug is rectangular sized with length δ its 300 micrometer, the thickness of the stationary liquid 2 2 variance is σ in j = δ 12 in the length domain). Te stand- is only in the micrometer range or lower) and thus give ard deviation of the fnal peak is then, according to the rise to severely leading peak profles. Terefore, the only 2 2 general additivity of variances, equal to σ in j +σ col with way to preserve the inherently high efciency of capil- 2 σ col being the variance caused by the dispersion process- lary column consists in introducing suitably small sam- es within the chromatographic column.[82] Zone disper- ple amounts in terms of mass and volume. However, as sion contributions from sources outside of the separa- liquid volumes smaller than about 0.1 microliters are tion column, such as the injected plug and the detec- hard to handle by syringes, capillary column generated tor volume, are generally referred to as extra-column the need for especially designed sample introduction efects, and inevitably impact the experimentally observ- devices, e.g. allowing for a partial injection of the sample able peak width. However, the impact of these generally vapor afer being mixed with the carrier gas. unfavorably extra-column contributions on the overall Tis was achieved by engineering injectors that split performance is quite diferent in severity for packed and the gas fow between an additional outlet and the col- capillary columns, respectively. umn. Typically, the outlet consisted of a restrictor which When using packed columns, samples are generally enabled adjusting the desired split ratio between column introduced by an injector which consists of an evapora- and the vent. In practice, split ratios (vent to column) tion chamber, a heated cylindrical glass or quartz tube were selected between several 10:1 and few 100:1. In this with about one mL volume. Te injector is coupled with way, extra-column peak broadening and sample overload an external gas supply which provides a continuous of the capillary column can be avoided. fow of the mobile phase carrier gas. At the outlet of the Tis simple split-injector design, developed in the injector the packed column is mounted, and the inlet of 1960s, and still in use nowadays, has some inherent the injector is typically sealed by a silicon rubber sep- disadvantages, e.g. the marked loss of analyte (via the tum. Te solid or liquid sample is dissolved in a highly vent) and the consequent loss of sensitivity of the ana- volatile solvent, and is injected through the septum by lytical method. To address these limitations, other, more a syringe into the evaporation chamber and where it is sophisticated injector types were developed, e.g. the fash-vaporized. Here the vapor is homogeneously mixed splitless (see e.g. ref. [83]), the Programmed Temperature with the gas phase before it fows into the packed col- Vaporizing (PTV)[84] and the on-column injectors. How- umn. Typically, the injected volumes of the liquid sam- ever, these developments will not be described here. ple solutions are in the microliter range, being volumes In this context it must be pointed out that with that can conveniently be handled by precision syringes. the advent of the capillary column also new sensitivity Tese volumes cause a considerable large injected zone requirements for the detecting systems arose. For packed of vapor at the column inlet, and typically contain a columns initially the thermal conductivity detector was relatively large amount of sample. As pointed out earlier, commonly in use and yet efective. Its relatively large this zone contributes to the width of the fnal peak in volume translates technically into low sensitivity, but addition to the dispersion produced within the column, given the large injected sample typically processed in However, due to the inherently low efciency of packed packed column GC provided satisfactory signal strength. column, the extra-column contributions associated with However, the typical sample amounts eluting from capil- sample introduction remain rather negligible relative to lary GC are much smaller, and would produce a low if the signifcant in-column dispersion. In addition, mass any signal with the traditional detection systems. It was overloading of the packed column by the relatively large therefore a fortunate coincidence, that Golay´s intro- amount of analyte does usually not occur due to the duction of capillary GC was synchronized with the frst large volume of the stationary phase available in the col- detailed technical description of the highly sensitive umn (more currently due to the large phase ratio). fame ionization detector (FID)[85], which could easily be While the described type of injection device is well miniaturized to meet the new demands. Tis detector suitable for sample introduction into packed columns, it responds to CH-groups and is, by the way, thus perfectly is incompatible with the requirements of capillary col- suitable for organic compounds of the binding media. umns. Firstly, a large input sample zone would dominate in-column peak broadening and thus obscure the high efciency of the capillary column. Secondly, introduction of a large amount of analytes would certainly Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 111

8. PYROLYSIS GAS CHROMATOGRAPHY are a well-defned and constant pyrolysis temperature, a precisely controlled heating rate, and a defned dura- As mentioned previously, certain classes of non-vol- tion of the pyrolysis. A gas chromatogram obtained with atile compounds can be addressed by GC analysis afer the pyrolysis GC prototype device mentioned above for a being degraded into smaller volatile fragments through synthetic polymer (applying 500°C for 30 sec in a helium chemically well-controlled reactions (e.g. for proteins atmosphere) is shown in Figure 11, lef panel. and polysaccharides through acid hydrolysis). However, First instrumental improvements of the pyrolysis this conventional approach is inapplicable to materials process were obtained by loading the sample onto an for which no suitable degradation reactions exist (e.g. electrically heated metal wire or into a spiral[86, 88], with high molecular mass condensed hydrocarbons, the main the temperature being calibrated by the melting points constituents of bituminous materials). For these non-vol- of suitable reference compounds. Alternatively, other atiles, pyrolysis GC ofers an attractive option for direct devices for accurate temperature adjustment were using analysis. Tis technique is carried out with GC instru- ferromagnetic metals which, upon placement into a high ments in which the injector system has been modifed frequency inductor coil, were heated exactly to the Curie into a pyrolysis chamber.[86, 87] Afer being loaded into point and maintained a constant temperature by self-sta- the pyrolysis cell, the sample is heated there to high tem- bilization. Modern commercial instruments regulate and peratures at which the contained analytes are thermally control all crucial operational parameters fully electroni- decomposed and the formed products are transferred by cally. Equipped with platinum flaments, state-of-the-art the mobile phase stream into the column. pyrolizers can be operated at temperature up to 1400°C Around 1959 early devices for pyrolysis GC were (which can be kept constant with an accuracy of 1°C), reported, one consisting of a metal loop, which replaced with fash pyrolysis heating rates up to 1000°C/s, and a gas sample loop at the inlet of the packed column.[87] can be programmed to execute in highly reproducible To accomplish thermal sample decomposition, the loop fashion sophisticated user-defned pyrolysis protocols. was heated in a bath of Wood’s Alloy, while the tem- Upon thermally cracking the sample under well- perature achieved by this heat source was measured by defned conditions, the resulting pattern of chromato- a thermocouple connected to a pyrometer. Tis design graphic peaks allows conclusions to be drawn about the showed in principle all necessary functional features of kind of the sample by comparison with reference materi- modern pyrolysis GC, but it lacked provisions for the als and the identifcation of marker peaks. Coupling MS accurate adjustment of the instrumental parameters, detection devices to the GC column provides the pos- which are essential for the reproducible sample decom- sibility to gain detailed structural information. A strik- position and subsequent chromatographic analysis. ing advantages of pyrolysis GC over more conventional Parameters crucial to reliable analysis by pyrolysis GC approaches is that samples can directly be analyzed

Figure 11. Introduction and development of Pyrolysis GC. Lef panel: Early Pyrolysis GC with packed column published in 1959. Sample: polymethylmetacrylate. Packed column, isothermal 100°C; Pyrolysis at 500°C for 30 sec in He atmosphere. Peak annotation: A, air; B, methanol; C, ethanol; E, methyl acrylate; G, methyl methacrylate. For details, see text and ref. [87]. From ref. [87] with permission. Right panel: Pyrolysis capillary GC recorded using a state-of-the-art instrument in 2005. T-programmed mode. Detection: TIC of MS. Sample from a painting in “oleoresin” technique. Peak annotation: 23, hop-22(29)-en-3-ol; 24, β-amyrin; 25, α-amyrin; all as TMS esters. For other peaks see ref. [89]. For details see Supplementary Information and ref. [89]. From ref. [89] with permission. 112 Ernst Kenndler, Norbert M. Maier without any time-consuming pretreatment steps, such In column (2) this fraction is chromatographed simul- as dissolution, hydrolysis, derivatization, etc. Moreover, taneously with the sample components remaining at simultaneous derivatization without elaborate pretreat- column (1). Given that the columns have been selected ment can be carried out in situ by admixing suitable appropriately in term of complementary selectivity, suc- derivatization reagents to the sample in the pyrolysis cessful separation (and detection) of the initially unre- chamber. solved peaks may be achieved. Te improvement of pyrolysis GC instrumentation Certainly, column switching is not limited to only becomes obvious upon comparing the chromatograms one unresolved peak pair, but may be repeatedly applied in Figure 11, lef panel vs. right panel.[89] For the latter to many peak clusters in the same chromatogram. In chromatogram the sample was taken from a painting case that the set of employed columns possess com- executed in “oleoresin” technique by the Mexican artist pletely different chromatographic properties (i.e. are Carmen Lopez, and for which the pyrolysis was carried “orthogonal”), they are considered to represent diferent out using an in situ silylation protocol. As the resulting chromatographic dimensions and the method is named chromatogram is rather shown for comparison, only two-dimensional GC. Tis terminus is taken from two- the peaks of three sample constituents are assigned, dimensional gel electrophoresis and thin layer chroma- i.e. those of the pentacyclic C30H50O triterpenols hop- tography. Te similarity of the chromatographic proper- 22(29)-en-3-ol, and α- and β-amyren. Te results are ties of the stationary phases (and thus the “dimension- indicative for a special resin, Mexican copal, as a constit- ality” of the two combined columns) may be established uent of the painting medium. quantitatively using chemometric methodologies[92, 93], As useful as pyrolysis GC has proven for the analy- and to some extent by comparison of their Rohrschnei- sis of binding media, we wish to place here a word of der-McReynolds indices for the polarity. caution. Despite of the maturity of the method, repro- To the best of our knowledge, column switching ducibility of the results remains an issue, especially approaches have not yet been applied to the analysis of when diferent experimental conditions are applied, and binding media, although it should have a high potential, instrumentation supplied from diferent vendors is used. especially for the separation of the components in very Moreover, it has been pointed out that the quantitation of complex mixtures of diferent binders in the same sample. specifc analytes with pyrolysis GC might be less reliable than that carried out with conventional chromatographic methods. Readers interested in a more in-depth treatise 10. ON-LINE COUPLING GC-MS of the contributions of pyrolysis GC to binding media in objects of cultural heritage are directed to ref. [90]. Technically, coupling a gas chromatograph with a mass spectrometer is a nearly ideal combination of two powerful analytical techniques, because in both meth- 9. COLUMN SWITCHING, HEART CUTTING, TWO- ods the analytes are present in the gaseous phase, albeit DIMENSIONAL GC at pressures that difer by about 8 orders of magnitude. First attempts to realize this attractive option of an on- When being challenged with highly complex sam- line combination of GC with MS were carried out with ples, not all compounds of interest may be successfully packed columns, but were complicated by the large vol- separated with GC employing a single column. Rather, ume of the GC gas fow, which, unsurprisingly, com- a number of analyte zones may co-migrate and form promised the high vacuum conditions required for MS overlapping peaks with the consequence that reliable operation. Solutions to this problem were sought by the identifcation and quantitation remain elusive. In this design of innovative interfaces to harmonize the mutu- case, application of a column with a diferent chromato- al fow requirements of the GC and MS module[94-98], graphic retention characteristic might resolve some criti- respectively, by jet-type or by membrane separators, or cal pairs, but still may fail to separate other compounds by direct open coupling as described in ref. [99]. With the of interest. A solution to this problem is to combine two latter interface packed columns with inner diameters columns of low chromatographic similarity in series, of up to 4 mm were successfully coupled with a double via a switching valve interface as introduced by D.R. focusing MS (see e.g. ref. [100]), though some losses in Deans[91]. Tis method, referred to as heart cutting or sensitivity and an impairment of the detection limit of column switching, enables the transfer of a certain frac- the GC-MS method were resulting. tion of eluate from the frst column (1) onto the second For analyses of binding media, application of packed column (2), both being equipped with separate detectors. column GC-MS coupling emerged in the 1970s, and a Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 113

