The beginnings of modern Mineralogy: from the end of the XVIII th century to the middle of the XX th century.

It is most probably between the end of XVIIIth century and the beginning of XIXth century that occur the true beginnings of modern mineralogy. The development of the geometrical crystallography which, started from the morphological study of natural crystals, rapidly lead to the analysis of symmetry laws and the identification of the crystallographic systems. J.B. Romé de l’Isle wrote the law of the constancy of dihedral angles, while R.J. Haüy identified the major symmetry laws of the crystals shapes and introduced the concept of " integrating molecule ". G Delafosse introduced the concept of discontinuity into the crystal. The elements of symmetry identified by Haüy (centre, mirror plane, rotation axis of order 2, 3, 4 or 6) made only possible to define the 32 classes of symmetry to which are referred all natural crystals, which one were established through the middle of XIXth century by A.Bravais . Bravais also recognised a certain number of lattices motifs, related to the internal structure of the crystals and used additional operators of symmetry, mirrors with translation and helix rotation axes. The, purely mathematical, analysis of the sets authorised by these symmetry operators, which is one of the bases of the groups theory, remains today one of the most beautiful examples of an application of the most abstractive mathematics to a science generally considered as naturalist. In 1879, Léonard Sohncke identified 66 groups of symmetry and opened a way for Schoenflies and Fedorov’s work who proved the existence of 230 symmetry groups which, still today, form the basis for the study of the crystals atomic structures.

It seems remarkable to note that the atomic structure, which had been intuitively taken as an hypothesis, but not demonstrated, from the geometrical analysis of the external shape of the natural crystals, will be definitively verified by the development of X-rays crystallography at the beginning of XXth century. After G Friedel who defined, in 1904, the concept of crystal motif, Max von Laue (1879-1960) performed in 1912 the first X-rays diffraction experiment on a sphalerite crystal (cubic ZnS). After this historical demonstration of the periodic structure of the crystals, X-rays crystallography will receive rapidly a considerable development with names such as Walter Friedrich and Paul Knipping, but especially the Bragg family William Henry, the father (1862-1942) and William Laurence, his son (1890-1971). They received the Nobel Prize in 1915 together to have shown that the study of the X ray diffraction diagrams " allowed to determine the structure of minerals on the scale of the atom ". In 1917, P. Debye (1884-1966) developed one diffraction technique using as a specimen a finely grounded mineral (powder diagrams), which under the name of " Debye-Scherrer method " will become the most usually used method for identification of mineral species. The first powder diagrams are carried out in 1917 and their authors will receive the Nobel Prize in 1936. At the same time, the data on the structure of various minerals are refined. As one of the X-rays crystallography pioneer, Bragg did some work on the structure of halite (Na Cl), which will be solved in 1924 confirming the structure model proposed by twenty years earlier by William Barlow. It is never so good to be in advance over his time, and Barlow’s ideas did not received many attention from scientists of his time. After the historical discovery of Bragg, the structures of the main mineral groups will be quickly identified: diopside, trémolite and vesuvianite by B.E. Warren in 1929, beryl, muscovite and biotite by J. West, the so significant group of feldspars by W.H. Taylor, etc. Circa 1935 the structures of the commonest minerals are known.

X-rays crystallography will be also a powerful mean of investigation for minerals with variable composition (solution solids), solving in a final way the controversy which had opposed, a long time ago, Haüy and Mitscherlisch concerning the principle of isomorphism. Radiocrystallography will then become an important method for the determination of the chemical properties and will be used as a complement to the direct methods of analysis. In 1922, Mauguin published a fundamental work on the structure of simple crystals with variable chemical composition: chlorides and sulphides. He also studied the structures of micas and chlorites. He demonstrated that the framework of these silicates is formed by a fixed number of atoms of oxygene:12 per unit cell for the micas and 18 for the chlorites. On the other hand, the number of cations in clays is variable. In 1929, L. Pauling established the complete structure of those complex minerals, by giving the unit cell sizes of the crystal and the thickness of the layers, the stacking of which possess a fixed repetition length (10 Angstroms for micas, 14 Angstrom for chlorites) and constitutes the mineral on a macroscopic scale. Those works, constitute the base for all the following researches on the structure of clay minerals. They opened a way for a generation of researchers, in France we have to mention among many others : J Wyart, S. Goldztaub, A.J. Rose, A. Rimsky, which make possible that today, the structure of most of all the minerals are known in detail providing a coherent physical base for the interpretation of their properties.

