|The triumph of Mendeleev's work lay in the ability to predict the existence of elements then unknown. The most famous of these predictions was for eka-silicon, or germanium; not only was the existence of the element postulated, but the properties of both it and its chloride and oxide, with considerable precision. The selection below is from his paper in the Journal of the Russian Chemical Society, 3: 25-56 (1871), which performs the same function for eka-boron.|
A NATURAL SYSTEM OF THE ELEMENTS AND ITS USE IN
And now, in order to clarify the matter further, I wish to draw some conclusions as to the chemical and physical properties of those elements which have not been placed in the system and which are still undiscovered but whose discovery is very probable. I think that until now we have not had any chance to foresee the absence of these or other elements, because we have had no order for their arrangement, and even less have we had occasion to predict the properties of such elements. An established system is limited by its order of known or discovered elements. With the periodic and atomic relations now shown to exist between all the atoms and the properties of their elements, we see the possibility not only of noting the absence of some of them but even of determining, and with great assurance and certainty, the properties of these as yet unknown elements; it is possible to predict their atomic weight, density in the free state or in the form of oxides, acidity or basicity, degree of oxidation, and ability to be reduced and to form double salts and to describe the properties of the metalloorganic compounds and chlorides of the given element; it is even possible also to describe the properties of some compounds of these unknown elements in still greater detail. Although at the present time it is not possible to say when one of these bodies which I have predicted will be discovered, yet the opportunity exists for finally convincing myself and other chemists of the truth of those hypotheses which lie at the base of the system I have drawn up. Personally, for me these assumptions have become so strong that, as in the case of indium, there is justification for the ideas which are based on the periodic law which lies at the base of all this study.
Among the ordinary elements, the lack of a number of analogues of boron and aluminum is very striking, that is, in group III, and it is certain that we lack an element of this group immediately following aluminum; this must be found in the even, or second, series, immediately after potassium and calcium. Since the atomic weights of these latter are near 40, and since then in this row the element of group IV, titanium, Ti = 50, follows, then the atomic weight of the missing element should be nearly 45. Since this element belongs to an even series, it should have more basic properties than the lower elements of group III, boron or aluminum, that is, its oxide, R203, should be a stronger base. An indication of this is that the oxide of titanium, TiO2, with the properties of a very weak acid, also shows many signs of being clearly basic. On the basis of these properties, the oxide of the metal should still be weak, like the weakly basic properties of titanium dioxide; compared to aluminum, this oxide should have a rnore strongly basic character, and therefore, probably, it should not decompose water, and it should combine with acids and alkalis to form simple salts; ammonia will not dissolve it, but perhaps the hydrate will dissolve weakly in potassium hydroxide, although the latter is doubtful because the element belongs to the even series and to a group of elements whose oxides contain a small amount of oxygen. I have decided to give this element the preliminary name of ekaboron, deriving the name from this, that it follows boron as the first element of the even group, and the syllable eka comes from the Sanskrit word meaning "one." Eb = 45. Ekaboron should be a metal with an atomic volume of about 15, because in the elements of the second series, and in all the even series, the atomic volume falls quickly as we go from the first group to the following ones. Actually, the volume of potassium is nearly 50, calcium nearly 25, titanium and vanadium nearly 9, and chromium, molybdenum, and iron nearly 7; thus the specific gravity of the metal should be close to 3.0, since its atomic weight 45. The metal will be nonvolatile, because all the metals in the even series of all the groups (except group I) are nonvolatile; hence it can hardly be discovered by the ordinary method of spectrum analysis. It should not decompose water at ordinary temperature, but at somewhat raised temperatures it should decompose it, as do many other metals of this series which form basic oxides. Finally, it will dissolve in acids. Its chloride EbCl3 (perhaps Eb2Cl6), should be a volatile substance but a salt, since it corresponds to a basic oxide. Water will act on it as it does on the chlorides of calcium and magnesium, that is, ekaboron chloride will be a hygroscopic body and will be able to evolve hydrogen chloride without having the character of a hydrochloride. Since the volume ot calcium chloride = 49 and that of titanium chloride = 109, the volume of ekaboron chloride should be close to 78, and therefore its specific gravity will probably be about 2.0 Ekaboron oxide, Eb2O3, should be a nonvolatile substance and probably should not fuse; it should be insoluble in water, because even calcium oxide is very slightly soluble in water, but it will probably dissolve in acid. Its specific volume should be about 39, because in the series potassium oxide has a volume of 35, CaO = 18, TiO = 20, and CrO8 = 36; that is, considered on the basis of a content of one atom of oxygen, the volume quickly falls to the right, thus, for potassium = 35, for calcium = 18, for titanium = 10, for chromium = 12, and therefore the volume for ekaboron oxide containing one atom of oxygen should be nearly 13, and so the formula Eb2O3 should correspond to a volume of about 39, and therefore anhydrous ekaboron oxide will have a specific gravity close to 3.5. Since it is a sufficiently strong base, this oxide should show little tendency to form alums, although it will probably give alum-forming compounds, that is, double salts with potassium sulfate. Finally, ekaboron will not form metalloorganic compounds, since it is one of the metals of an even series. Judging by the data now known for the elements which accompany cerium, none of them belong in the place which is assigned to ekaboron, so that this metal is certainly not one of the members of the cerium complex which is now known.
Mendeleev on Periodicity: 1 Periodicity index Home Page