Figure 12. One of the frst on-line GC-MS for binding media analysis: computer reconstructed plots of (a) TIC of a 16th century ink hydrolysate; (b) mass chromatogram at m/e 458 plus 443 specifc for gallic acid (all analytes as TMS derivatives). Packed column, T-program; MS scan speed 1 set/decade in cyclic mode, period 4 s. Peak numbering: peaks 1, 3, 5, 7, 8: furanose and pyranose epimers of arabinose and galactose, resp.; peaks 10 and 11, α- and β-glucopyranose, resp.; peak 18, gallic acid. For details see Supple- mentary Information and ref. [101]. From ref. [101] with permission. chromatogram documenting one of these early eforts is shown in Figure 12. Tis study was reported in 1977 and aimed at identifying plant gums and gallic acid as possible constituents of an ink sample of a European manuscript on parchment from the 16th century.[101] Prior to Figure 13. GC-MS of the polysaccharide fraction from a sample of analysis, the ink sample was hydrolyzed and the emerg- the Cope of the Virgin Mary (15th century). Te Cope measures 330 ing products subsequently silylated. A packed column of cm in width and 164 cm in length. Peaks: α- and β-glucopyranose. (TMS derivatives afer hydrolysis and concomitant group-separa- 2 mm i.d. was connected to the MS with a single-stage tion by an ion exchanger from the proteinaceous fractions). Insert: jet-type separator. Te chromatograms were reconstruct- GC-MS of barley starch; same procedure as the for sample of the ed by summing up the ion current of each scan. In trace Cope. Time in min. Peaks recorded from TIC. For details see text, (a) the total ion current (TIC) is recorded, and arab- Supplementary information and ref. [102]. From ref. [102] with per- inose, galactose and glucose are identifed. Being isobar- mission. Bottom and lef picture: Long shot and detail of the Cope of ic, the epimers of glucopyranose could not be diferenti- the Virgin Mary. Photographic images by courtesy of KHM-Muse- umsverband, Vienna, Austria. ated by their mass spectra, but plausibly identifed based on their chromatographic retention order. Te extracted dual ion monitoring trace depicted in trace (b) with two at reasonable costs have made GC-quadrupole MS the m/e values specifc for silylated gallic acid allowed its most popular analytical method in laboratories devot- reliable identifcation. Based on these results, the authors ed to the investigation of organic materials in museum concluded that the investigated manuscript was written objects. using ferro-gallic ink containing gum Arabic as a binder. In Figure 13 the result of directly coupled capil- Te mismatch between the gas fow volumes at the lary GC-quadrupole MS for the analysis of a polysac- outlet of the packed GC columns and the inlet into the charide sample is shown[102], processed afer hydrolysis MS could largely be avoided by the use of capillary col- and silylation. Te sample was taken from the pluviale, umns. Due to the lower gas volume emerging from the known as the Cope of the Virgin Mary, a liturgical vest- capillary, the column outlet could directly be coupled via ment of the Order of the Golden Fleece, housed in the a heated transfer line to the MS without impairing its Imperial and Ecclesiastical Treasury in Vienna, Aus- vacuum, enabling the introduction of the entire fow vol- tria (see Figure 13). It was fabricated between 1425 and ume into the ion source of the MS. Moreover, for routine 1440 in Burgundy and most likely designed by the Mas- analyses the sector MS instruments were replaced by the ter of Flémalle (highly probably identifed with Robert less expensive quadrupole mass spectrometers with their Campin, c. 1375 - 1444). Te vestment has been executed much faster scan rate, and with their very user-friendly using two embroidery techniques, viz. in needle paint- operability. Evidently, the combined benefts of a high ing and in or nué, a lazur embroidery technique. Te level of technical maturity, and the ready availability 114 Ernst Kenndler, Norbert M. Maier

