Ulvospinel in the Titanomagnetites of Mt. Patyn

T. A. Nikolskaya

(Presented by the scientific seminar of the Department of Mineral Resources)

Ulvospinel \((\mathrm{Fe}_2\mathrm{TiO}_4)\) was established by Mogenson [8] in 1946. As an independent formation this mineral apparently does not occur, and it is known only as a product of the decomposition of titaniferous magnetite. It is recognized only at thousandfold magnification, and therefore is both detected and diagnosed with great difficulty and, quite possibly, is often overlooked. For this reason ulvospinel is mentioned extremely rarely in the literature; in the most complete mineralogical handbooks it is only noted, but not described [1, 2, 8]. It is absent from Mikheev [3]. The most complete characterization of this mineral is given by Ramdohr [6, 7] on the basis of his studies of titanomagnetite ores from Sweden, Germany, Africa, and other countries. The most convincing data on the reality of ulvospinel are found in Shchepochkina [4], who studied artificial compounds of titaniferous iron.

We discovered ulvospinel in the titanomagnetite ores of Mt. Patyn (Gornaya Shoria), in samples submitted for processing by S. S. Il’enko, by whom, incidentally, ulvospinel had previously been found in the titanomagnetite ores of Mt. Kultayga (oral communication).

The ore occurrence of Mt. Patyn is confined to a complex differentiated body of basic intrusive rocks. Among them, ulvospinel was encountered only in gabbro, which contains magnetite in an amount of 6.9%, isolated grains of ilmenite, always occurring in association with magnetite, and also isolated ocherous inclusions of limonite. Magnetite and ilmenite are later formations in relation to the main mass of silicates; they cement them, filling the interstices between grains of plagioclase and, chiefly, pyroxene, not infrequently forming a chain of small grains at the contact between them. The sizes of magnetite grains range from 0.3 to 0.7 mm; the ilmenite grains are always somewhat smaller.

The magnetite of Mt. Patyn is, in essence, titanomagnetite, which is confirmed both by chemical analysis and by the presence of characteristic decomposition structures of ilmenite in magnetite. The sizes of the largest ilmenite grains—plates—are \(0.15 \times 0.66\) mm. In addition, in the largest magnetite grains spinel is also observed as a decomposition product; it is sometimes detectable even at relatively low magnifications. The spinel is arranged along the \((100)\) faces of the magnetite, forming a distinctive lattice composed of rare, sharply elongated, needle-like segregations tapering at both ends and never intersecting one another (Fig. 1). The spaces between these needle-like se—

... inclusions prove to be covered with a dense network of ulvöspinel exsolutions, which appears in the form of thin, sometimes mutually intersecting prismatic crystals oriented parallel to the spinel, i.e., along the (100) face of magnetite.

The largest grains of ulvöspinel are always located near the spinel, but do not directly adjoin it; rather, they are separated from it by a thin “strip” of magnetite. Moreover, these largest ulvöspinel exsolutions are always considerably smaller than the spinel, not to mention the exsolutions that are situated at some distance from it, which are always much thinner and shorter. The ulvöspinel grains do not lend themselves to direct measurement, but, apparently, individual grains do not exceed 0.01 mm along the long axis. Thus, the largest spinel grains are 0.02 × 0.004 mm (Fig. 1).

Fig. 1.

The mineral described is very similar to ilmenite. The similarity is expressed in its close reflectivity, color, hardness, and form of occurrence. In contrast to ilmenite, ulvöspinel is isotropic and is confined to the cube faces (100) of magnetite, whereas ilmenite is arranged parallel to the octahedron faces [6, 7]. In addition, ulvöspinel never forms exsolutions cutting magnetite grains from one edge to the other, which is characteristic of ilmenite; it is characterized by shorter prismatic (Fig. 3) exsolutions.

Fig. 2.

Figures 1–3 show the exsolution structure of titanomagnetite. The black “needles” are spinel against a background of light magnetite. The dark-gray network within it is ulvöspinel. The white spots below near the crack (Fig. 2) are hematite. Magnification 1320 (with immersion).

In addition to the minerals described, hematite is sometimes encountered in magnetite at high magnifications in considerable quantity, in the form of white rims along cracks and along the periphery of magnetite grains, as well as along the contacts of ilmenite. Small grains of titanomagnetite are almost

... began to be martitized, whereas the decomposition products—the spinel and ulvöspinel—remain unchanged.