Today, a great importance is given to the study of the crystal lattices defects. The knowledge acquired in mineralogy finds immediate applications in other disciplines, in particular metallurgy and materials science. To give a recent example, it is interesting to note that the structures of the first " high temperature " supraconductor materials were determined by mineralogists of the Geophysical Laboratory at Carnegie Institution (Washington D.C., U.S.A.). Doing that work they participate in a decisive way in one of the greatest adventures of modern physics.

A mineral is defined as a natural body with an ordered structure - in rare exceptions some minerals possess a vitreous structure, such as the natural glass known as lechatelierite- and with precisely codified chemical composition:

This official " authorisation " corresponds to the main part of the discussions within the international commissions of nomenclature. As an example of the discussed question :

- must one consider or not "oligoclase" an albitic plagioclase feldspar with approximately anorthite 20%, like a separate mineral species or a variety of another mineral which is plagioclase ?

The discussions between members of the commission are often tough. By times, non entirely scientists considerations appear. In theory, many rules to propose a mineral name exist indeed :

related to the location of the type deposit (e.g. andalusite from Andalousia),

related to its physical properties (kyanite = from the ancient Greek for " blue color ")

related to its chemical formula (fluorite = because it contains fluor),

etc.

Nowadays there is a general tendency to dedicate a new mineral to a known mineralogist, often a colleague of whom proposed the new mineral. But in rare cases the acceptance of a new name or, much more difficult, the removal of an insufficiently characterised species, can also lead to some political problems.

These few remarks point out the fact that chemical composition is as significant as structure for the characterisation of a given species. The first classification which one can regard as modern, by A.G. Werner from Freiberg, in Saxony, was only based on the physical properties of minerals. But since 1758, Cronstedt introduced chemical data, which quickly were considered as important and later as more important than physical properties. During this period a major interaction between mineralogists and chemists happened. As an example, the discovery by J Dalton (1766-1844) of the stoechiometry laws, provided a coherent base to the concept of the atomic structure of minerals. Many scientists of this time, such as Ebelmen or Vauquelin, were, at the same time, famous chemists and the principal craftsmen of the " chemical mineralogy ". The Swedish school was particularly remarkable, with T. Bergman (1735-1884), and especially J.J. Berzelius (1779-1848), who developed a classification of the minerals based on the electronegativity of the elements. At the same time were introduced some class of mineral such as oxides, halides, phosphates, sulphates and silicates which form still today the basis of the modern classifications. One have to add to the pioneers list, the name of Lavoisier, which was himself much interested in mineralogy - with an important personal collection of " mineral substances ". Unforunately, this great chemist was so buzy with his chemical researchs that he could not develop the mineralogy of Haüy toward the chemical way into which he should have probably lead it.

For R.J. Haüy, indeed, chemistry plays only one additional role in the definition of a mineral. If it can reveal the nature and the quantity of the components, it cannot reveal its structure (assembly) which is defined only by geometrical criteria (crystallography). For Mitscherlisch, on the contrary, two minerals can be different although they have a same structure, if significant differences exist on the level of the chemistry. This is the principle of isomorphism and the source of a violent controversy which will tarnish the last years of the long life of R.J. Haüy (which will even bring his adversaries into the justice court!). This fundamental point will be definitively solved nearly one century later, when V.I Vernadsky, V.M. Goldschmidt and some others generalise the concept of solid solution, establishing in an irrefutable way the accuracy of Mitscherlich ideas.