Cope measures 330 cm in width and 164 cm in length. Te background structure of the Cope consists of several layers of textile materials stitched together. Origi- nally, these textiles may have been partially congluti- nated, as small white round-shaped scales of about one square millimeter in size were found between the layers, probably applied as a gluing agent. For restoration and conservation purposes, it was of interest to identify the kind of this gluing material. For plausibility reasons, an analytic strategy capable of diferentiate between animal and plant glues was chosen for this project. Prior to GC-MS analysis, a mild hydrolysis protocol suited for a potentially present mixture of proteinaceous and polysaccharide binders was employed, using a cation exchange resin in the H+-form as a hydrolysis reagent.[103] Importantly, this particular procedure provides additional benefts as it permits the separation of the protein hydrolysate from polysaccharide-based materials and prevents these classes of compounds from undergoing condensation reaction with each other. Afer the group-specifc separation of the two binder classes, hydrolysis of the polysaccharide fraction was completed, Figure 14. GC-MS of binding media in objects from Antiquity and and both hydrolysate solutions were subjected to appro- Renaissance. Top panel: TIC chromatogram of sample of embalm- priate silylation protocols. GC-MS analysis of the amino ing materials in Egyptian skulls (neutral fraction from extraction). Sample from the archaeological collection, Natural History acid fraction did not provide any products consistent Museum, Florence, Italy. According analytes as TMS-derivatives. with amino acids, thus the presence of proteinaceous Column, fused silica capillary; T-program, MS: single quadrupole. binder could be safely ruled out. However, in GC-MS For details see Supplementary Information and ref. [107]. From ref. analysis of the derivatized saccharide fraction two main [107] with permission. Bottom panel: capillary GC-MS of amino acid peaks were detected, both of which were identifed by fraction of egg tempera sample from polyptych “Annunciation and Saints” (ca. 1385), Giovanni del Biondo (ca. 1356–1398), Galleria MS being epimeric glucopyranoses; yet eforts towards dell’Accademia, Florence, Italy; sample pretreatment and experi- unambiguous stereochemical assignment of the individ- mental conditions see ref. [104]. Taken from ref. [108] and modifed. ual epimers proved difcult due to their almost identical mass spectra. However, as the α-epimer has a smaller retention factor than the β-epimer, the two peaks could analysis of organic matter in museum objects has signif- [101] be reliably assigned (see also ref. ). Given the fact that cantly increased over the last two decades. Tis is mani- no other saccharides were detected, the unknown mate- fested by the growing number of groups being active in rial employed as a gluing material in the vestment was the feld as well as the ever-increasing number of papers positively identifed as starch (see insert in Fig. 13). Sub- published on the topic. Te objects investigated range sequently, this conclusion was independently confrmed from archeological fnds and artifacts from antiquity[106, by consistent results obtained by analyzing the other 107] (see Figure 14, top panel) over those from Middle possible polysaccharides as reference samples under Ages[108] (Figure 14, bottom panel) up to contemporary identical conditions. art works. It should be pointed out that advanced protocols for As an addendum we mention that the GC-MS pro- group-specifc isolation of lipids, animal and plant glues, tocols developed for the determination of proteinaceous waxes and resins prior to GC-MS analysis are described binders are exceptionally capable of distinguishing [104, 105] in refs. . Tese procedures involve mainly liquid- between classes of parent proteins, i.e. egg, casein or col- solid, solid phase, separation sorbent-tip, and clean-up lagen, but unsuitable for identifying the species of the by ion exchange extraction steps, respectively. animal from which this protein originates. Tis question Historically, only a limited number of investigation can be addressed using bioanalytical methods, origi- of organic matter in museum objects with GC-MS were nally developed for clinical diagnostics, such as enzyme published before the turn of the millennium, but the linked immunosorbent assay (ELISA) or by modern pro- acceptance of this technique as an enabling tool for the teomics protocols. ELISA methodology is based on selec- Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 115 tive binding of the protein of interest to a target-specifc FTIR-MS to the analysis of organic binding media in antibody. Proteomics protocols, recently adapted to museum objects. binding media studies[109], involve the enzymatic diges- tion of the protein sample of interest using appropriate hydrolytic enzymes (typically trypsin), separation and 11. CONCLUSIONS identifcation of the formed peptides via LC-MS techniques, and fnally matching of the observed peptides Te development of GC has been a long journey, with appropriate databases for identifcation of the origi- which started in the late 1800s with the pursuit of meth- nal protein(s). ods for preparative gas separation and ultimately led to The suitability of ELISA and proteomics proto- the establishment of one of the most versatile and sen- cols for the species-specifc identifcation of proteins in sitive modern analytical separation techniques. It needs the binding media has been recently demonstrated for to be emphasized that the evolvement of GC was made Egypt-Romano Portraits dating to 180-200 A.D., for possible through the ingenious contributions of many which cow hides could be established as protein source. outstanding scientists both in the felds of theory and [110] ELISA and proteomics assay were also employed, applied research. Important historic milestones in the together with a number of spectrometric techniques, for development of GC are numerous; these involve i) the the identifcation of the proteinaceous material found early efforts to adapt known adsorption liquid solid in late medieval mortars, and their merits and limita- chromatographic concepts to gas solid separations; ii) tions were critically discussed.[111] A study into the pro- the subsequent transition from solid adsorptive station- teinaceous binders present in the Giant Buddha statues ary phase to solid-supported liquid stationary phases, of Bāmiyān in Afghanistan (largely demolished in 2011) dramatically expanding the scope of interactions forces was carried out using a combination of GC-MS and pro- for tuning separation selectivity; iii) the establishment of teomics techniques. In this case, egg tempera was found a thorough understanding of the parameters that con- in the original paint layer, while cow and goat milk was trol retention and dispersion in chromatographic col- detected in historical overpaintings.[112] umns, and the theoretical understanding of how these Certainly, ELISA and proteomics protocols are use- contributions impact the overall achievable separation ful to complement the knowledge accessible with estab- performance; iv) the following theory-guided transition lished GC-MS methods, allowing identifcation of spe- from packed columns to open tubular capillary columns, cies-specifc sources of proteinaceous binders. Currently, obviating the performance-degrading particle beds and however, these methods are rarely applied. Reasons for allowing for tremendous improvements in separation this reluctance may be the need for specialized instru- efciency by using vastly enhanced column lengths; v) mentation and the high level of expertise required; and and the parallel occurring improvements in chroma- the question whether or not the high costs associated tographic materials and instrumentation, such as inert with these techniques are justifable by additional scien- column materials, thermally highly stable stationary tifc information potentially gained. phases, improved injection systems and highly sensitive To round of the discussion about the role of GC-MS detectors, and, fnally, vi) the optimized coupling of GC in binding media studies, we wish to refect on a very with mass spectrometric detection devices. particular coupling technique evaluated at the end of the The potential of GC for the analysis of binding 1980s by one of the authors (E.K.). In co-operation with media has been recognized as early as in the mid-1960s, a major instrument vendor, the opportunity arose to and has found ever-increasing appreciation as the tech- test an innovative GC-FTIR-MS prototype. In this con- nology approached maturity. State-of-the-art GC instru- fguration, a low-volume IR cell was directly connected ments ofer the benefts of low sample requirements, to the outlet of the capillary column, allowing the non- high sensitivity and separation selectivity, and straight- destructive detection of GC efuents prior to transfer forward analyte identifcation when coupled with mass to the quadrupole MS. Tis GC-FTIR-MS combination spectrometric detection devices. Special GC techniques, was successfully used for the identifcation of some non- such as pyrolysis GC, can be employed to provide valu- resinous compounds detected in the chromatograms of able insights into the compositions of samples difcult to samples of anatomic wax models from the 18th century. characterize by other techniques, such as high molecular [60] However, there may have been little general interest mass compounds and polymers. Given these advantages, in GC-FTIR-MS at this time, as the instrument never GC is currently appreciated as a standard tool in the made it to the market. Tus, our scientifc exploits may feld, and routinely employed for the characterization of well have been the frst and the last application of GC- all important media classes. 116 Ernst Kenndler, Norbert M. Maier

Other techniques are increasingly added to the REFERENCES repertoire of analytical tools employed binding media analysis; especially worth mentioning are dedicated 1. R. Mazzeo, Analytical Chemistry for Cultural Herit- Raman and IR spectrometric techniques, which ofer age, 1st ed., Springer, Switzerland, 2017. the beneft of being fully conservative. Also, new chro- 2. J. S. Mills, R. White, The Organic Chemistry of matographic methods, in particular those involving LC- Museum Objects, 2nd ed., Butterworth-Heinemann, MS proteomics methodology, are fnding recognition Oxford, 2000. as tools for characterization of protein-based binding 3. M. Doerner, Malmaterial und seine Verwendung im media to identify not only the class but also the species- Bilde, 23rd ed., Christophorus Freiburg, 2010. specifc source. For the same purpose, also bioanalytical 4. A. Schönemann, M. Eisbein, A. Unger, M. approaches based on immunoafnity binding, such as Dell`Mour, W. Frenzel, E. Kenndler, Stud. Conserv. ELISA are employed. 2008, 53, 118. While the advent of these new methods will cer- 5. A. Schönemann, W. Frenzel, A. Unger, E. Kenndler, tainly enhance the information accessible through bind- Stud. Conserv. 2006, 51, 99. ing media analysis, it is unlikely that they will replace 6. L. Reed, Proc. Amer. Chem. Soc. 1883, 9, 123. GC as the prime enabling technique. GC, while being a 7. D. T. Day, Proc. Amer. Phil. Soc. 1897, 36, 112. micro-destructive analytical method, is uniquely suited 8. M. S. Tswett, Trudy Varhavskago Obshchestva este- for binding media analysis with regard to its low sam- voispytatelei Otd. Bio.l (T.r Warsawsk Obst. Jestesv. ple requirements, the broad scope of materials that can Otd. Biol.); Proc. Warsaw Nat. Sci. Biol. Sec. 1903, be analyzed, its inherently high sensitivity, and its per- 14, 20. fect compatibility with mass spectrometric devices. Also, 9. M. S. Tswett, Ber. Dtsch. Botan. Ges. 1906, 24, 316. there is a uniquely rich body of expertise and experience 10. M. S. Tswett, Ber. Dtsch. Botan. Ges. 1906, 24, 384. with GC in binding media analysis that has been com- 11. R. Kuhn, E. Lederer, Naturwissensch. 1931, 19, 306. piled over fve decades, which facilitates the interpreta- 12. R. Kuhn, E. Lederer, Ber. Dtsch. Chem. Ges. 1931, tion of results, and is invaluable for guiding the design 64, 1349. and execution of new studies. 13. R. Kuhn, E. Lederer, Zeitschrift physiol. Chemie Future progress in the field of GC-based binding 1931, 200, 246. media analysis may involve improvements in detection 14. R. Kuhn, A. E. Winterstein, E. Lederer, Hoppe-Sey- sensitivity by new generations of mass spectrometric lers Z. physiol. Chem. 1931, 197. devices, which is anticipated to further reduce sample 15. L. Zechmeister, L. von Cholnoky, in Die chroma- amount requirements. Also, future binding media analy- tographische Adsorptionsmethode, Springer, Vienna, sis may potentially proft from the implementation of the 1937. emerging comprehensive two-dimensional GC technology 16. A. J. P. Martin, R. L. M. Synge, Biochem. J. 1941, using stationary phases with diferent retention character- 1358. istics, resulting in enhanced information outputs to better 17. A. J. P. Martin, R. L. M. Synge, Biochem. J. 1941, justify the use of precious and unique sample materials. 1364. Arguably, GC will remain a technique of central 18. E. Berl, K. Andress, Angew. Chem. 1921, 34, 378. importance in the feld of binding media analytics in 19. G. S. Bohart, E. O. Adams, J. Am. Chem. Soc. 1929, cultural heritage research and preservation. 42, 523. 20. N. Schilow, L. Lepin, S. Wosnessensky, Koloid-Z. 1929, 49, 288 ACKNOWLEDGEMENTS 21. G. Damköhler, H. Teile, Die Chemie 1943, 56, 353. 22. E. Cremer, R. Müller, Microchim. Acta 1951, 36/37, The authors gratefully acknowledge KHM-Muse- 553. umsverband, Vienna, Austria, for the kind permission to 23. E. Cremer, R. Müller, Z. Elektrochem. 1951, 55, 217. use the photographic material of the Cope of the Virgin 24. E. Cremer, F. Prior, Z. Elektrochem. 1951, 55, 66. Mary. 25. A. V. Kiselev, K. D. Stscherbakowa, in Gas Chro- matographie 1961 (Eds.: M. Schröter, K. Metzner), Akadem.-Verl., Berlin, 1961, p. 207. COMPETING INTERESTS 26. I. G. Kolomnikov, A. M. Efremov, T. I. Tikhomiro- va, N. M. Sorokina, Y. A. Zolotova, J. Chromatogr. A Te authors declare no competing interests. 2018, 1537, 109. Gas Chromatography and Analysis of Binding Media of Museum Objects: A Historical Perspective 117

