The first periodic table? The telluric screw of Chancourtois

In 1869, Dmitri Mendeleev (1834-1907) presented his first periodic table of chemical elements and his name is generally associated with the discovery of the periodic table today. However, Mendeleev was not the first to identify a periodicity in the properties of elements when they were ordered according to increasing atomic weights. Based on publication dates, this credit goes to Alexandre-Emile Béguyer de Chancourtois (1820-1886), who in 1862 published his classification of elements based on a helix drawn on a cylinder, known as the 'telluric screw'. Here, we discuss this discovery.

Nineteenth-century approaches to classification

In the wake of Lavoisier’s redefinition of the chemical element in 1789, chemists sought to classify the growing number of chemical elements according to their chemical properties. This qualitative classification somewhat foreshadowed the columns of the current periodic table: for example, the families of alkali metals (lithium, sodium, potassium...), alkaline earth metals (beryllium, magnesium, calcium, strontium, barium...) and halogens (fluorine, chlorine, bromine, iodine) were identified during the first decades of the nineteenth century. 

Meanwhile, chemists were also conducting quantitative research based on the atomic weights that were attributed to elements from the early nineteenth century. Thus, in 1817, the German chemist Döbereiner identified a first "triad": three elements (calcium, strontium, and barium) arranged so that the mass of the middle element is equal to the average of the masses of the other two. This concept was further developed with three other triads published in 1829, aligning elements that are superimposed in the current periodic table. It was the first discovery of quantitative relationships between the masses of elements in the same family, and therefore in some way a first step towards the current periodic table. In 1843, Gmelin combined triads for the first time (he is actually the one who coined this name) in a table containing 55 elements. Although this classification could not yet be considered periodic, it could be considered a precursor in the sense that it already correctly grouped the elements in columns 1, 2, 15, 16, and 17 of the current periodic table, at least for its first 3 rows. 

Figure 1. Representation of Döbereiner's triads in relief on the first 6 lines of the current periodic table. Author’s adapation of an image in the public domain.

Quantitative research continued in the 1850s. During this period, more numerical relationships between elements were identified both within chemical families and between neighboring families. To create a table, it would have been enough to arrange the elements by increasing atomic masses across the rows and columns. Still, one had to think of it!

Moreover, many of the atomic weights that chemists adopted during this time were not correct from today’s point of view, which also made it more difficult to identify periodicity. However, in the 1860s, very rapid progress was made in chemical classification, with multiple chemists independently identifying periodicity. This may have been due to the work of the 1860 International Congress of Karlsruhe, which helped to organize the determinations of atomic masses and fostered a consensus in this field among researchers from different countries. On the other hand, the rapid discovery of new elements, notably through spectroscopy, facilitated classification research. It was in this context that Chancourtois' innovative approach led to the first true demonstration of periodicity.

A geologist among the chemists

Figure 2. Alexandre-Emile Béguyer de Chancourtois

Chancourtois, a polytechnician (class of 1838) and mining engineer, was a professor of geology at the School of Mines (École des Mines) in Paris and the president of the Commission for the Detailed Geological Map of France. From 1852 onwards, he taught geology, first as an assistant to Elie de Beaumont, then as the chairholder in geology from  1875. While working on this course, Chancourtois sought a way to logically present all the notions relative to minerals and their constituents. He ultimately had the idea to arrange the elements by increasing atomic mass: he aligned them along segments inclined at 45° to create a table that can be described as oblique. Thus, he correctly placed today’s columns 1, 2, 13, 14, 15, 16, and 17: chlorine was positioned directly below fluorine, calcium below magnesium, and silicon below carbon, for example. Then, he formed a cylinder by wrapping this table around a cylinder so that the segments formed a continuous helical line; the families were then located on the generators of the cylinder. However, from today’s view, the table became incorrect from calcium onwards: there were increasing  difficulties in identifying and classifying transition metals without taking into account columns 3 to 12, which should have been ten generators more on the telluric screw. 

Chancourtois presented his findings at the French Academy of Sciences and later published a pamphlet detailing his classification. Published in 1862, this is the first system to highlight the periodic recurrence of the properties of chemical elements on the basis of their atomic weight. Chancourtois named his classification the telluric screw, "based on its mode of realization and origin”, as he explained at the French Academy of Sciences on April 7, 1862. A little later, on May 5 of the same year, while specifying that the name was suggested to him especially by the central place of the element tellurium on the screw, he wrote that "the epithet telluric (...) very happily recalls the geognostic origin, since tellus means earth in the most positive, most familiar sense, in the sense of nourishing earth." ([4] p. 11). This statement clearly shows that geology was his starting point. He saw the helical shape as ideal for representing periodicity: thus, he used comparable methods for establishing "telluric" classifications of earth movements, and the different geological formations. 

