DISTRIBUTION OF GOLD IN ROCKS AND MINERALS OF GRANITOID MASSIFS OF ALTAI, TRANSBAIKALIA, AND THE URALS

T. A. ANOSHIN, I. N. BUSHLYAKOV, V. V. POTAP’EV
(IG and G, Siberian Branch, Academy of Sciences of the USSR)

In recent years the number of works on the geochemistry of gold has sharply increased. This, in our opinion, is connected above all with two reasons: a) the attempt to find more precise criteria for the connection of gold mineralization with magmatic activity, and b) the creation of reliable and sensitive methods for determining the clarke contents of gold in various objects. In this respect, work has proceeded both along the line of introducing modern nuclear-physical methods (radioactivation analysis) and along the line of improving classical, chemical and spectral methods associated with concentration of the substance (for example, the spectrochemical method for determining gold) [3, 13]. In this report we shall consider certain questions of the geochemistry of gold obtained as a result of studying its distribution in granitoids.

As is known, geologists and geochemists have long been discussing the question of so-called specialized intrusions. And indeed, in many regions a connection is established quite clearly between intrusions of a definite age and mineralization of specific composition. Tin-bearing, tantalum-bearing, gold-bearing, etc., intrusions are accordingly distinguished. Much work has been carried out to clarify the reasons for the ore specialization of intrusions; however, many contradictions in this problem have still not been eliminated.

Many investigators come to the conclusion that a sign and proof of the ore specialization of an intrusion is the increased content of the ore element in its rocks. On the basis that the total content of ore elements in deposits and ore occurrences accompanying an ore-bearing intrusion does not exceed 0.01–0.1% of the total amount of these elements in the magmatic body (calculations by N. I. Safronov [7]), the conclusion is drawn that their concentration during recrystallization of intrusions is not noticeably disturbed and that from the content of the ore element in the magmatic rock one may judge its content in the melt. Especially many works in this direction have been carried out for Sn and Ta. Thus, according to V. L. Barsukov [1], the content of Sn in tin-bearing granites is 2–3 times greater than in non-tin-bearing granites. According to the data of E. B. Znamenskii [6], Zalashkova N. E. [5], A. A. Sitnin, and some other investigators, granitoids accompanied by tantalum mineralization are characterized by an increased content of tan-

gallium and niobium. Thus, if one adopts the position of these authors, it must be assumed that, for the potential ore-bearing capacity of an intrusion, it is necessary that the magma that formed it was enriched in the given ore element. It should be said that even for these two elements, which are cited as the most convincing examples of specialization of granitoid magmas, in some cases the analytical material obtained is contradictory and does not permit such a conclusion to be drawn [10].

Data on elevated Au contents in the rocks of ore-bearing intrusions, as though indicating their specific character, are also cited in the literature, but much more rarely, since in general there are exceptionally few data on the distribution of gold in intrusive rocks. It may be said that, before the work of Yu. G. Shcherbakov and G. A. Perezhogin [11, 12], there were only isolated data on the gold content in granitoids. There was also a series of determinations of gold in the international granite standard G-1. The gold content in granite G-1 is, on average, 0.0045 g/t. As studies of the distribution of gold in granitoids advanced, a tendency emerged toward lowering the Clarke value of gold in granitoids, both as a whole and for individual regions.

For the purpose of a detailed study of the distribution of gold in granitoids, we selected intrusive bodies from different regions characterized by different ore mineralization: the Kolyvan massif in Altai, the Khangilai-Shilinsk massif in Eastern Transbaikalia, and the Verkh-Iset and Suunduk massifs in the Urals.

A detailed geological study of these intrusions was carried out beforehand, and work was performed to establish the relationship of ore mineralization with the process of their formation. The multiphase character of the formation of these intrusions was established. For three massifs, a relationship was established between mineralization of different composition and the crystallization of granites of different phases.

Gold deposits of the quartz-vein type are genetically associated with the Verkh-Iset and Suunduk massifs; weak manifestations of gold mineralization are noted in association with the Kolyvan massif. The Khangilai-Shilinsk massif, with which rare-metal occurrences of various compositions are known, is not accompanied by gold mineralization.

Most analyses of bulk samples and all monomineralic fractions were performed by the neutron-activation method developed by G. A. Perezhogin. This procedure is widely known and has been published [9]. As applied to the present work, the procedure was somewhat modified, especially for the analysis of monomineralic samples.

It should be noted that the radioactivation method is at present the most sensitive for determining gold. However, in our opinion, it is now necessary to conduct the analysis of standard samples on a broader scale in order to increase the interlaboratory reproducibility of this method. High sensitivity by no means excludes the presence of systematic error.

Bulk samples from the Suunduk massif were analyzed by the spectrochemical method improved by V. G. Tsimbalist [13]. We note that good agreement was achieved between the results obtained by the two methods.