27. H. Brunschwig, Liber de arte Distillandi de Com- Stud. Conserv. 1968, 13, 105. positis. Das buch der waren kunst zu distilieren die 53. L. Masschelein-Kleiner, J. B. Heylen, Stud. Conserv. Composita, Johann Grüninger, Straßburg 1512. 1968, 13, 87. 28. A. Bittel, Tesis (Univ. Tübingen), 1957. 54. J. S. Mills, Phytochem. 1973, 22, 2407. 29. H. G. Struppe, in Handbuch der Gaschromatogra- 55. J. S. Mills, R. White, Stud. Conserv. 1975, 20, 176. phie, 3 ed. (Eds.: E. Leibnitz, H. G. Struppe), Aka- 56. J. S. Mills, Conserv. Restor. Pict. Art 1976, 69. dem. Verlagsges. Geest & Portig, Leipzig, 1984, pp. 57. J. S. Mills, R. White, Conserv. Restor. Pict. Art 1976, 22. 72. 30. A. T. James, A. J. P. Martin, Analyst 1952, 77, 915. 58. J. S. Mills, R. White, Stud. Conserv. 1977, 22, 12. 31. A. T. James, A. J. P. Martin, Biochem. J. 1952, 50, 59. R. White, Stud. Conserv. 1978, 23, 57. 679. 60. E. Kenndler, F. Mairinger, Fresenius J. Anal.Chem. 32. A. I. M. Keulemans, A. Kwantes, P. Zaal, Anal. 1990, 338, 635. Chim. Acta 1955, 13, 377. 61. L. S. Ettre, LCGC 2001, 19, 49. 33. E. F. G. Herington, in Vapor Phase Chromatography 62. M. J. E. Golay, in Gas Chromatography 1958 (Ed.: (Ed.: D. H. Desty), Butterworths Sci. Publ., London, D. H. Desty), Butterworth Sci. Publ., London, 1958, 1957, p. 5. pp. 36. 34. A. A. Zhukhovitskii, N. M. Turkeltaub, T. V. Geor- 63. M. J. E. Golay, Nature 1958, 182, 1146. gievskaya, Nauk SSSR 1953, 92, 987. 64. M. J. E. Golay, in Gas Chromatography (Eds.: V. J. 35. A. A. Zhukhovitskii, O. V. Zolotareva, V. A. Coates, H. J. Noebels, I. S. Fagerson), Acad. Press, Sokolov, N. M. Turkeltaub, Dokl. Akad. Nauk SSSR New York, 1958, pp. 1. 1951, 77, 435. 65. D. H. Desty, in Gas-Chromatographie 1958 (Ed.: 36. S. Dal Nogare, in Gas Chromatography 1958 (Ed.: D. H.-P. Angele), Akadem.-Verl., Berlin, 1959, p. 176. H. Desty), Butterworths Sci. Publ., London, 1958, p. 66. D. H. Desty, A. Goldup, B. H. F. Whyman, J. Inst. 242. Petroleum 1959, 45, 287. 37. S. Dal Nogare, L. W. Safranski, Anal. Chem. 1958 67. R. P. W. Scott, Nature (London) 1959, 183, 1753. 30·, 894. 68. D. H. Desty, J. N. Haresnape, B. H. F. Whyman, 38. J. C. Giddings, Anal. Chem. 1962, 34, 722. Anal. Chem. 1960, 32, 302. 39. J. C. Giddings, Dynamics of Chromatography; Part 69. G. Schomburg, R. Dielmann, H. Husmann, F. 1, Principles and Teory, Marcel Dekker, New York, Weeke, J. Chromatogr. 1976, 122, 55. 1965. 70. M. Novotny, K. Tesarik, Chromatographia 1968, 1, 40. J. N. Wilson, J. Am. Chem. Soc. 1940, 62, 1583. 332. 41. E. Glueckauf, in Ion Exchange and Its Applications, 71. L. Masschelein-Kleiner, in PACT Vol. 13, Athens, Vol. 4061, Society of the Chemical Industry, Lon- Greece, 1986, pp. 185. don, 1955, pp. 1540. 72. L. Blomberg, J. Buijten, J. Gawdzik, Chromatograph- 42. E. Glueckauf, in Ion Exchange and Its Applications, ia 1978, 11, 521. Vol. 4061, Society of the Chemical Industry, Lon- 73. K. Grob, G. Grob, K. J. Grob, J. HRC&CC 1979, 31. don, 1955, pp. 34. 74. B. E. Richter, J. C. Kuel, L. W. Castle, B.-A. Jones, J. 43. J. J. van Deemter, F. J. Zuiderweg, A. Klinkenberg, S. Bradshaw, M. L. Lee, Chromatographia 1983, 17, Chem. Eng. Science 1956, 5, 271. 570. 44. L. Lapidus, N. R. Amundson, J. Phys. Chem. 1952, 75. S. R. Lipsky, W. J. McMurray, M. Hernandez, J. E. 56, 984. Purcell, K. A. Billeb, J. Chromatogr. Sci. 1980, 10, 1. 45. J. C. Giddings, Anal. Chem. 1963, 35, 439. 76. C. Madani, M. Rigand, J. Durand, P. Chebronx, J. 46. E. Kováts, Helv. chim. Acta 1958, 41, 1915. Chromatogr. 1976, 126, 161. 47. E. Kenndler, F. Mairinger, W. P. Bauer, F. Rainer, K. 77. V. Pretorius, J. C. Davidtz, J. HRC&CC 1979, 2, 703. Schmidt-Beiwl, Stud. Conserv. 1988, 33, 115. 78. P. Sandra, G. Edant, E. Schacht, M. J. Verzele, J. 48. L. Rohrschneider, J. Chromatogr. 1966, 22, 6. HRC&CC 1981, 4, 411. 49. W. O. McReynolds, J. Chromatogr. Sci. 1970, 8, 685. 79. G. Schomburg, H. Husmann, F. Weeke, J. Chroma- 50. N. Stolow, in Application of Science in the Examina- togr. 1974, 99, 63. tion of Works of Art, Vol. II, Museum of Fine Arts, 80. R. Dandeneau, R. H. Zerenner, J. HRG&CC 1979, 2, Boston, 1965, pp. 172. 351. 51. J. S. Mills, Stud. Conserv. 1966, 11, 92. 81. R. P. Evershed, C. Heron, L. J. Goad, Analyst 1990, 52. L. Masschelein-Kleiner, J. Heylan, F. Tricot-Marckx, 115, 1339. 118 Ernst Kenndler, Norbert M. Maier

82. J. C. Sternberg, in Adv. Chromatogr., Vol. 2 (Ed.: J. 99. D. Henneberg, in Gas Chromatography 1960 (Ed.: R. C. Giddings), Marcel Dekker, New York, 1966, pp. P. W. Scott), Butterworth Sci. Publ., London, 1960, 205. pp. 129. 83. K. Grob, G. Grob, J. Chromatogr. Sci. 1969, 7, 584. 100. J. F. K. Huber, E. Matisova, E. Kenndler, Anal. 84. F. S. Poy, F. Visani, F. Terrosi, J. Chromatogr. 1981, Chem. 1982, 54, 1297. 217 8. 101. P. Arpino, J. P. Moreau, C. Oruezabal, F. Flieder, J. 85. I. G. McWilliam, R. A. Dewar, in Gas Chromatogra- Chromatogr. 1977, 134, 433. phy 1958 (Ed.: D. H. Desty), Butterworths, London, 102. U. Schneider, E. Kenndler, Fresenius J. Anal. Chem. Amsterdam 1958, pp. 142. 2001, 371, 81. 86. R. C. Lehrle, J. E. Robb, Nature (London) 1959, 183, 103. E. Kenndler, K. Schmidt-Beiwl, F. Mairinger, M. 1671. Pöhm, Fresenius J. Anal.Chem. 1992, 342, 135. 87. E. A. Radell, H. G. Strutz, Anal. Chem. 1959, 31, 104. A. Andreotti, I. Bonaduce, M. P. Colombini, F. 1890. Modugno, E. Ribechini, N. Tr. Anal. Environ. Cult. 88. J. Janak, Collect. czechoslov. chem. Commun. 1960, Herit. Chem. 2008, 389. 25, 1780. 105. A. Lluveras, I. Bonaduce, A. Andreotti, M. P. 89. J. de la Cruz-Canizares, M. T. Domenech-Carbo, J. Colombini, Anal. Chem. 2010, 82, 376. V. Gimeno-Adelantado, R. Mateo-Castro, F. Bosch- 106. J. J. Lucejko, F. Modugno, M. P. Colombini, M. Reig, J. Chromatogr. A 2005, 1093, 177. Zborowska, J. Cult. Herit. 2012, 13, S50. 90. G. Chiavari, G. C. Galletti, G. Lanterna, R. Mazzeo, 107. J. J. Lucejko, A. Lluveras-Tenorio, F. Modugno, E. J. Anal. Appl. Pyrolysis 1993, 24, 227. Ribechini, M. P. Colombini, Microchem. J. 2012, 91. D. R. Deans, Chromatographia 1968, 1, 18. 103, 110. 92. C. Muscat Galea, D. Mangelings, Y. Vander Heyden, 108. M. Serefdou, S. Bracci, D. Tapete, A. Andreotti, L. in Chemometrics in Chromatography, 1st ed. (Eds.: Biondi, M. P. Colombini, C. Giannini, D. Parenti, Ł. Komsta, Y. Heyden, J. Sherma), CRC Press, Boca Microchem. J. 2016, 127, 187. Raton, 2018, pp. 45. 109. C. Tokarski, E. Martin, C. Rolando, C. Cren-Olivé, 93. J. F. K. Huber, E. Kenndler, G. Reich, J. Chromatogr. Anal. Chem. 2006, 78, 1494. 1979, 172, 15. 110. J. Mazurek, M. Svoboda, J. Maish, K. Kawahara, S. 94. R. S. Gohlke, Anal. Chem. 1959, 31, 535. Fukakusa, T. Nakazawa, T. Yoko, e-Preservation Sci- 95. D. Henneberg, Z. Anal. Chem. 1961, 183, 12. ence 2014, 11, 76. 96. J. C. Holmes, F. A. Morell, Appl. Spectrosc. 1957, 17, 111. I. Krizova, J. Schultz, I. Nemec, R. Cabala, R. Hynek, 86. S. Kuckova, Anal. Bioanal. Chem. 2018, 410, 189. 97. R. Rhyhage, Anal. Chem. 1964, 36, 759. 112. A. Lluveras-Tenorio, R. Vinciguerra, E. Galano, 98. J. T. Watson, K. Biemann, Anal. Chem. 1964, 36, C. Blaensdorf, E. Emmerling, M. P. Colombini, L. 1135. Birolo, I. Bonaduce, PLoS ONE 2017, 12, 1. Firenze University Press www.fupress.com/substantia