Figure 3. The telluric screw, collection of École des Mines de Paris  

Chancourtois had an almost mathematical vision of matter, according to which "the properties of bodies are the properties of numbers". This vision could explain the priority he gave to numbers rather than chemical properties. Thus, Chancourtois did not correctly place iodine below chlorine and fluorine as his successors did: the issue is that, curiously, iodine (element 53) is lighter than tellurium (element 52) and an inversion in the order of atomic weight is required in order for iodine to fit with the halogens. Furthermore, this mathematical vision led him to attempt to predict chemical properties based on a factorization of atomic masses, which were assumed to be whole numbers. In this spirit, he wanted to bring the notion of element closer to that of prime number. 

Although he was wrong about various things from today’s point of view, the importance of the atomic number today somewhat confirms Chancourtois's intuition that there is a close relationship between numbers and the nature of substances. Similarly, his idea that these numbers could be used to predict and explain the spectral lines of elements could be seen as a kind of precursor to Moseley's law of 1913, which connects the frequencies (ν) of the lines to the atomic number (Z) of the element. At the time however, Chancourtois’s telluric screw did not receive much attention, especially since Chancourtois was not part of the circle of chemists.

Figure 5. Simplified representation of the telluric screw, established by the author

During the seven years following the publication of the telluric screw by the geologist Chancourtois, four chemists developed increasingly accurate periodic tables. For the most part, these classifications start to contain errors from our contemporary point of view from the third row onwards, as the transition metals remained very difficult to classify correctly. In 1865, Odling was the first to separate these transition metals from the other columns. However, it took a few more years to arrive at systems that took these elements into account: Lothar Meyer's system (1870), which previewed only nine columns of transition metals. Finally, Mendeleev's system (1869) included correctly for the first time ten columns of transition metals. These two tables were not the first to be published, but they were by far the most accurate. 

Figure 6. Representation of Mendeleev's periodic table, version of 1872 [5], with its main predictions. When moving the block of surrounded elements, we almost exactly find today’s periodic table, although in a horizontal position. Image by the author.

Then, Mendeleev's table was the first to be composed of 17 columns – or actually 17 rows, given the horizontal layout of his first table. When Chancourtois’s telluric screw had included 58 elements, Mendeleev increased this number to 63. To classify them, Mendeleev arranged the elements by increasing atomic weight, but he also paid attention to the chemical similarities between them. Thus, in addition to swapping the positions of iodine and tellurium to respect their belonging to chemical families (which Chancourtois had not done), he adjusted the weights of beryllium and uranium based on their positions in the table. Moreover, Mendeleev did not only classify already known elements. He quickly predicted extremely precisely the properties of three yet unknown elements that he called eka-aluminium, eka-boron, and eka-silicon, letting three empty places for them in his first table of 1869.

Three discoveries eventually confirmed these predictions: eka-aluminium was discovered under the name gallium in 1875 by Lecoq de Boisbaudran, eka-boron under the name scandium by Nilson in 1879, and eka-silicon under the name germanium by Winkler in 1886. The fact that these elements were actually found significantly contributed to the fame of Mendeleev and his table. However, it was virtually impossible for Mendeleev and the other precursors to foresee the 18th column. It was Ramsey and Rayleigh who demonstrated in 1900 that the noble gases constituted this 18th column that completed the table [1], much like the keystone that was still missing to ensure the stability of a building [6].

Epilogue

Nevertheless, one can be surprised by such a contrast between Mendeleev's fame and the oblivion into which Chancourtois has fallen, considering the priority of the latter in identifying periodicity. However, Mendeleev’s contribution was far more significant for the shape of the periodic table as it is known today. Mendeleev expanded it to 17 columns, and was able to predict previously unknown elements. In conclusion, if Mendeleev is the main builder of the periodic table of chemical elements, which Ramsay later completed, it is indeed Chancourtois who laid its first foundations.

This article is based on a publication in la Jaune et la Rouge (alumni journal of the École polytechnique), n° 749, 2019, p. 35. 

References

[1] Scerri, E., The Periodic Table: Its Story and Its Significance, Oxford University Press, 2007. 

[2] Van Spronsen, J. W. L’histoire de la découverte du système périodique des éléments chimiques et l’apport de Béguyer de Chancourtois, Conférence donnée au Palais de la découverte, Alençon : Bernard Grisard, 1965. 

[3] Fuchs, E., « Notice nécrologique sur M. A.-E. Béguyer de Chancourtois, inspecteur général des mines », Annales des Mines 1887, 11, p.505.

[4] Chancourtois, A.-E. Béguyer de, Vis tellurique, classement naturel des corps simples ou radicaux obtenu au moyen d’un système de classification hélicoïdal et numérique, Paris : Mallet-Bachelier, 1863. 

[5] Mendeleïev, D. I., « Die periodische Gesetzmässigkeit der chemischen Elemente », Annalen der Chemie und Pharmacie, 1872, vol. suppl. 8(2), p. 149. 

[6] Avenas, P., How the element names reveal their history, Comptes Rendus. Chimie, Vol. 23, 2020, n°. 3 p. 221-230, translation posted online: 2023-11-06.