Discovery of the New Chemical Elements 43 and 75.

A. I. Rabinovich.

Already Mendeleev had the possibility of predicting the properties of new, as yet undiscovered elements on the basis of the periodic dependence, discovered by him and by Lothar Meyer, of the chemical and physical properties of the elements on atomic weight. Moseley’s discovery made it possible to establish precisely the number of as yet undiscovered elements and to calculate in advance the frequencies of oscillations in their X-ray spectrum. The latter proved to be an indispensable method for identifying newly discovered elements and made it possible at once to insert them into the vacant places of the periodic system. Bohr’s theory of atomic structure, extended by him in 1922 to the whole system of elements, led to the discovery by Coster and Hevesy (von Hevesy) of a new element with atomic number 72, which before them had been incorrectly sought for (and erroneously found) among the rare earths1. Bohr’s theory, on the contrary, led to the conclusion that element 72 must be a homologue not of the rare earths, but of zirconium metal. Since similar elements are often encountered together, the search was directed toward zirconium minerals and led in 1923 to the discovery of a new element with the predicted X-ray spectrum, which received the name hafnium (Hf).

After this, the following places in the periodic system remained unfilled: in the group of rare-earth elements—61; in the seventh group, 43 and 75, where the homologues of manganese should be located; 85, situated under iodine, where one may expect the weakest halogen; and in the first group, 87, situated under cesium, where one may expect the strongest alkali metal. The last two elements are placed in the last rows of the periodic system, and therefore it is very probable that they are radioactive. It is possible that the rate of radioactive decay in them is very great, that the “lifetime” is very small, and that precisely for this reason they have not yet been detected.

Elements 43, 75, and 61 should not be radioactive. They have long attracted the attention of investigators. Unfortunately, many works directed toward finding them remained unpublished, since they did not lead to the desired result. Miolati2 advised looking for element 43, a homologue of manganese, in iron ores, which often accompany manganese. Bosanquer and Keeley3 searched for it in manganese ores and minerals, applying the sensitive method of X-ray spectroscopy, but without success.

Prandtl and Grimm4 carried out a very thorough chemical and spectrographic investigation of the rare-earth metals in search of element 61, but did not find even traces of this substance. The unsuccessful outcome of their attempts prompted the authors5 to suppose that element 61 does not exist at all and that its absence is connected with the absence of elements 43 and 75. Indeed, if the rare-earth elements are written in consecutive order, rather than placed all in one third group, as is usually done to express the extremely close similarity between them in chemical respect, then element 61 will prove to be in group VII under 43 and above 75, i.e. formally as if a homologue of manganese. The authors expressed

...hypothesis that, in general, the homologues of manganese (43, 61, 75, and 93, which should have stood after uranium) are incapable of existence.

However, this assumption has now been refuted by the work of Noddack, Tacke, and Berg (Walter Noddack und Ida Tacke, Otto Berg und Ida Tacke),¹ who succeeded in discovering two new elements at once: the 43rd and the 75th.

By their position in the periodic system, both are homologues of manganese. At the suggestion of Mendeleev, before their actual discovery they bore the names ekamanganese and dvimanganese.

The success of the latest attempt to detect the presence of these substances in the earth’s crust is to a considerable extent explained by the fact that, as in the case of hafnium, the rocks among which they had to be sought were correctly chosen. Noddack and Tacke express the following considerations on this point:

The metals of the last rows of the periodic system, beginning with scandium and ending with uranium (with the exception of the alkali and alkaline-earth metals), are grouped in nature into two most important groups. One of them is usually encountered as impurities in platinum ore, consisting predominantly of platinum and iron; these are — Cr, Mn, Co, Ni; Ru, Rh, Pd, Os, Ir; Cu, Ag, Au, forming as it were a closed island in the system (see Table 1):

TABLE 1.

The other is found in minerals—columbite, tantalite, and similar ones—consisting chiefly of oxides of Fe, Nb, Ta and containing as impurities several percent of Ti, V, Cr, Mn, Zn, As, Y, Zr, Sn, W, Pb, U, and also very small quantities (0.001–0.1%) of Se, Ga, Ge, Mo, Cd, In, Sb, La (and other rare earths), Hf, Bi, Th. As is seen from the table, the group of elements occurring in columbite minerals occupies in the periodic system two regions separated by the group of elements of platinum ore. Cr, Mn, and Fe occur both in this ore and in columbite minerals, i.e. at these elements the boundaries of both groups, as it were, overlap. Whether these boundaries pass below manganese, as indicated in the table, or also overlap, was unknown until the work of Noddack, Tacke, and Berg, who established that they do in fact overlap, i.e. that elements 43 and 75 occur in rocks of both groups.

The position of these elements in the system made it possible to predict with some probability not only where they should occur, but also in what quantity.

From Table II it is seen that, in the vertical groups adjacent to the manganese group, the homologues occur in the earth’s crust in approximately equal quantities. Consequently, ekamanganese and dvimanganese should also be distributed approximately equally. Furthermore, in this part of the periodic system elements with odd numbers are distributed approximately 10–20 times less abundantly than the even elements following them.² Consequently, the 43rd should be approximately

¹ Die Naturwissenschaften, 13, 567, 571 (1925).
² See, for example, Harkins, Journ. Amer. Chem. Soc. 39, 856 (1917).

so many times rarer than ruthenium, and the 75th—osmium, by as many times as Mn is rarer than Fe. This makes it possible to estimate the abundance of the 43rd and 75th, respectively, at \(10^{-13}\) and \(10^{-12}\) of the total substance of the earth’s crust (by weight).