Historical Article Exact Time: the First Scientifc Application of Radiocommunications Citation: M. Calamia, M. Gherardelli (2018) Exact Time: the First Scien- tific Application of Radiocommunica- tions. Substantia 2(2): 119-123. doi: Mario Calamia, Monica Gherardelli 10.13128/substantia-65 Dipartimento di Ingegneria dell’Informazione, Università degli Studi di Firenze (Italy) Copyright: © 2018 M. Calamia, M. E-mail: [email protected]; [email protected] Gherardelli. This is an open access, peer-reviewed article published by Firenze University Press (http://www. Abstract. Marconi’s frst experiment of signal transmission by means of Hertzian fupress.com/substantia) and distribuit- waves was carried out in 1895. In the following years, wireless telegraphy progressed ed under the terms of the Creative steadily and worldwide eforts were made to exploit the potential ofered by new tech- Commons Attribution License, which nologies. In those years Guido Alfani, a young Florentine Piarist teacher of promise in permits unrestricted use, distribution, Seismology, joined the Ximeniano Observatory in Florence where he found the ideal and reproduction in any medium, provided the original author and source environment for his experiments and his insights. He understood the importance of are credited. having the exact time in Seismology, to temporally characterize the telluric movements and therefore accurately characterize them. In 1910 when the Paris radio station Data Availability Statement: All rel- located at the Tour Eifel began regular broadcasts of exact time, he laid down the issue evant data are within the paper and its of its reception. As far as pendulums and chronometers were concerned, no doubt his Supporting Information fles. expertise as seismologist was signifcant, while problems arose when it came to the Competing Interests: The Author(s) radio station, due to the novelty of such situation. For this reason he arranged contacts declare(s) no confict of interest. and managed to set the frst Italian radio station to be used in a weather station. Tus, on the night of March 16-17, 1912, he received for the frst time the time signal for a particular scientifc application. He wrote to Marconi and in 1912 Marconi expressed words of great appreciation and encouragement for such work. Father Guido Alfani’s radio station is certainly the frst one applied in Seismology and among the frst radios made in Italy. It is an extremely important application which demonstrated that the new technique could provide solutions in diferent situations.

Keywords. Exact time, radiocommunications, Osservatorio Ximeniano.

INTRODUCTION

Te second half of the nineteenth century has been a period where science met with great ferment. Important pages on discoveries or great scientifc activities have been written and ofen people are lef wondering on how it was possible to collect such amazing results in such a short and precise time span. To keep to the topic of our contribution, we will point out that in the second half of the nineteenth century Seismology became a science thanks to some great scientists. Among them, father Giuseppe Mercalli (the Mercal- li intensity scale used nowadays to classify earthquakes goes back to those years, namely 1880-90) and all over the world eforts were made to compare both results and experiences.

Substantia. An International Journal of the History of Chemistry 2(2): 119-123, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-65 120 Mario Calamia, Monica Gherardelli

Since the second half of the eighteenth century the ed his eforts towards “fxing and keeping” the exact Osservatorio Ximeniano has been working in Florence time by means of spyglasses, so as to measure culmina- as a research institution founded by the Jesuit father tion in star transit, as well as by means of pendulums to Leonardo Ximenes and later on run by the Piarists. In keep the correct measure of time. Let’s read once again 1872 father Filippo Cecchi was appointed Director of his words: the Osservatorio Ximeniano; he was a skilful meteor- ologist who has lef important contributions in weather In modern seismic stations, the accuracy of time is not only forecasting. Furthermore, he was not at all indiferent one of the important element, but even the most important to the great deal of progress made by Seismology and one. For such reasons, there will never be too much care to nowadays his works are considered of greatest value. He get such accuracy and keep it unchanged and exact. As a matter of fact, this Observatory has been working mainly conceived the frst “three-component seismograph” and on Seismology, therefore I had great care to defne the exact many other tools, which can still be seen in the exhibi- time and make it fawless. Terefore, I have arranged a tion room dedicated to his studies in the Osservatorio very special equipment to guarantee results.2 Ximeniano. Finally, the very last years of the nineteenth centu- …omissis…. ry could witness for other important scientifc developments, thanks to the arrival to the Osservatorio Xime- Until 1912 (this was the year I could realize the very frst niano of a young Florentine teacher of mathematics and set of equipment related to a radiotelegraphy station dedi- physics. Father Guido Alfani, who belonged to the Pia- cated to radio time signals) the exact time reckoning has rist order, was skillful and self-assertive, which helped been carried out on a regular basis, nearly every evening, when he became Director of the Osservatorio Ximeni- and at least every two days by means of star culmination measured at meridian circle.2 ano in 1905.1 In the wake of Father Filippo Cecchi, Father Guido The “very special equipment” is in fact the first Alfani became soon an important reference in Seismol- Radio Station he realized in Italy. ogy, he conceived new seismographic instruments, he In 1910 the Bureau Central in Paris, namely the kept on improving already existent seismograms, while French Observatory located at the Tour Eifel, began to studying deeply any electrodynamics tools. transmit the exact time signal and it was at that point His contributions to the seismology framework have Father Alfani realized that such transmittance could been of primary importance. serve perfectly the purpose of supplementing the star culmination method. He needed a suitable receiver, he integrated his training, he searched for information and FATHER GUIDO ALFANI AND THE SERVICE OF EXACT TIME he was able to set up a radiotelegraphy station and in the night between the 16th and 17th of March, he received Te main focus of this paper is not only the Father for the frst time the exact time from Paris. Alfani’s experiences, but through his experiences we It was a crucial step forward in Seismology. No would like to pay homage to many Scientists who could longer depending on star observation at a specifc time, understand the importance of interdependency in the but relying on a regular time signal, as the one trans- scientifc felds. mitted from Paris, that was an absolutely exceptional As to Seismology, the activities carried out by Father result. Father Guido Alfani was the frst to achieve such Alfani have been presented in many excellent papers, reception and he dedicated the discovery to Guglielmo which have already highlighted the importance of his Marconi, thus writing: “I would like to dedicate to Gug- lielmo Marconi these pages on the frst radiotelegraphy contribution to that area. Terefore it is not our inten- 3 tion to focus on such aspect. station working in an Italian Observatory”. We would like to focus on the work of Father Alfa- Te frst radio station was nothing else than a start- ni from a diferent perspective: he was a mathematician ing point for Father Alfani and other seismologists who with strong interests in Seismology and even though he followed in his footsteps. Te course of events related to was not a radio engineer, his insights enabled him to those years is well described in the quoted passage dedi- understand that what he found in a diferent scientifc cated to Guglielmo Marconi. feld could be of great help to Seismology. Main components of his radio equipment were He sensed that knowing the exact time was a funda- the receiver itself and the aerial (the antenna). As to mental issue in earthquake studies. Terefore, he direct- the receiver, he resorted to the expertise and helpful- ness of Ducretet & Roger Company and he used “elec- Exact Time: the First Scientifc Application of Radiocommunications 121

Figure 1. Te Ximeniano Observatory seen from South and showing the antenna of the radiotelegraphy station.3 trolytic diodes” because “they are much more constant than crystal sets”. He had to deal with some drawbacks he promptly pointed out to the Company, thus quickly managing to have reliable components. He kept on working on the improvement of the receiving system; thus in 1923 he exploited the coherer as radio detector in his receiver, only to change in favor of vacuum tubes, Figure 2. Drawing of the second (upper) and third (lower) radio once their checkout had turned out to be trustworthy. equipment related to the aerial.3 Yet, it is the work on the antenna, which took up almost his time. He sensed that the shape of the antenna and the location where to place it were very important spy ring. As a scientist he cannot accept nor justify any factors in defning the quality of the received signal and stop to the progress of Science; therefore he builds and without such signal there was not too much to detect. adopts quad antennas for his receiver device: once over- When describing the types of “aerials” he realized at come the problems of tuning them with the receiver’s the Ximeniano Observatory (Figure 1 and 2) he wrote: interface, the reception capacity of such antennas turns out to be greater and allows him to work. One of such aerials is really great; it is dipole and has its Figure 3 shows quad antennas, two with rhombic point of support in the Dome of the Florence Duomo. It is shape and one with round shape. made of two wires in phosphor bronze and it is about 300 Tis is not just a storytelling retrieved from books, meters long and 110 meters high. Tere is an iron pole (15 which is of great value in itself, but it has been a work of meters long) which has in its upper end some pulleys to reconstruction based on collected parts which have been be used to rise or drop the aerial’s bays by means of steel brought into use again, thus allowing the recovery of a cables.2 priceless heritage, along the last ten years. In the storerooms at the Ximeniano Observa- However, not everything goes so easy as it might tory many items were found and they were supposed appear: the country is going to war (1915-1918) and to be components of the radiotelegraphy station built therefore the government imposes restrictions such as by Father Alfani: only a picture could witness that in the interruption of the broadcasting station services and the booklet already mentioned and entitled Te time the removal of the antennas at the Observatory. Tere is service.2 One of the authors of this paper, prof. Mario a very interesting exchange of letters describing the heat- Calamia, was informed about the possible and yet pared argument between Father Alfani and the Government tial recovery of that equipment: the task seemed to be Authorities. His aim is to stand up for his activities with not easy due also to the lack of students to be involved. the radiotelegraphy station described as a support to his Eventually in 2006 a solution was within reach thanks seismology studies and not at all meant to be part of a to Giovanni Manneschi, an engineer and Arezzo “CEIA 122 Mario Calamia, Monica Gherardelli

Figure 5. Radio receiver built in 1923 according to the model of Marconi’s receiver dating back to 1895.4 A part from the battery box, all the other components like the coherer (right lower part, near to the antenna) are original.

