On the Discovery of the Periodic System of Elements¹

A. V. Rakovskii, Moscow.

The modern doctrine of the structure of atoms, which has so closely linked such distant fields of knowledge as spectroscopy and the periodic system of the chemical elements, upon careful consideration of the question leads us to admiration for the genius of D. I. Mendeleev, who created so perfect a system on the basis of very scanty data. The perfection of this system became clear only with the emergence of the doctrine of atomic structure and was demonstrated by physicists; from the chemical point of view it was unprovable. In the chemical literature one sometimes encounters the opinion that the creation of the periodic system of elements was the fruit of strictly inductive thinking and resulted from an attentive study of the facts. However, a brief historical account of D. I. Mendeleev’s first steps in this field will show that this opinion is incorrect.

In order properly to assess D. I. Mendeleev’s creative work, it is necessary for a brief moment to transport ourselves mentally to the era when Mendeleev began his search for the periodic law. This was at the end of the 1860s of the last century. The passionate disputes between the adherents of the old chemical theories—the theories of types and radicals—and the new theory of valence had only just died down. Fierce disputes were being waged between the two camps of the new theory, the supporters of constant and variable valence of the elements. Only a decade had passed since the appearance of Cannizzaro’s historic work; the major idea of the necessity of a clear distinction between the concepts of equivalent, atomic, and molecular weights had not yet entered into general use. Let us recall that the famous French chemist Berthelot only in the 1880s of the last century passed from the language of equivalents to the language of atomic weights. The very concept of atomic weight did not seem to all chemists as important and cardinal as it does to us now.

¹ From the address “On the Works of D. I. Mendeleev,” delivered at the Central House of Scientists in Moscow on the occasion of the 20th anniversary of the death of D. I. Mendeleev.

By the end of the 1860s chemistry had identified about 64 elements. Taking atomic weight as the fundamental property of the elements, Mendeleev arranged them in increasing order of atomic weights and observed not only individual groups of elements with recurring properties, as had already been seen by Döbereiner (triads) and Newlands (octaves), but also saw that all elements are subject to a general law, the law of periodicity, which he expressed in the following words: “elements arranged according to the magnitude of their atomic weight exhibit an evident periodicity of properties.” But Mendeleev did not confine himself to the discovery of the periodic law; in the same year, 1869, he made the first attempt to give a system of the elements on the basis of the law he had discovered (Table 1).

TABLE 1.

1869

In this table Mendeleev did indeed stand on the ground of facts, but the table itself was not of great interest.

In the following year, 1870, an article by L. Meyer appeared, in which he cites Mendeleev and gives a table essentially identical with Mendeleev’s table; what was new in this article was the now well-known curve of atomic volumes. Looking at this diagram, we are indeed convinced of the existence of the periodic law, but one so complex that the words of L. Meyer become entirely understandable: “It would be premature to alter the atomic weights accepted up to now on the basis of so insecure a starting point.” At the same time Mendeleev, developing his ideas further, paid special attention to the periodicity not of the physical but of the chemical properties of the elements. Considering, on the one hand, atomic weights, and on the other—the highest valence of an element in salt-forming oxides, Mendeleev noticed the possibility of a considerably stricter and more regular periodicity, though at the cost of a major step—the alteration of the atomic weights of a number of elements, i.e. a step,

venture, which L. Meyer did not consider possible. Already in 1871 Mendeleev gave a new table, essentially coinciding with the modern generally accepted table; here the elements are already divided into 8 groups and into periods. It is necessary, however, to draw attention to the revolution that Mendeleev was compelled to make in the generally accepted atomic weights in order to construct his table of elements.

In Table 2 I have placed only those elements whose position either gave rise to no doubts at all or gave rise only to some perplexity (the underlined elements).

TABLE 2.

Thus, out of 64 elements, the placement of only 35 elements gave rise to no doubts; 8 elements, underlined in the table, already caused perplexity for Mendeleev himself as early as 1869. It was somewhat unclear that Zn, Cd, and Hg fell into one group with Mg, Ca, Sr, and Ba, and likewise that Cr and Mn fell, the former into the group of S, Se, and Te, the latter into the group of the halogens. Aluminum also caused some difficulties; the point is that the determination of the vapor density of aluminum chloride led to the formula Al₂Cl₆; consequently, aluminum had to be regarded as a tetratomic element. The placement of lead in group IV also seemed abnormal, since at that time compounds of tetratomic lead were not yet known. But these difficulties were small in comparison with those encountered in the attempt to place the remaining elements in the table: here a radical measure was required—either a sharp change in the generally accepted atomic weights or the placement of elements in positions not corresponding to their atomic weights.