TABLE II.

Thus the new elements are far rarer substances than the recently discovered hafnium.

By interpolation one may obtain an idea of some of the physical and chemical properties of the new elements. Their atomic weights should be equal to 98.0—99.5 and 187—188, their densities—11.5 and 21, their melting points—about 2300° and 3300° abs. Oxides of the types \(XO\), \(X_2O_3\), and \(XO_2\) should be dark substances, insoluble in acids; oxides of the type \(X_2O_7\)—light-colored, melting at 350°—500° and readily subliming at a somewhat higher temperature. Mercurous nitrate \(HgNO_3\), which gives precipitates with solutions of salts of all the neighboring elements, should also give precipitates with salts of the 43rd and 75th. Finally, it may be assumed that these elements will either not form sulfides capable of existing in aqueous solution at all, or will form only very unstable ones. This conclusion is drawn on the grounds that, in neighboring groups, elements of the same horizontal rows as the 43rd and 75th show a certain similarity in chemical properties to the elements of the preceding row standing one group to the left. Thus, Nb and Ta are similar in some respects to Ti; Mo and W—to V; Ru and Os—to Mn. It may therefore be considered that the 43rd and 75th will be similar to Cr, which, as is known, does not form a stable sulfide in water.

All the properties predicted here have not yet been tested experimentally, owing to the extremely small quantities of the new elements, which moreover have not yet been isolated in the individual state.

Without dwelling in detail on the methods of “enriching” the minerals containing the new elements, let us say only that one of them consisted in the following. 80 grams of platinum ore were dissolved in aqua regia, the solution was evaporated to dryness, and the residue was alternately ignited in a stream of hydrogen and oxygen. In this process \(OsO_4\) and \(RuO_4\) distilled over; black \(RuO_2\) and white \(As_2O_3\) accumulated on the walls of the quartz tube, as did a very small quantity of a white deposit consisting of microscopic soft needles. By all indications it coincided with the expected oxide of element 75. When hydrogen sulfide was passed through, it darkened, evidently giving a sulfide; when oxygen was passed through, it again became white; it dissolved in water, giving an acid reaction; with \(H_2S\) and \((NH_4)_2S\) it gave no precipitate. During further treatment the small quantity of the substance obtained was lost, and since the authors had no more platinum ore, they turned to columbite.

About a kilogram of the ground mineral was fused with caustic soda and saltpeter and extracted with water. In this process all Fe, as well as Nb and Ta in the form of Na salts, remained in the insoluble part. By repeatedly passing a stream of hydrogen sulfide through the solution, to which alkali and then acid were added in turn, all heavy metals were precipitated. Hydrochloric acid was added to the filtrate, and the solution was evaporated to a volume of 50 cm³. In this way the new elements were converted into chlorides, and then precipitated with mercurous nitrate together with Nb, Ta, As, Mo, W, Zn, Sn. The precipitate

DISCOVERY OF THE NEW CHEMICAL ELEMENTS 43 AND 75

was reduced in a stream of hydrogen, fused again, and precipitated as before. In this process part of the indicated elements again separated into the precipitate, and the content of the 43rd and 75th increased respectively to 0.5% and 5%. This preparation, as well as others obtained by enriching the minerals sperrylite, fergusonite, gadolinite, tantalite, and wolframite, were subjected to X-ray spectrography in the physical laboratory of the Siemens & Halske works.

The preparations were reduced by a stream of hydrogen to metals so that volatile oxides would not sublime from the hot anticathode. To obtain clear lines, 1 mg of a substance containing 0.1% of the sought element was sufficient. The substance was placed on the copper anticathode of a metal X-ray tube with an aluminum window and a heated cathode. The spectrograph was a Bragg spectrograph with a rotating rock-salt crystal. It made it possible to obtain spectra of elements beginning with Ti (22nd). In order that the substance should not remain too long on the anticathode, two rock-salt crystals were placed in the spectrograph at once, at different angles, giving simultaneously two different spectral regions. The results are given in the following Table III.

TABLE III.

The lines attributed to elements 43 and 75 cannot be interpreted as lines of higher orders of other elements, for example, the Kα₂ of neodymium, since the corresponding first-order lines are absent in the spectrogram. The line 75Lβ₁ coincides with the Lβ₅ line of tungsten (from the cathode), but the high intensity of the latter (about 9 in arbitrary units) shows that here there is in fact not one line, but two.

Thus the presence of element 43 in the minerals—columbite, sperrylite, and gadolinite—was proved on the basis of the coincidence of three spectral lines, and that of element 75 in columbite (traces of it were also found in tantalite and wolframite) on the basis of the presence of 4–5 lines, their intensity decreasing precisely in the order that was to be expected.

The investigation of other physical and chemical properties of the new elements has been postponed until they are obtained in more considerable quantities, but at present the spectral data leave no doubt that we are dealing precisely with elements 43 and 75 of the periodic system.

Noddack, Tacke, and Berg proposed for them the names: masurium (Ma) for the 43rd and rhenium (Re) for the 75th, after the names rejected by Germany’s provinces of East Prussia and the Rhine region.

1. On the discovery of element 72 and the dispute over priority caused by it, see the article by F. Paneth, “Achievements of the Physical Sciences,” vol. 4, 80 (1924). ↩

2. Abegg’s Handbuch der anorg. Chem. IV, 2, 629. ↩

3. Phil. Mag. [6] 48, 145 (1924). ↩

4. Zeitschr. f. anorg. u. allg. Chem. 136, 283 (1924). ↩

5. Cf. also Loring. Chem. News 225, 309, 386 (1922). ↩