The average gold content in the rocks of the Verkh-Iset massif (Table 1): diorites and granodiorites [14]—0.0011 g/t; biotite, biotite-amphibole, leucocratic granites, plagiogranites [18]—0.0017 g/t.

In the rocks of the Suunduk massif: diorites and granodiorites [7]—0.0005 g/t; biotite granites [4]—0.0005 g/t. In biotite granites

of the Kolyvan massif [20]—0.0012 g/t. In the biotite granites of the Khangilai–Shilinsky massif [7]—0.0012 g/t.

All analyses were processed by methods of mathematical statistics. Comparison of variances was carried out using Fisher’s \(F\)-test; comparison of means was carried out using Student’s \(t\)-test. In most cases, at the 5% significance level, hypotheses concerning differences in means and variances were rejected. Thus, a statistical study of the distribution of gold in granitoids of different regions confirms the absence of real differences in its content. This fact of similar gold contents in granitoids, both “ore-bearing” and

Table 1

Gold content in granitoids

“barren,” of different ages and different regions, does not allow one to conclude that their magmas were enriched in gold. We also note the fact of similar gold contents in granitoids of different phases, accompanied by ores of different composition and composed of rocks of different composition.

The question of the gold content in intrusive rocks of different composition was first raised by Vincent and Crocket in their work devoted to the study of the geochemistry of gold in the Skaergaard differentiated intrusion of basic composition (Greenland) [16]. Vincent and Crocket came to the conclusion that the gold content in the Skaergaard intrusion changes little from more basic rocks to more acidic ones. In later works, Yu. G. Shcherbakov [11, 12], on the basis of study of the Sarakokshinsky massif (Gorny Altai) and the Anakit trap intrusion (Siberian Platform), speaks of a certain decrease in gold during the transition from early phases of more basic composition to later phases of acidic composition. In the work of N. T. Voskresenskaya and N. F. Zvereva [4], on the basis of study of the distribution of gold in the granitoids of Northern Kazakhstan, a conclusion is also drawn regarding similar gold contents in rocks of the granitoid series, from diorites to granites. According to their data, differences in gold contents in rocks of different phases are statistically insignificant.

The authors of the present report have also carried out a detailed study of the distribution of gold in rock-forming and accessory minerals of granitoids. It should be noted that, although the question of the distribution of gold in rock-forming minerals has been considered by individual researchers [12, 15, 16], there are no clear ideas in the literature concerning the distribution of gold among the minerals of granitoids.

We analyzed more than 160 samples of rock-forming and accessory minerals of granitoids. These results are given in Table 2.

Table 2

Au content in rock-forming and accessory minerals of granitoids

Note: $s^2_{lg}$ is the variance of the logarithms of Au contents.

As can be seen from Table 2, all minerals of granitoids show similar gold contents. In any case, the differences in Au content among them are statistically insignificant. On the basis of these data, we can state with confidence that none of the minerals studied is a mineral concentrator of gold, whereas the existence of such mineral concentrators has been well demonstrated for a number of other elements. Thus, one may speak of the dispersion of gold in the minerals of granitoids. But at present it is difficult to answer the question: in what form does the dispersion of gold occur in minerals and rocks—atomic or ionic? As is known, Vincent and Crocket [16] advanced the hypothesis that gold is dispersed in rocks and minerals in the form of uncharged atoms. Their conclusions were based mainly on a study of the Skaergaard differentiated intrusion. Later, Yu. G. Shcherbakov [12] proposed a hypothesis of a partial chemical bond of gold in silicates on the basis of the positive correlation of Au contents with Cu, Fe, V, and chiefly with Mg. Our data support the hypothesis that the form of Au dispersion is that of uncharged atoms. Indeed, from the classical standpoint of isomorphism it is difficult to explain, for example, isomorphism between $\mathrm{Mg}^{+2}$ and $\mathrm{Au}^{+}$ in magnesian silicates, since the ionic radius of $\mathrm{Mg}^{+2}$, according to Ahrens [14], is 0.66 Å, and the radius of $\mathrm{Au}^{+}$ is 1.37 Å, i.e., the radii differ by more than a factor of two, whereas, as a rule, isomorphism permits a difference in the ionic radii of replacing ions of no more than 15%. In addition, the chemical bond Mg—O in silicates is considerably more ionic than the Au—O bond. The electronegativity of $\mathrm{Mg}^{+2}$ is 1.2, and that of $\mathrm{Au}^{+}$ is 2.3 [2], i.e., the incorporation of gold into the lattices of magnesian silicates will have a weakening effect.

True, recently V. I. Lebedev [8] proposed using in geochemistry and crystal chemistry a new system of radii, the so-called orbital atomic radii. From this standpoint he was able to explain a whole series of deviations from Goldschmidt’s rules of isomorphism. At the same time, he explained the absence of isomorphism between elements having similar effective ionic radii. For example, the absence of isomorphism between $\mathrm{Na}^{+}$ and $\mathrm{Cu}^{+}$, $\mathrm{K}^{+}$ and $\mathrm{Au}^{+}$, to which attention was drawn...