CONCLUSIONS Figure 3. Te radiotelegraphy station of Father Alfani at the Xime- niano Observatory.2 Our goal has been focused on informing about events which may seem of minor importance, but which had a great social and scientifc impact. Tis generally happens when Science manages activities in some facilities. Te Ximeniano Observatory has been one of such facilities and yet its renown is mainly related to Mete- orology and Seismology. It has a very relevant map pro- jection Department and it is also well-known for Father Eugenio Barsanti’s work on the internal combustion engine whose frst version he invented. On the other hand, the Observatory played a leading role in radio engineering during the frst half of the 20th century, but this aspect has been a bit underestimated. Figure 4. Radio receiver built in 1912, the model of it is in the Te research carried out by Father Alfani had the merit booklet Father Alfani dedicated to Marconi during his visit held in 1912.4 Te diode, head-phones and the capacitor are original. of improving radio engineering as well, though his name became notorious for meteorology and seismology studies. S.p.A” executive, but also very keen on radio engineer- He understood that the main shortcoming in Seis- ing. Since then a very methodical work has begun to mology was the inability to correlate the diferent tellu- recover, restore and put that Laboratory (Figure 3) back ric phenomena detected along far-of areas. Tis could on its feet. It took about six/seven years to fnish the be solved if every Observatory was able to beneft from work, but nowadays more than 95% of Father Alfani’s the exact time service and he had the insight that this Laboratory is on exhibit at the Ximeniano Observatory.4 was possible when he heard of Paris radio station’s regu- It is made of 44 parts ranging from the radiotelegraphy lar broadcast of time signal. station built in 1912 and already depicted in Figure 4 to He did not purchase a receiver, which was not pos- several other equipments dating back to 1940 and per- sible, but he improved his learning on radio engineering, fectly operative. Te radiotelegraphy station has been so as to defne the requirements which could make the restored by using the original Ducretet & Roger thermi- reception more and more accurate. onic diode, which was found by chance in the Observa- What he wrote in his short, and yet regularly dated tory’s store-rooms. Actually the station has been tuned and published notes, shows clearly his cleverness. Gug- again so as to receive the national broadcast “RAI” in lielmo Marconi, invited to Florence, met full of admi- amplitude modulation. Another noteworthy article is the ration Father Alfani at the Observatory. In fact, Father radio receiver dating back to 1923 and shown in Figure Alfani had used radio broadcast not only in telecommu- 5, with its original coherer. nication engineering, but also in a scientifc feld of great social impact. Exact Time: the First Scientifc Application of Radiocommunications 123

Finally, we would like to stress that he was never REFERENCES particular about his studies, experiments and outcomes. As a scientist, he was persuaded that the more the 1. D. Barsanti, Padre Guido Alfani (1876-1940), Observatories adopted the proposed solution, the greater Osservatorio Ximeniano dei PP. Scolopi, Firenze, would be the usefulness of that research. 1992. His words are revealing: 2. G. Alfani, Il servizio dell’ora, Pubblicazione 136, Osservatorio Ximeniano dei PP. Scolopi, Firenze, Te frst radiotelegraphic reception occured in the night of 1928. the 19th March 1912 , based on Paris time signals. Shortly 3. G. Alfani, La stazione radiotelegrafca, Pubblicazione aferwards, other colleagues joined and followed my exam- 115, Osservatorio Ximeniano dei PP. Scolopi, Fire- ple. I went personally to set up new equipments. One of nze, 1912. the frst radio station I installed was at the Montecassini 4. M. Calamia, G. Manneschi, Kernes: la rivista del res- Observatory on the 26th July 1913.2 tauro, N. 83 (Eds: Nardini), Firenze, July-September 2011.

Firenze University Press www.fupress.com/substantia

Historical Article Te dextrorotatory sweet asparagine of Arnaldo Piutti: the original product is conserved in Citation: L. Colli, A. Guarna (2018) The dextrorotatory sweet asparagine Florence of Arnaldo Piutti: the original product is conserved in Florence. Substantia 2(2): 125-130. doi: 10.13128/substantia-66 Laura Colli*, Antonio Guarna Copyright: © 2018 L. Colli, A. Guarna. This is an open access, peer-reviewed Dipartimento di Chimica “Ugo Schif” dell’Università degli Studi di Firenze, Via della article published by Firenze University Lastruccia 3-13, Polo Scientifco e Tecnologico, 50019 Sesto Fiorentino – Firenze (Italy) Press (http://www.fupress.com/substan- E-mail: [email protected] tia) and distribuited under the terms of the Creative Commons Attribution License, which permits unrestricted Abstract. In 1886, Pasteur presented a note on the work of the Italian chemist Arnaldo use, distribution, and reproduction Piutti concerning the diference between the two physical isomers (enantiomers) of in any medium, provided the original asparagine. Te octahedral crystal of asparagine appeared only as “levorotatory hemi- author and source are credited. hedralism” but, in principle, should also exist as a dextrorotatory asparagine with a Data Availability Statement: All rel- symmetric crystalline form. In 1886 Arnaldo Piutti isolated the dextrorotatory aspara- evant data are within the paper and its gines while he was working as an assistant of Ugo Schif in Florence. He obtained also Supporting Information fles. another unexpected information, of which only Pasteur immediately understood the importance: the dextrorotatory aspargine had a sweet taste. Competing Interests: The Author(s) Te dextrorotatory sweet asparagine of Arnaldo Piutti is conserved in the Schif Col- declare(s) no confict of interest. lection of the Department of Chemistry “Ugo Schif” at the University of Florence, and is the frst compound where a relationship between the optical isomerism of a molecule and a diferent response of human receptors, in this case the taste, was observed.

Keywords. Asparagine, Chemical Heritage, history of enantioselectivity.

Tis research stems from a purely museological question. A few years ago, during the reordering of the products synthesized by Ugo Schif (1834- 1915) at that time conserved at the Department of Chemistry “Ugo Schif” of the University of Florence, and now merged into the Museum of Natural History, a small bottle was found. It carried a “Ugo Schif” museum label, written by “our” German chemist, referring to: “Asparagina destrogira dolce” (Dextrorotatory sweet asparagine)1. Afer reading the work by Joseph Gal2-4 in 2008, Antonio Guarna (the scientifc director of the museum project for historical chemical fnds) immediately had the intuition that the product could contain the famous dextrorotatory sweet asparagine isolated by Arnaldo Piutti. Te historical research that was conducted, within the Chemical Her- itage project of the University of Florence, proved him right. In this contribution we discuss some insights about the discoverer and how this discovery took place.

Substantia. An International Journal of the History of Chemistry 2(2): 125-130, 2018 ISSN 1827-9635 (print) | ISSN 1827-9643 (online) | DOI: 10.13128/substantia-66 126 Laura Colli, Antonio Guarna