Beyond these controversies, during the end of XVIIIth century and the beginning of XIXth will happen an exceptional scientific knowledge growth. The majority of " significant " minerals and 25 new chemical elements will be discovered in a few tens of years (of 1790 to 1830), establishing in a final way the chemical base of descriptive mineralogy and consolidating Berzelius's ideas. A decisive event will occur in 1837 with the publication of the first edition of the " System of Mineralogy " written by James Dwight Dana. In the 4th edition, which appears in 1854, Dana adopts Berzelius’s classification, which one is surprisingly very near of the classification presented in today manuals of mineralogy. The " Dana " will remain for decades the " bible " of mineralogists, supplanting in a final way the former treatises written by R.J. Haüy, Dufrenoy and F Beudant for France, J.F.L. Hausmann (1813) for Germany and R. Jameson (1816) for United Kingdom.

However, if gradually , chemical data take an increasingly large importance, it should be well understood that, during the whole 19th century, the analysis of most of the minerals is still a difficult task due to the technologies of that epoch. With rare exceptions, the only way to obtain good quantitative analysis was after their entire dissolution in aqueous solutions, a particularly difficult operation for many minerals like the silicates. Initially, the presence of certain elements - especially metals -, could be analysed by the "dry way" methods like the pyrognostic tests with the blowpipe. Those studies were lead to a very high level of sophistication by specialists such as A. Braly or R. Berthier. Definitively forgotten today, these techniques still constituted a significant part in learning Mineralogy in universities shortly after the Second World war.

Quantitative analyses used traditional methods of analytical chemistry:

We have to mention the use of some additional clever techniques, for example : microchemistry, which leads to the precipitation under the microscope of microcrystals considered as characteristic index of the presence of a certain element, e.g. work of Berhens and Boricky in 1877,de L Bourgeois in 1893, unfortunately, those methods have to be considered as giving qualitative or semi-quantitative results and for that reason their development remained relatively confidential.

In every country, only some rare research centres were sufficiently significant to have all the hardware and especially the analysts sufficiently qualified: the Museum of Natural History in Paris (works of A. Lacroix and J Orcel), laboratories of the Carnegie Institution in USA, universities of Berlin or Heidelberg in Germany, research centres of Vienna (Austria) or Zurich (Swiss), etc. In spite of these difficulties, an enormous analysis work of thousands of minerals and rocks were performed out between 1790 and the years following the First World War, allowing the advent of a new major discipline of Earth Sciences : Geochemistry, which will appear circa 1920 starting from at least three independent, but complementary, currents of thoughts: the group of F.N. Clarke in U.S.A., which will develop chemical petrography (in particular for petrographic classifications and estimation of the average compositions of the various major structures of our planet), V.M. Goldschmidt (1888-1947) in Norway, then in Germany (unfortunately taken in the storm of WW2) (application of thermodynamics law to chemical balance between minerals) and V.I. Vernadsky, then A. Fersman (1883-1945) with Paris (Museum of Natural History), then Russia (Moscow) (concept of geochemical cycle, interaction between the different envelopes of our planet, from atmosphere to inner domains). The majority of current specialist considers V.M. Goldschmidt and V.I. Vernadsky as the true initiators of modern Geochemistry, which took a rapid development since the possibility to measure and interpret " geochemical tracers " ( trace elements and isotopic geochemistry).

Due to the difficulties encountered to perform precise chemical analysis on minerals, scientists have very early the idea to mix them with the study of the other properties of minerals, in particular their physical properties. Basic physical properties: color, luster, density, hardness, aspect (cleavages, break) remain significant means of determination, especially as field methods of mineral recognition. The principles of their study did not vary a lot since the beginning of the 19th century: circa 1820, the Austrian mineralogist F. Mohs, who gave the first rigorous definition of the 7 crystallographic systems, also published the famous mineral hardness scale divided into 10 degrees universally used still today (1: talc, 2: gypsum, 3: fluorite, 4: calcite, 5: apatite, 6: orthoclase, 7: topaz, 8: quartz, 9: corundum, 10: diamond). Two other physical properties have a particular importance for minerals: the interaction of light within the crystal (optical properties) and their thermal properties, very used for the analysis of certain minerals having very small dimensions (clay minerals).

