ELECTRICAL PROPERTIES OF CRYSTALS OF SOLID SOLUTIONS OF ALKALI-HALIDE COMPOUNDS OF THE KCl—KBr SYSTEM

Yu. M. Annenkov, V. A. Grishukov

(Presented by the scientific seminar of the Laboratory of Energetics and Electronics of Ionic Structures)

Dielectric losses and electrical conductivity of solid solutions of KCl—KBr have been studied by a number of researchers [1–5]. According to the data of G. I. Potakhova [1], at frequencies \((3 \cdot 10^5—10^7)\) Hz, both at room temperature and at higher temperatures, the dielectric losses of the solid solution vary with composition along a curve with a maximum in the equimolar region. Somewhat later N. P. Bogoroditskii and B. A. Kulik [2] measured the dielectric losses of this same system and obtained, at a frequency of \(10^6\) Hz and a temperature of \(200^\circ\mathrm{C}\), an additive dependence on composition.

Schulz [3], studying the electrical conductivity of the KCl—KBr system, came to the conclusion that the magnitude of the electrical conductivity of the solid solution at a temperature of \(645^\circ\mathrm{C}\) does not greatly exceed the electrical conductivity of KCl and KBr. The maximum falls at the composition having the minimum melting temperature. According to the measurements of M. N. Treskina [4], the character of the dependence of the electrical conductivity of the solid solution in the region of intrinsic-conductivity temperatures and in the structure-sensitive region is the same and has a complex form, with a maximum for the composition 84 mole % KBr in KCl.

Our investigations, carried out on this same system [5], showed that \(\tg\delta\) and \(\sigma\), as functions of the composition of the solid solution, vary along a curve with negative nonadditivity. In a subsequent, more detailed study of the system [6], it was found that specimens of the same composition, but cleaved from different single crystals, have considerable differences in the values of \(\tg\delta\) and \(\sigma\). These differences increase in passing from KCl to KBr. It was shown that the considerable dependence of \(\tg\delta\) and \(\sigma\) on the composition of the solid solution and the large scatter of the values of these quantities for different single crystals of the same composition are due to unequal contents of uncontrolled divalent cation impurities in the specimens studied. Consequently, in order to determine the character of the change in electrical conductivity and \(\tg\delta\) caused by the nature of the solid solution, it is necessary to eliminate the influence of polyvalent impurities in the specimens studied.

There are reports in the literature [7, 8] that, when such anion-substituting impurities as OH\(^{-}\) and CO\(_3^{-}\) are introduced into alkali-halide crystals, an effect of compensation of the cationic divalent impurity is observed. It is assumed that the anionic impurities, combining with Ca\(^{2}\), form insoluble compounds Ca(OH)\(_2\) and CaCO\(_3\). To clarify

the nature of the dependence of the electrical properties of the solid solution on its composition, single crystals with additionally introduced OH\(^{-}\) impurities were grown and investigated.

Single crystals were grown by the Kyropoulos method from salts of “chda” and “khch” grades. The specimens were cleaved in the form of plates measuring \(18 \times 18 \times 0.5\) mm in planes perpendicular to the growth direction of the single crystal. Measurements of \(\operatorname{tg}\delta\) and \(\sigma\) were carried out on an apparatus with a sensitivity of the measuring circuit for \(\operatorname{tg}\delta\) of \(1 \cdot 10^{-4}\), and for current of \(1 \cdot 10^{-14}\) A. The measurements were made in a vacuum of not less than \(1 \cdot 10^{-2}\) mm Hg.

The investigations showed that, when OH\(^{-}\) impurities are introduced into the crystals, the electrical conductivity of the specimens sharply decreases in comparison with “pure” crystals. This decrease increases on going from KBr to KCl. The maximum decrease (by a factor of 10) in the electrical conductivity for KBr is observed when 0.02 mol % KOH is introduced into KBr, while for KCl an electrical conductivity 1000 times smaller is observed at a KOH concentration of 0.2 mol %. These results show that, indeed, when OH\(^{-}\) impurities are introduced, an effective compensation of the influence of the divalent impurity is observed. The fact that, in the case of KCl, the greatest amount of KOH is required to achieve the minimum electrical conductivity confirms the conclusion we previously drew, namely that the instability of the value of the electrical conductivity in individual specimens of one composition, and the strong dependence of the electrical conductivity of the solid solution on composition, are most probably due to the unequal content of “uncontrolled” divalent cation impurities [6]. The considerably greater content of divalent impurities in KCl in comparison with KBr also indicates the significant decrease in the electrical conductivity in KCl.

The values of the electrical conductivities of crystals of the solid solution corresponding to maximum compensation of the isovalent cation impurities were selected and compared. The results of such a comparison are given in Fig. 1. As is seen from Fig. 1, whereas in the case of “pure” crystals (curve 1) the electrical conductivity decreased on going from KCl to KBr, for crystals with an OH\(^{-}\) impurity (curve 2) the electrical conductivity increases on going from KCl to KBr, varying along a curve with a positive deviation from additivity. The fact that the nature of the dependence of the electrical conductivity on the composition of the solid solution obtained for crystals with an OH\(^{-}\) impurity is analogous to the dependence obtained by Schulz [3] in the temperature range where the influence of divalent impurities is not manifested makes it possible to assume that: first, the nature of the dependence of the electrical conductivity on the composition of the solid solution obtained by us upon introduction of OH\(^{-}\) impurities is determined by the structure of the solid solution; second, the method of introducing anionic

Fig. 1. Change in the electrical conductivity of “pure” specimens (curve 1) and specimens with an OH\(^{-}\) impurity (curve 2) of the KCl–KBr solid solution as a function of composition at a temperature of 202°C.

impurities in some cases can be used effectively to eliminate the influence of isovalent cation impurities.

In the temperature interval \((20—260)^\circ\mathrm{C}\) and the frequency range \((30—5000)\) Hz, \(\operatorname{tg}\delta\) varies along a curve with two characteristic regions [9]. At high temperatures, where \(\operatorname{tg}\delta\) is determined by losses due to through conductivity, the nature of the variation of \(\operatorname{tg}\delta\) with the composition of the solid solution is analogous to the variation of electrical conductivity and is not presented in this article.

REFERENCES

1. G. I. Potakhova. Dissertation. Tomsk, 1956.
2. N. P. Bogoroditskii, B. D. Kulik. ZhTF, 10, 2165, 1958.
3. Schulze H., Thesis, Göttingen University, 1952.
4. M. N. Treskina. Dissertation. Tomsk, 1965.
5. Yu. M. Annenkov, V. A. Grishukov. Izv. TPI, vol. 139, Tomsk, 1965.
6. Yu. M. Annenkov, V. A. Grishukov. Izv. vuzov—Fizika (in press).
7. N. A. Tsal’, V. M. Pashkovskii, I. D. Nabitovich. FTT, 6, 6, 1828, 1964.
8. M. I. Kornfeld, V. V. Lemanov. FTT, 7, 7, 2249, 1965.
9. Yu. M. Annenkov, E. K. Zavadovskaya. Izv. TPI, vol. 139, Tomsk, 1965.