In the following table are shown the elements whose atomic weights Mendeleev had to change; in the second column are shown

atomic weights accepted before Mendeleev, in the third—those changed by Mendeleev.

TABLE 3.

For the following elements Mendeleev changed the order in which they follow one another:

TABLE 4.

For Ni and Co, as well as Te and J, Mendeleev compared the atomic weights (59 for the first pair, 127 for the second); for titanium he considered a lower atomic weight (\(<50\)) probable.

Thus, out of 63 known elements, 8 were placed in the table with a violation of chemical kinship, and for 20 elements it was necessary either to change their atomic weight or to put them in a position abnormal for them.

Let us suppose that we are living in 1871 with our psychology, with our respect for facts; let us imagine that a chemist comes to us and declares that he has discovered a periodic law and that on it he is building a natural system of the elements, but... for this it is necessary to do greater or lesser violence to 28 out of 63 elements, even to the point of sharply changing their atomic weights. I shall hardly be mistaken if I say that the majority of us would have reacted negatively to such a discovery and would have called such a system of elements unnatural.

There is no doubt that Mendeleev discovered his system not on the basis of facts, but in spite of the facts. Here Mendeleev revealed that side of his mind which distinguishes genius from talent—a great intuition, a great and rare gift of nature, allowing one to see the truth through the husk of incorrect facts, inaccessible to the gaze of the overwhelming majority of people.

How great Mendeleev’s faith in the correctness of his idea was is evident from the fact that he took yet another great step forward and

in the same year 1871 predicted not only the existence of a number of unknown elements, but also their future properties, as well as the properties of their compounds. Of special historical interest are the predictions of the properties of 3 elements, named by Mendeleev ekaboron, ekaaluminium, and ekasilicon.

We should not be surprised that Mendeleev’s ideas were received in science, at best, with restraint; Kolbe’s warnings against attempts to replace experiment by speculations are understandable, as are Lyubavin’s remarks on the “roughly approximate” character of Mendeleev’s law, on the possibility of constructing new systems no worse than Mendeleev’s, and so on. On the contrary, we may be surprised at Richter, who already in 1874, in his textbook of inorganic chemistry, based the plan on the periodic system of the elements.

Many years were needed for the confirmation of Mendeleev’s ideas; the discovery of new elements was needed, fully confirming Mendeleev’s predictions, before the learned world paid due attention to his ideas.

It is interesting that the first major fact in favor of Mendeleev was not the confirmation of corrections he had made in the atomic weights of already known elements, but the confirmation of his predictions of the properties of new elements. In 1875 Lecoq de Boisbaudran discovered the new element gallium; in August of that year he gave a preliminary description of the method of discovering gallium and of some of its properties. In November of the same year Mendeleev published an article in which he indicated that the gallium discovered was nothing other than the ekaaluminium predicted by him, and that it must have an atomic weight of about 68, a specific gravity of 6.0—5.9, and an atomic volume of 11.5. The first determination of the specific gravity, made by Lecoq de Boisbaudran, gave the figure 4.7; having obtained larger quantities of the element, he determined its specific gravity and found it equal to 5.96. The atomic weight of gallium proved to be 69.9, the atomic volume—11.71.

In 1879 Nilson discovered scandium; in 1880 he was able to show that scandium possessed the properties predicted by Mendeleev for ekaboron2.

In 1884 the disputes over the atomic weight of beryllium were finally settled in favor of Mendeleev; gradually, long investigations confirmed, one after another, the corrections of atomic weights made—

given by Mendeleev in 1871; thus, in 1884 the atomic weight of uranium was finally recognized as equal to 240, in 1885 the atomic weight 138 was established for cerium, and in 1888, 113 for indium. Work that definitively confirmed the corrections, on the order of magnitude, of the atomic weights of the platinum metals continued until the 1890s.