The discovery of the double refraction in the Iceland spar by Erasme Bartholin in 1672 and its first scientific explanation by C Huygens in 1674 marked the beginning of the study of the optical properties of crystals. Research will slowly progressed during all 18th century, more devoted to geometrical crystallography, but from 1800 it will rapidly accelerate, since new optical scientific instruments (microscopes) appeared to be well adapted to minerals study. Circa 1810, Arago studied polarisation and optical activity in quartz, while Biot defined the first principles of what will become the crystal optics, by differentiating in particular uniaxial and biaxal crystals. It is well known that certain minerals interfere with the polarisation plane of the light and it is the reason why numerous works on microscopes optics will be performed. The first researches are devoted to the crystals themselves and interestingly the first objects which struck observers when they started to look at crystals under the microscope will be the crystal imperfections like fluid inclusions. At the beginning of XIXth century, Davy, then Brewster discovered inside certain crystals a mysterious " expansive liquid ", of which they analysed with precision the physical properties (refraction index, etc.) and the behaviour with variable temperature. But its composition will be known only more than fifty years later (1871), by the spectroscopic analyses of H. Vogelsang and F. Geisler in Delft: the misterious liquid was nearly pure CO2, one of the main components of fluid inclusions.

In 1818, Mallus discovered chromatic polarisation and defines the polarised ray of light. Circa 1830, Nicol improved the Mallus’s prisms and built the famous " nicol ", allowing the construction of modern polarising microscopes adapted to the observation of thin sections of minerals and rocks. With some mocking remarks of his French colleague, Monsieur de Saussure, which was amused by the fact that " one could even think to apprehend a mountain under a microscope ", the English H. C Sorby generalised the use of the polarizing microscope in mineralogy and in petrography. This constituted a true technological revolution which have nearly no more recent equivalent excepted the remarkable invention of the electron microprobe by Castaing one century and half later. Thanks to the remarkable technical progress made during a few years by some major microscope manufacturers, like Nachet in France, Leitz and Zeiss in Germany, Reichert in Austria, the determination of the optical characters of minerals became one of greatest scientific adventure of the 19th century

Mineralogists seek the relations between optical properties, structure and composition within the main groups of rocks minerals: initially feldspars, micas, pyroxenes and amphiboles. The polarizing microscope became, more than just another means of determination, a genuine tool for analysis. Crystal optics is a complex science, in which illustrate themselves famous physicists, such as Fresnel, who defined in 1820 the indexes ellipsoide. Then two major schools will dominate the research :

* France, which remained almost exclusively mineralogical thanks to the researchers of Museum of Natural History and of School of Mines and

* Germany including Austria, which turned itself towards the systematic description of the assemblies of minerals within a rock, a fascinating goal thus a way more clearly petrographic.

In Paris, A. Descloizeaux studied within thick sections, the optical properties of 468 mineral species, and A. Michel-Michel-Levy, circa 1880, defined the theoretical bases allowing the interpretation of birefringence measurements. The so significant problem of the determination of the group of the feldspars, which will allow the development of most of the modern petrographic classifications based on the composition of " white " minerals (rich in alkaline - quartz, feldspars and feldspathoides), will be worked by De Cloizeaux in 1875, continued by P. Fouqué and A. Michel-Michel-Levy who finally solved it in his masterpiece work published in 1894. " the study of the determination of feldspars in thin sections from the point of view of the classification of the rocks ", printed in Paris by Librairie Polytechnique Baudry and which will be then resumed, in a more or less direct way, by all the subsequent works explaining the handling of the petrographic microscope to petrologists.

At nearly the same time (1893), the Russian V.S. Fedorov introduced the first universal stage, allowing a three-dimension orientation of the thin sections of rocks under the microscope. These complex apparatus authorise a precise location of the optical elements (axes in particular) versus the crystal. The universal stage will be at the origin of a series of significant work (Duparc and Rheinhart in Switzerland, Emmons in the USA), which remain remarkable from the theoretical point of view, but which lost many of their practical interest in analytical mineralogy since the invention of the electron microprobe. On the other hand this technique remains very used in a simplified form (the needle stage) for the study of minerals in grains, as for microstuctural measurements.