“PIUTTI’S ASPARAGINE”: WAS IT AN IMPORTANT proposed that compounds with asymmetric carbon may DISCOVERY? exist in two diferent forms corresponding to the two optical antipodes, which could be separated into two Te work of Gal on the discovery of Piutti reports in optically active compounds. Van’t Hof’s theory, consid- the conclusions, the following statements4: ered too audacious and abstract, was almost rejected by the academic world. Only when Hermann Emil Fischer Piutti was a highly original chemist who carried out nota- (1852- 1919), a German chemist, almost contemporary of ble investigations in a wide variety of research topics. His Schif, and Nobel Prize winner in 1902, in 1894 adduced discovery of a diference in the taste of D- and L-aspara- evidence in favor of this thesis by synthesizing the levo- gine was a milestone frst observation of enantioselectiv- rotatory (not natural) antipode of glucose, the proposi- ity at a biological (human) receptor. Te discovery was also the frst observation of stereoselectivity of any kind in tions of van‘t Hof became accepted by the Academics, 7,8 taste; the frst fnding of biological enantioselectivity in an paving the way to modern stereochemistry . organism higher than microorganisms; the frst example of biological enantioselectivity in an efect other than enzyme action; and one of the two earliest reports of the prepara- ARNALDO PIUTTI BIOGRAPHY tion of a D-amino acid. Arnaldo Piutti was born on January 23, 1857 in Te sample preserved in the Schif Collection of the Udine. He graduated in 1875 at the Technical Institute University of Florence therefore has a considerable his- of Udine, in the Physics-Mathematics section and from torical and scientifc relevance. Te discovery had a good there he moved to Turin where he enrolled in the Fac- resonance at the time of Piutti. Louis Pasteur (1822-1895) ulty of Natural Sciences9. In 1879 he brilliantly com- himself showed a great interest in this research, and in pleted his studies under the guidance of Ugo Schiff 1886, at the French Academy of Sciences, commented on (1834-1915), a German chemist that served as professor the work of the Italian chemist with these words5: of chemistry in Florence for ffy years and in his Floren- tine laboratory discovered the bases and the Schif reac- Why this big diference in taste between these two aspara- tion10. Piutti had the opportunity to meet Schif just in gines? One might assume the existence of a very special the two years when the German scientist was lecturing isomerism, but I think otherwise […] if two dissymmetrical in Turin. In fact, Schif moved from Florence to Turin in inverse bodies ofer in their interaction with inactive bodies, physical and chemical properties that are very similar 1877, because of the disagreements he had with the man- and almost identical, these dissymmetrical inverse bodies agement of the Florentine Royal Institute of Practical will give combinations of diferent absolute properties when and Advanced Studies (Regio Istituto di Studi Pratici e di they merge with asymmetric and optically active bodies. Perfezionamento). He eventually returned to Florence in Te active dissymmetrical bodies that will interact with the 1879 afer the promise of more funds for his laboratory11. nervous system, leading to a sweet taste in one case and Afer his graduation, Piutti remained in Turin as almost tasteless in another, won’t be anything else in my an assistant to Angelo Mosso and Icilio Guareschi, two opinion than like the nervous matter itself, dissymmetri- important Italian chemists of the time. In 1881 he joined cal, just like all the basic substances of life: albumin, fbrin, Ugo Schif in Florence as assistant at the Laboratory gelatin, etc. of General Chemistry. In Florence, Piutti helped with the course of General Chemistry and later in 1885 also of Organic Chemistry. In 1886 he received a degree in THE SCIENTIFIC CONTEXT OF ARNALDO PIUTTI’S DISCOVERY Pharmacy, a professional pratice that was quite usual at that time. In this way he was able to apply for a position At the end of the Nineteenth century stereochem- in Pharmaceutical Chemistry and in the same year he obtained a professorship at the University of Sassari in istry did not exist yet. In 1867 the German chemist 9 Friedrich August Kekulé von Stradonitz (1829-1896) Sardinia . proposed for the frst time the tetrahedral structure of In 1888 he moved from Sassari to Naples, where in 1890 he was appointed as full professor, with the maxi- carbon. Two years later the Italian chemist and politi- 6 cian Emanuele Paternò (1847-1935) applied this hypoth- mum score (Figure 1). He also won the position in Gen- esis to saturated organic compounds6. On the relation- eral Chemistry in Padua, but he preferred to remain in ship between structure and optical properties of a mol- Naples where he could organize the new Institute, following the example of Schif in the reorganization of the ecule, in 1874, Jacobus Henricus van’t Hof (1852-1911), 11 a Dutch chemist awarded with the Nobel Prize in 1901, Chemistry Institute Florence . Te dextrorotatory sweet asparagine of Arnaldo Piutti: the original product is conserved in Florence 127

only when they were declared unft. During that period, not being able to carry out experimental work, he devoted himself to theoretical studies on the spatial representation of chemical elements, arranging them in an alternative way to the Mendeleev table9. He died in Conegli- ano, near Treviso, on October 19, 1928.

ISOLATION OF ASPARAGINE

Asparagine is an α-amino acid (Figure 2) that was identifed for the frst time by the French chemists Louis Nicolas Vauquelin (1763-1829) and Pierre Jean Robiquet (1780-1840) in 1815, in the asparagus sprouts. Short- ly aferwards Joseph Bienaimé Caventou (1795-1877), another French chemist who pioneered the research in amino acids and plant alkaloids, and Heinrich Hlasivetz isolated asparagine from Glycyrrhiza glabra (liquorice) and Robinia pseudoacacia. Later Asparagine was found in many other plant species: in the roots of althea, in potatoes, in hop, in legumes sprouted in darkness and in sweet almonds12. To isolate the asparagine molecule, the juice squeezed from the plant was boiled until it formed an abundant coagulation of albumin. At this stage it was fltered, purifed and crystallized, obtaining crystals that Figure 1. A portrait of Arnaldo Piutti in his laboratory at Naples9. looked like “sugar candy”12,13. Te isolated asparagine was an optically active molecule that caused a lef-hand- Piutti was a member and correspondent of numer- ed rotation of the polarization plane. In 1835 William ous academies and scientific societies, including the Hallowes Miller (1801-1880) a British mineralogist and prestigious Accademia dei Lincei. He held the office physicist, determined the crystallographic constants and of Dean of the Faculty of Science and of the School of measured its refractive indices. Pharmacy in Naples for several years, he became Vice- In 1848 Pasteur identifed a relationship between the Rector and represented the Minister of Education at the crystalline form of the asparagine and the rotation of the International Congresses of Applied Chemistry in 1896, polarization plane, arguing that in principle, a dextrorota- 1890, 1900, 1903 and 1910. In Naples he was for a long tory asparagine with symmetrical crystalline form should exists, albeit the occurrence of the octahedral crystal of time appreciated Director of the Institute of Pharmaceu- 14 tical Chemistry and Toxicology and he personally fol- the asparagine only as “levorotatory hemihedralism” . lowed with great commitment the construction of the Even Karl Friedrich Rammelsburg, (1813-1899), a German new headquarters in San Marcellino9. mineralogist and chemist, in 1855 advanced the possibility of crystallization in the form of a lef or right-handed Piutti’s scientifc contributions are mainly focused 15 on the study of aspartic acid, asparagines and their tetrahedron , but the isolation of the dextrorotatory derivatives and on the optical rotation of organic com- asparagine was carried out only in 1886. pounds. His research ranged also in other fields of chemistry. He studied the toxicity of the combustion products of locomotives and he traced the presence of helium in a mineral of the Vesuvian area, demonstrating the difusion of this gas in solids. He patented the Piran- tina solubile Piutti (soluble Pirantina Piutti), an antipy- retic and analgesic9. Despite having personally contributed to the construction of the new laboratories in San Marcellino, he fnally abandoned those in the place called “Il Salvatore” Figure 2. Te two forms of asparagine. 128 Laura Colli, Antonio Guarna

THE DISCOVERY OF DEXTROROTATORY “SWEET” nal notation). From the former derivative he obtained ASPARAGINE the amidate, and fnally the two rotatory asparagines, identical to the natural products and called by Piutti as Piutti guessed that the failure in identifying the dex- a whole “β-asparagine”, with the formula: CONH2-CH2- trorotatory asparagine was presumably due to its low CHNH2-COOH (Figure 4). Tis is of course a historical abundance in Nature. In the factory of Mr. Galgano Par- nomenclature, which does not comply with the com- enti, near Siena, Piutti prepared a large quantity of the mon rules currently in use. From the second amidate he product from sprouted vetch, a legume15. Trough frac- obtained the inactive “α-asparagine”, an isomeric form tional crystallization he observed the precipitation of of the β-asparagines. Tus, he came to the conclusion two species of crystals: ordinary levorotatory asparagine, that he had obtained three asparagines. He did not know described at that time as “almost tasteless”, and dextroro- whether the inactive form, called “α-asparagine”, could tatory, which turned out to be “sweet”. Tus the dextro- be separated into two optically active asparagines. rotatory asparagine could be detected and separated from He successfully separated the β-asparagines and the levorotatory form on the basis of its taste (Figure 3). found that the compounds, like the natural products, difered in the rotation of the plane of polarization as well as in taste, in solubility and in density. Piutti wrote:

Te synthetic asparagines, thus obtained, difer from each other, just like the natural ones. Besides the diferent hemihedralism and the opposite rotatory power, even the taste, which in the levorotatory is insipid, while the dextrorotatory is “sweet”. Furthermore “…as the rotatory asparagines have the same chemical composition, they are to be considered as physical isomers.

Piutti continued: Figure 3. Te words of Piutti about the frst discover of sweet asparagine: “From the mother liquor from which the asparagine was Tis result […] also shows how the dextrorotatory aspara- crystallized, some crystals formed whose pronounced sweet taste I gine, discovered by me in the vetch and now obtained by 15 immediately caught” . synthesis, is the physical isomer of the ordinary asparagine15.

According to Piutti “the second [amidate ] supplied Te two types of crystals were analyzed and found to the inactive species, chemically isomeric, and until today be chemically identical, with the same refractive indices unknown” (inactive α-asparagine) that he specifed with and same plane of optical axes. In the laboratory of Phys- ics headed by Prof. Roiti at the Royal Institute of Practi- cal and Advanced Studies in Florence, Piutti determined the optical rotation of the two species with a Laurent polarimeter, measuring the following values: [α]D = -5.43 for the ordinary or levorotatory asparagine and [α]D = +5.41 for the dextrorotatory sweet asparagine15 . Te dextrorotatory asparagine, whose existence was theoretically proposed by Pasteur, was fnally discovered.