Thus, the titanomagnetite described represents a complex aggregate of minerals: magnetite, ilmenite, spinel, ulvöspinel, and hematite. The quantitative content of these components cannot be determined because of their extremely small size. Only in areas with the greatest concentration of coarse-grained spinel was it possible to carry out its quantitative count (from a photomicrograph) within part of a magnetite grain. The spinel content in such areas is 4.2%; consequently, its average content in the magnetite will be somewhat lower than this figure. Ulvöspinel, as is evident from Fig. 1, is present in considerably larger amounts. Naturally, the maximum quantity of decomposition products, as well as their largest segregations, is associated with the largest magnetite grains, in particular with their central parts. The abundance of decomposition products in the magnetite leads to the surface of its grains being so strongly pitted that it gives the impression of being poorly polished; in this respect titanomagnetite differs sharply from well-polished ilmenite grains.

Fig. 3.

We attempted to carry out a more or less complete study of the titanomagnetite in order to identify ulvöspinel in it. An analysis of the magnetic fraction was performed. The magnetite was ground into the finest powder, after which it was subjected to repeated magnetic separation in order to obtain as pure a magnetic fraction as possible. Nevertheless, as the chemical analysis showed (Table 1), it was not possible to obtain a pure magnetic fraction. Only after partially recalculating the chemical-analysis data to the mineralogical composition, allowing for removal of apatite, which is sometimes encountered in the magnetites, and of pyroxene, which is often found intergrown with them, do we obtain data that rather closely characterize the composition of the magnetite.

According to the classification of Dene [3], our magnetite, on the one hand, is undoubtedly titaniferous and, on the other, aluminous. The elevated content of $\mathrm{Fe_2O_3}$ is explained by significant martitization of the magnetite, and possibly in part also by the fact that the chemical analysis was carried out without taking the necessary care to separate ferric and ferrous iron.

Spectral analysis gives us a somewhat more complete picture of the composition of the magnetite (Table 2), which could not be established chemically because of the extremely limited amount of material at our disposal.

The thermal analysis carried out gave us a thermogram on which the heating curve shows an insignificant endothermic effect at a temperature of 500–700°, which corresponds to the decomposition temperature.

ulvöspinel with its transition into ilmenite and magnetite [4], as well as the loss of magnetism.

The specific gravity of the titanomagnetite fraction is very low—4.55, which may be explained by the abundance of inclusions and the admixture of silicates.

Table 1

Chemical-analysis data¹)

Table 2

Spectral-analysis data²)

X-ray structural analysis was carried out with magnetite powder and with the same powder, but partially dissolved in HCl in order to enrich it in spinel. The exposures were made on various X-ray instruments under different conditions. Chamber diameter—57.3 mm; copper or iron anticathode; diameter of the specimen column 0.5–1 mm; 35–40 mV; 18 mA; exposure from 2 to 12 hours; correction of the Debyegrams by a special exposure with NaCl. In all cases the data obtained reveal the interplanar spacings of magnetite and ilmenite, but in none of the Debyegrams are even the most characteristic lines of ulvöspinel recorded, as established for artificial ulvöspinel by Shchepochkina [4]. It is possible that this is the result of the fact that the principal interplanar spacings of ulvöspinel prove to be very close to those of mag-

¹) Performed in the laboratory of the Tomsk Integrated Expedition.
²) Performed in the spectral laboratory of the Tomsk Polytechnic Institute by L. Simakhina.

netite and ilmenite, sometimes even within the limits of permissible error (1.509 at \(i=10\) for ilmenite, 1.511 at \(i=9\) for ulvöspinel), and their superposition occurs. In our work these lines were always obtained broad and somewhat diffuse. The data available in published works concerning the dimensions of the unit cell are contradictory; they fluctuate within wide limits, namely \(a_0 = 8.47\ \text{Å}—8.53\ \text{Å}\) (5, 8); interplanar spacings are not given.

On the basis of the investigations carried out, we come to the conclusion that the presence of ulvöspinel in ores can be detected only by microscopic examination in immersion at high magnifications.

REFERENCES

1. Winchell A. N. Optical Mineralogy. Foreign Literature Publishing House, Moscow, 1949.

2. Dana J. D., Dana E. S. et al. The System of Mineralogy, vol. 1, part 2. Foreign Literature Publishing House, Moscow, 1951.

3. Mikheev V. I. X-ray Determinative Tables. Gostekhgeolizdat, 1957.

4. Shchepochkina N. I. Physicochemical investigations of barium and iron titanates. Transactions of the Institute of Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry, issue 11, Moscow, 1958.

5. Chevallier R., Bolfa I. et Mathieu. Titanomagnetites et ilmenites ferromagnetiques. Bull. Soc. frans. Miner. Crist. LXXVIII, 307–46, Paris, 1955.

6. Ramdohr Paul. Ulvöspinel and its signification in titaniferous iron ores. Econ. Geol., 48, 1953.

7. Ramdohr Paul. Die Erzminerale und ihre Verwachsungen. Drad. verlag, Berlin, 1955.

8. Ramdohr Paul. Die Beziehungen von Fe-Ti Erzen ans Magmatischen Gesteinen. Bull. de la commis. geol. de Finlande N 173. Helsinki, 1956.