In Austria-Germany, research was much more oriented towards petrography. German petrographic classifications are especially based on the " colour index " which is the relative percentage between white minerals (containing K, Na, Ca) and coloured minerals (rich in Fe-Mg). The colour index is immediately perceptible under the microscope, without being necessary to know more accurately the composition of the various mineral groups. The contribution of the German school to the avancement of microscopic mineralogy is however far from being negligible. In 1864, P. Tschermak defined in a rigorous way the concept of solution solid and, in 1870, he described the microscopic properties of the principal groups of coloured minerals: micas (biotite), pyroxenes (augite) and amphibole (hornblende). Circa 1890, the Viennese mineralogist F. Becke indicated that, when one defocused the thin section under the microscope, the direction of displacement of a luminous fringe in contact with two contiguous minerals is done according to the relative value of their refractive indexes. This famous method called " Becke 's fringe" remains still today an universal mean of determination, not only on minerals out of thin section, but also on insulated minerals (mineral in grains).

Near 1900, theoretical knowledge is sufficient to approach the realisation of the great systematic treatises, whose realisation will continue until WW2. Those treatises related to systematic mineralogy as well as petrography, with still certain differences between France and Germany. In France, one privileges especially systematic mineralogy, thanks in particular to the eminent A. Lacroix, Professor at Museum of Natural History, who will publish during more than 20 years from 1893 his monumental masterpiece about the " Mineralogy of France and Madagascar ". In Germany, the treatises of H. Rosenbusch (Mikroskopische Physiographie of Mineralien und Massige Gesteinen) whose first edition goes back to 1873, will be constantly reedited and supplemented until 1924 by E. Wulfing.

Those books remain essential references, and especially they are the direct source of all the work used still today for the teaching of mineralogy and petrography microscopic (e.g. / Mr. Roubault in France, A. Kerr in USA, etc)

For a long time, only transparent minerals could be studied with the transmission microscope, in which the ray of light pass through the observed material. In 1927, J. Orcel extended this technique to opaque minerals. He developed a polarising reflection microscope making possible to make qualitative observation on the opaque minerals, their colour, hardness, cleavage, polarisation colour, etc. The reflective capacity could also be measured in a quantitative way, by a relatively complex equipment.

One thus could prepare determinative tables (F Uytenbogaardt and E.A.J. Burke) very used in metallogeny. Current research is oriented towards microscopes with an infra-red lighting source, which thanks to an image converter placed in the eyepiece make possible to in transmitted light study the minerals (sulphides, native elements) opaque in the visible spectrum.

It should be noted that the theoretical study of transparent minerals in polarised light lead to a quasi-universal use of thin sections with a standard thickness of 30 micrometers. The preparation of those thin sections destroy most of the fluid inclusions contained in certain minerals (quartz particularly) because usually such inclusions possess comparable or even larger dimensions. An important field of study of those fluid inclusions, currently under development, is done on special preparations (polished sections 3 to 4 times thicker than the normal sections thin), thanks also to a microscope equipped with an heating and cooling stage, allowing the observation of the preparations in a wide range of temperature (microthermometry, between -200 and +600, even +1500°C).

An unexpected discovery due to the polarizing microscope is the discovery of liquid crystals:

In 1899, Lehmann discovered the particular behavior under the petrographic microscope of cholesterol benzoate, which between 145°C and 178°C presents both an optical anisotropy and a certain fluidity. He gave the name " liquid crystal " to this new phase, which is found under other conditions, in particular on the surface of drops of very viscous liquids (F Grandjean). F. Wallerant was also interested in these substances and improved their methods of study. But it is CH. Mauguin who performed the discovery of a very significant phenomenon: by heating liquid crystals in a magnetic field, the molecules behave as an uniaxial crystal whose optical axis is parallel with the lines of the magnetic field. In 1922, G Friedel replaced the expression " liquid crystal " by the expression " mesomorphic phase ". Ray (1936), then Weygand (1948) demonstrated that the liquid crystal state appears only if the molecule plane is very elongated. One knows today many substances having a liquid crystal state. For a long time, they were considered as laboratory curiosities, but with the computers age them became industrial products. Liquid crystals are indeed likely to be directed under the influence of a low intensity electromagnetic field, therefore they are used to visualise signs or images on a screen. They are now used on very a large scale in many applications related to electronics.