THE SYNTHESIS OF SWEET ASPARAGINE

Afer separating the dextrorotatory asparagine in Figure 4. Nomenclature given by Piutti to the “asparagines” 1886, Piutti succeeded in setting up the synthesis in he found: on the left the inactive asparagines that he called hislaboratory in 188716. Te synthesis was carried out α-asparagine and the “mixture” of the two optically active asparagi- through the reduction of the oxime of oxaloacetate nes, that he called β-asparagine (α-asparagine and β-asparagine are ether. Piutti separated two monoethyl esters, written two racemic mixtures; Piutti reported this observation, but was not 2 5 2 able to confrm this assumption). On the right the products obtained with the formulae: COOC H -CH-CHNH -COOH and by separation of the mixture of β-asparagine: the asparagine levoro- 2 5 2 2 COOC H -CHNH -CH -COOH (according to the origi- tatory, tasteless, and the dextrorotatory, with a sweet taste. Te dextrorotatory sweet asparagine of Arnaldo Piutti: the original product is conserved in Florence 129 the formula: CONH2-CHNH2-CH2-COOH. He conclud- present in the plant itself. The dextrorotatory sweet ed by saying: asparagine is the D-aminoacid: Piutti demonstrated its presence in nature. I intend to complete the study of this asparagine [inactive During the years in which the theories of van ‘t α-asparagine] and to determine whether or not it is separa- Hof on optically active compounds were not yet easily ble from the two rotatory asparagines of the same composi- accepted, Piutti successfully identifed the two enanti- tion (such as the presence of an asymmetric C would sug- omers of asparagines. He demonstrated the correspond- gest) when I will be able to prepare it in greater quantities. ence between the crystallographic form and the optical properties of a chiral molecule and discovered a correla- In 1890 Piutti modifed the preparation method 17. tion between the diferent optical rotation of the mole- Tis time the asparagines were obtained from the silver cule and the diferent response of the human receptors17. salt of “acido γ-ossimmidosuccinico” in ether via solvent Finally, a few years later, he observed the enantio-specif- evaporation and fltration of the iodide excess. Te result ic use of amino acids by plants 18. is an oil: “nitrilosuccinato dietilico”, a nitrogen derivative Te existence of the two rotatory asparagines in of diethyl succinate. Te oil was treated with bromine in lupins and their selective use in the plant and the obser- acetic acid solution yielding the compound C H N O Br 6 7 2 3 vation that only one of the two enantiomers interacts which is optically inactive. Tis compound was reduced with our receptors, giving the sweet taste, represented in acetic acid solution with sodium amalgam. Te moth- two major discoveries in biological chemistry. These er liquor was lef resting for a long time together with results were of fundamental importance that, perhaps copper acetate; fnally: Piutti (unlike Pasteur) did not fully understand. with a fne sieve the [inactive] α-asparagine, reduced to powder by the loss of its crystallization water, is separated from the crystals of the rotatory β-asparagines, in their THE ASPARAGINE OF PIUTTI IS CONSERVED IN THE turn recognizable and separable by their taste or by their MUSEUM OF NATURAL HISTORY IN FLORENCE diferent hemihedralism17. Te original sample of dextorotary sweet asparagine In this way Piutti understood that the formation obtained by Piutti in 1886 is conserved in the Schif Col- of the rotatory asparagines is independent on the syn- lection of the University of Florence. thesis procedure. Tis observation also emphasizes the Recently, the sample of dextrorotatory sweet aspara- instability of the inactive asparagine for what he ref- gine has been included as a cultural asset to the scien- ereed to as “reasons of physical order”. However, his tifc-technological heritage (“PST”, Patrimonio Scienti- goal changed: he was prompted to isolate the inactive fco e Tecnologico), according to the Italian Ministry for α-asparagine, and to separate it into the two rotatory Cultural Heritage and Activities (Ministero per i Beni e forms. He wrote: le Attività Culturali). Te information concerning this sample are available online through the SigecWeb sys- Te resolution of the inactive α-asparagine into two cor- tem of the Central Cataloging and Documentation Insti- responding rotatory asparagines acquires therefore more tute (ICCD)19. Te scientifc and private correspond- interest now and I am confdent of having the means to ence of Ugo Schif is also conserved at the “Ugo Schif” 17 experiment later . Department of the University of Florence, including the correspondence between Schif and Piutti of those years. A study of these letters will elucidate how and why the PRESENCE IN NATURE OF THE SWEET FORM sample of the asparagine of Piutti remained in Florence in his master’s laboratory. Furthermore, it can shed light In 1915 Piutti made another important discovery: on Piutti’s arguments related to his experimental work the two rotatory β-asparagines “coexist in the products and whether he really understood or not the importance 18 of germinated lupins” . According to Piutti, as the dex- of those discoveries. trorotatory sweet asparagine is mainly used by the plant itself, much more than its optical antipode, only a small quantity can be isolated. Finally the dextrorotatory ACKNOWLEDGMENTS asparagine disappears with the process of germination18. Te presence of dextrorotatory sweet asparagine was The authors gratefully acknowledge the current not due to racemization during the extraction from the Head of the Department of Chemistry “Ugo Schif”, plant, but according to Piutti’s hypothesis, it was already 130 Laura Colli, Antonio Guarna

Prof. Andrea Goti, the past Directors and all the staf 8. J. I. Solov’ev Jurij Ivanovič, L’ evoluzione del pensiero of the Department, for the support to their research on chimico dal ‘600 ai giorni nostri, Edizioni Scientifche the historical roots of the chemistry works carried out e Tecniche Mondadori, Milano 1976. in Florence. We thank Dott.ssa Mariagrazia Costa for 9. L. Pescitelli, Arnaldo Piutti, Napoli, Tip. Cimmaruta, her constant and timely attention. We are also grate- Della R. Università, 1914 ful to the Museum of Natural History for the aid to the 10. T.T. Tidwell, Angewandte Chemie, 2008, Volume Chemical Heritage project, in particular we thank Prof. 47, Issue 6, 1016-1020 Marco Benvenuti, Dott.ssa Angela Di Ciommo and Dott. 11. L. Colli, L. Dei, A. Guarna, M. Costa, in La Palazzina ssa Maria Giulia Maraviglia, Prof. Guido Chelazzi and dei Servi a Firenze. Da residenza vescovile a sede uni- Prof. Giovanni Pratesi. We acknowledge also Fondazione versitari, a cura di Cristina De Benedictis, Roberta Cassa di Risparmio di Firenze for funding the Chemical Roani, Giuseppina Carla Romby, Edifr Edizioni Fire- Heritage project. nze, 2014 12. L. Cantoni Appunti chimici sull’asparagina, Milano, Stab. Della Antica Casa Edit. Dott. Francesco Vallar- REFERENCES di, 1889 13. R. Piria Cimento, 1846, Gennaio-Febbraio 1. A. Guarna, L. Colli, M. Costa, in Rendiconti della 14. L. Pasteur, Annales de Chimie et de Physique, 1848, Accademia Nazionale delle Scienze detta dei XL, serie XXIV, 442. V, vol. XXXIII, parte II, tomo II, 391-400, 2009. 15. A. Piutti, Gazzetta Chimica Italiana, 1886, XVI. 2. J. Gal, Chirality, 2008, 20 (1), 5-19. 16. A. Piutti, Ricerche fatte nel laboratorio di Chimica far- 3. J. Gal, Chirality, 2008, 20 (10), 1072-84. maceutica della R. Università di Sassari, 1887-88 4. J. Gal, Chirality, 2012, 24 (12), 959–976. 17. A. Piutti Rend. della R. Accademia delle Scienze Fisi- 5. L. Pasteur, Académie des sciences, 1886, Comptes ren- che e Matematiche di Napoli, 1890, Fasc. 3° dus hebdomadaires des séances de l’Académie des 18. A. Piutti Estratti dalla R. Accademia delle Scienze sciences / publiés... par MM. les secrétaires perpé- Fisiche e Matematiche di Napoli, 1915, Allegato alla tuels seduta del 4 Dicembre, Fasc. 11° e 12°. 6. G. Montaudo, Bollettino Accademia Gioenia di Scien- 19. http://www.iccd.beniculturali.it/index.php?it/118/ ze Naturali, 2000, Catania, Vol. 42 N. 371, 41-50 sistema-informativo-generale-del-catalogo-sigec. 7. J. H. van’t Hof, La chimie dans l’espace, Rotterdam, Accessed on 13/07/2018. 1875.

Figure 5. Te dextrorotatory sweet asparagine of Piutti: on the lef the compound in the original bottle with the original label handwritten by Ugo Schif; on the right, a particular of the crystals of original asparagines contained in the bottle. Tis item is conserved in the Schif Collection of “Ugo Schif” Chemistry Department at University of Florence (class: Collezione Schif, DCO20). Courtesy of the Museum of Natural History of the University of Florence.

Finito di stampare da Logo s.r.l. – Borgoricco (PD) – Italia September 2018 Vol. 2 – n. 2 Substantia An International Journal of the History of Chemistry 5 7 27 43 73 81 93 19 119 125

of Arnaldo Piutti: the original product is Arnaldo Piutti: of

dextrorotatory sweet asparagine conserved in Florence Historical Perspective Monica Gherardelli Mario Calamia, Application of Radiocommunications Scientifc the First Time: Exact Antonio Guarna Laura Colli, The Anna Maria Papini the endless endorphins: e.g., morphine to endogenous opioid peptides, From quest for the perfect painkiller Maier Norbert M. Ernst Kenndler, A Analysis of Binding Media of Museum Objects: Gas Chromatography and John Elliston 1 – Part Veins of Quartz Hydrationin the Formation of Silica and Its Role Franza Annarita Vincenzo Lusa, Address Application of Chemical Imaging to The Visualizing Chemistry. Scientifc Challenges in Space Research Juan Manuel Garcia-Ruiz The Crystal Monolith 2001: Juergen Heinrich Maar the the Mining School of Ouro Preto, Almost a Discovery – Henri Gorceix, Monazite Sand of Bahia and the Chemistry of Didymium Barry W. Ninham Barry W. Australian National of a Building atThe End the Goodnight and Goodluck: University Helge Kragh Origin and Early History Formula: The Lorenz-Lorentz Table of contents Table No walls. Just bridges No walls.

Vol. 2 – n. 2 2 – n. Vol.

September 2018 September