Arens’s data [14]. If we turn to the orbital atomic radii, Au—1.187 Å and Mg—1.370 Å, it can be seen that the difference between them is less than 15%, i.e., according to V. I. Lebedev’s concept, isomorphism between them may quite well be admitted. However, this line of inquiry has not yet been developed.

Sincent and Crockett [16] drew attention to the association of Au in basic rocks with copper sulfides. From the standpoint of effective ionic radii and atomic orbital radii, isomorphism between \(\mathrm{Cu}^+\) and \(\mathrm{Au}^+\) may to a certain extent be admitted. However, in the already mentioned work N. T. Voskresenskaya and N. F. Zvereva show that they found no sufficiently reliable positive correlation relationships between Cu and Au.

Conclusion

The study carried out on the distribution of gold in granitoid massifs accompanied by various mineralization does not allow one to speak of an ore specialization of granitoid magmas. In the present report this conclusion is substantiated only by material on gold, but we have analogous material for Ta, Nb, and a number of other ore elements.

Those cases in which an elevated gold content is recorded in the rocks of intrusions associated with Au mineralization should be investigated from the standpoint of the possibility that disseminated hydrothermal mineralization was superimposed on them, since study of rocks of intrusions that have undergone hydrothermal metamorphism indicates the mobility of gold in the process of hydrothermal metamorphism and the possibility of its redeposition.

At the same time, in a number of cases enrichment of a magmatic melt with gold is possible, but, apparently, enrichment with one or another ore element, including gold, can occur only in separate parts of intrusions—for example, in cupolas and apophyses, where movement and enrichment of the melt by volatiles, the phenomenon of boiling, etc., take place; and it will thus have an acquired character.

LITERATURE

1. Barsukov V. L. On the metallogenic specialization of granitoid intrusions. In: Chemistry of the Earth’s Crust, vol. 11; Moscow, Nauka Publishing House, 1964.

2. Batsanov S. S. Electronegativity and Chemical Bonding. Novosibirsk, Publishing House of the Siberian Branch of the Academy of Sciences of the USSR, 1962.

3. Voskresenskaya N. T., Zvereva N. F., Rivkina L. A. Spectrochemical determination of gold in silicate rocks and minerals. Zh. “Analit. khimiya”, vol. XX, issue 12, 1965.

4. Voskresenskaya N. T., Zvereva N. F. Some problems of the geochemistry of gold in connection with the ore-bearing capacity of magmatic complexes of Northern Kazakhstan. Geokhimiya, No. 4, 1968.

5. Zalashkova N. E., Sitnin A. A. Features of the distribution of rare elements in tantalum-bearing apogranites and biotite granites. Geologiya rudnykh mestorozhdenii, No. 6, 1967.

6. Znamenskii E. B. On the average contents of niobium and tantalum in granitoids. In: Chemistry of the Earth’s Crust, vol. 11, Moscow, Nauka Publishing House, 1964.

7. Kontev-Dvornikov V. S., Rub M. G. On the geochemical specialization of magmatic complexes. In: Metallogenic Specialization of Magmatic Complexes, Moscow, Nedra Publishing House, 1964.

8. Lebedev V. I. Orbital sizes of atoms—a new system of radii—and regularities determining interatomic distances and isomorphism. Vestnik LGU, No. 6, 1967.

1. Perezhogin G. A., Alimarin I. P. Neutron-activation determination of gold in rocks and meteorites. Zh. “Analit. khimiya,” vol. XX, issue 8, 1965.

2. Potap’ev V. V., Malikova I. N., Grebennikov V. M., Dorosh V. M. Geochemical indicators of tantalum mineralization in granitoids. DAN SSSR, vol. 173, No. 4, 1967.

3. Shcherbakov Yu. G., Perezhogin G. A. On the geochemical association of gold mineralization with intrusions and enclosing rocks. “Geokhimiya,” No. 9, 1963.

4. Shcherbakov Yu. G., Perezhogin G. A. On the geochemistry of gold. “Geokhimiya,” No. 6, 1964.

5. Tsimbalist V. G. Determination of microquantities of gold in geochemical objects. “Geologiya i geofizika,” No. 12, 1968.

6. Ahrens L. H. The significance of the chemical bond form controlling the geochemical distribution of the Elements. Part 1. Physics and Chemistry of Earth. 1964, vol. 5.

7. Mantei E. J., Brownlow A. Variation in gold content of minerals of the Marysville quartz diorite Stock, Montana. — Geochim. et cosmochim. acta, vol. 31, No. 2, 1967.

8. Vincent E. A., Crocket J. H. Studies in the geochemistry of gold I–II. Geochim. et cosmochim. acta, vol. 18, Nos. 1–2, 1960.