However, apparently the strongest impression on contemporaries was made by Winkler’s discovery in 1886 of germanium, whose properties agreed astonishingly with the predicted properties of ekasilicon.

In Winkler’s opinion, one could demand no stronger proof of the correctness of the doctrine of the periodicity of the properties of the elements than this agreement between the properties of ekasilicon and germanium; it was not merely a confirmation of an ingenious theory, but “it signifies an eminent extension of the chemical field of vision, a mighty step into the realm of knowledge.”

These brilliantly justified predictions represent a triumph not only of Mendeleev: they are a great triumph of the human mind. When Leverrier and Adams discovered the new planet Neptune “at the point of a pen,” this was the pride of astronomy, and there is no textbook of astronomy in which this discovery is not mentioned. I believe that Mendeleev’s discovery and predictions are still more astonishing. Leverrier and Adams discovered Neptune by relying on visible irregularities in the motion of Uranus and on Newton’s universally recognized law. Mendeleev discovered elements and predicted their properties by relying on the empty spaces in the system he himself had created, and on a law he himself had discovered and that was far from universally recognized.

The discovery of the inert gases in the nineties of the last century not only did not create difficulties for the periodic system of the elements, but proved to be another triumph for it. These elements, which do not possess valence, formed a zero group in the system, transitional from the strongly electronegative elements of Group VII (the haloids) to the strongly positive metals of Group I. Following the discovery of the first representatives of the zero group (helium and argon), J. Thomsen, using Mendeleev’s method, predicted the existence of other elements of this same group and their atomic weights, which subsequently was also brilliantly confirmed.

TABLE 5

Atomic weights

Mendeleev also predicted other elements, which even now have not yet been discovered with complete certainty.

It is quite understandable that, with the passage of time, all chemists were bound to reckon with the periodic system, if only because of the services it rendered in the definitive establishment of the atomic weights of the elements; in this respect the periodic system had to be placed on a level with the laws of Avogadro, Dulong and Petit, and Mitscherlich (isomorphism). In the course of time the periodic system appears more and more often in manuals and textbooks of inorganic chemistry as a classificatory principle. And nevertheless, even at the beginning of the twentieth century, we encounter a very reserved attitude toward the periodic law and toward Mendeleev’s system among a number of major investigators. Thus Ostwald, in his Principles of Inorganic Chemistry, assigns the last pages to the periodic system, emphasizes its imperfections, and says (1900): “Here we are dealing not with a law of nature in the strict sense of the word, but with a principle of classification of something not fully defined.”

How is such a reserved attitude toward the periodic law to be explained? Undoubtedly, there were reasons for it. It was not a matter of the fact that, until recently, it was impossible to explain the irregularities in the placement of certain elements, such as iodine and tellurium, nickel and cobalt, argon and potassium; nor was it only a matter of the difficulty of placing the rare-earth elements. The chief cause of dissatisfaction with the periodic system lay in the obscure transitions in atomic weights from one element to another, in the fractional values of atomic weights, and in the incomprehensible entry of a number of elements into company alien to them (for example, Mn among the haloids). If the elements are placed in correctly arranged cells, then why are the differences between atomic weights so varied and so strangely varied?

Only the last decades have brought the solution, if not to all, then to the most important of these questions. The discovery of isotopy and of the ordinal number of the elements, in connection with modern theories of the structure of atoms, answered these questions and at the same time showed that the periodic law is a “law of nature in the strict sense of the word.”

Fate did not allow D. I. Mendeleev to live to see this final triumph of his principal work.

1. “Il n’est pas besoin d’insister, je crois, sur l’extrême importance qui s’attache à la confirmation des vues théoretiques de M. Mendeléef concernant la densité du nouvel élément” [Lecoq de Boisbaudran, Compt. rend. 83, 613 (1876)]. ↩

2. “... So bestätigten sich dadurch auf das augenscheinlichste die Spekulationen des russischen Chemikers, welche nicht nur die Existenz der genannten Grundstoffe (галлия и скандия) voraussehen liessen, sondern, auch die wesentlichsten Eigenschaften derselben im voraus anzugeben vermochten.” Nilson, Ber. d. deutsch. Ges. 13, 1442, 1450 (1880). ↩