If the optical properties of minerals are especially used for their determination, in particular for the petrographic applications, certain properties can also have other applications. It is the case in particular optical activity, which made it possible to better know the structure of certain minerals. In 1837, Delafosse highlights the fact that the optical activity is related to the lack of symmetry of the medium. His student Louis Pasteur, in 1860, distinguished two forms of merihedries of which one only is compatible with the existence of the optical activity. This property will be notably used for the analysis of sugars. Des Cloizeaux discovered the optical activity of cinnabar, F Wallerant explained it in quartz, and A. Longchambon, in 1922, developed a method of measurement making possible to establish a relation between optical activity and crystal symmetry in the biaxal crystals. Other physical properties were the object of many studies and applications: piezoelectricity or pyroelectricity, properties that certain crystals posses to make appear electric charges under the influence of pressure changes (piezoelectricity) or temperature changes (pyroelectricity). These properties are directly related to the anisotropy of the crystal system (in particular merihedry of the rhomboedral or hexagonal crystals like quartz or tourmaline. The existence of pyroelectricity have been recognised since the end of the 18th century (experiment of R.J. Haüy on the tourmaline), but it is the piezoelectricity, studied in 1880 by Jacques and Pierre Curie, which presents the most l practical interest (continuous vibration with frequency under the influence of a field electric, phenomenon used in the radio transmitters or in " quartz watches ").

The magnetic properties of certain minerals also present one very great importance. Certain minerals, like magnetite, are very strongly magnetic. Thus, they constitute remarkable natural magnets, but the mineral species concerned with this kind of magnetism (ferromagnetism) are relatively rare (essentially, double oxides with spinel structure). On the other hand almost all the minerals, especially when they contain chemical elements with variable valence, are more or less magnetic, but at much smaller level than ferromagnetic minerals. Magnetism present in a given mineral phase disappears above a certain temperature (Curie’s temperature), but is likely able to reappear during cooling, fossilising the direction of the terrestrial magnetic field at this moment. It is the principle of paleomagnetic studies, which played an essential role for the comprehension of the modes of ocean floor accretion starting from the semi-oceanic ridges. The concept of plate tectonics, which literally revolutionised the Earth Sciences during the fifty last years, is a direct consequence of these paleomagnetic studies, which still comprise many unknown factors (for example, on the precise origins of the inversions of the terrestrial magnetic field, repeated many times during Quaternary). To those electric and magnetic properties, one must add also mechanical properties, which render possible to understand the the deformations of minerals in deep domains which are not directly accessible for us. But in fact especially the thermal properties made - and continue to be the subject of most of the applied studies. Thermal conductivity (or heat-storage capacity, which is directly dependent from this one) of minerals is an essential parameter, used as well for the interpretation of the geophysical data (interpretation of the heat flows measured on Earth’s surface, estimation of the geothermal degree,etc.) as for the thermodynamic calculation of balances between mineral phases. A very significant application is the thermal measurements related to the analysis of clays, too small minerals and complexes to be analysed easily by all other technique. It is since 1930 that those techniques were generalised. If an hydrated mineral is heated, it is noted that it passes by various phases of dehydration, disorganisation and recrystallisation, which result in losses of weight (thermoponderal analysis), of size variations (dilatometric analysis) or calorific, dependent on the endo- or exothermic reactions (thermodifferential analysis). In connection with X-rays diffraction analysis, those techniques are essential for the study of clay minerals, a very important domain of the sedimentary geology.