EFFECT OF DEFECTIVENESS OF THE CATION SUBLATTICE ON THE FORMATION OF COLOR CENTERS

I. Ya. Melik-Gaikazyan, V. M. Lisitsyn, L. I. Roshchina

(The article was presented by the scientific seminar of the Radiation Spectroscopy Laboratory of the Research Institute of Nuclear Physics)

Despite the great progress achieved in recent years in the development of radiation physics, one of the main problems—the mechanism of defect formation by radiation—has still not been finally solved. The chief difficulty in solving the problem is that the efficiency of the formation process and the stability of the defects formed depend strongly on many factors: temperature, the presence of impurities, dislocations, etc. Clarifying the nature of the influence of one of these factors on radiation properties makes it possible to simplify the solution of the entire problem.

The possibility of a controlled change in the concentration of impurities with various known properties in a crystal makes it promising to carry out studies of the influence of impurities on the radiation properties of alkali-halide crystals. Therefore, much attention has recently been devoted to this question. The mechanism of the influence of an impurity on the formation of defects by radiation in crystals has not yet been fully clarified; however, two hypotheses have been formulated. Crawford and Nelson [1], on the basis of the results obtained, assumed that under the action of ionizing radiation anions may be displaced into cation vacancies introduced into the crystal together with a divalent cation-substitutional impurity, with the formation of \(F\)- and \(X_2^{-}\)-centers. An \(X_2^{-}\)-center is a molecular halide ion consisting of two halides oriented along the \([100]\) axis, one of which is located in a cation vacancy. This assumption was further developed in works [3, 4]. In work [4] it was shown that the presence of an Sr impurity in KCl and KBr crystals leads to an increase, by several times, in the rate of generation of anion vacancies even at room temperatures. It is assumed that the observed increase in the rate of generation of anion vacancies is caused by the displacement of halides into cation vacancies with the formation of \(F\)- and \(V_2\)- or \(V_3\)-centers. Another assumption was put forward by Sibley and Sonder [5]. They believe that the presence of an impurity in a crystal promotes stabilization of halides displaced into interstices; this leads to an increase in the rate of accumulation of \(F\)-centers at \(80^\circ\mathrm{K}\). Sibley and Sonder advance this assumption with reference to the work of Itoh et al. [6], where it was shown that in KBr crystals with a KCl impurity the stability of the interstitial defects formed is higher than in a pure crystal.

KBr. However, in the work cited it is well proven that the impurity itself is “stabilized” in interstitial positions, but in no way stabilizes the halides of the host substance. In addition, in the work of Faraday and Compton [7] it is shown that, in a crystal irradiated at 80°K, a quite definite relation is always observed between the concentration of $F$-centers and the sum of the concentrations of $V_{2}$-, $V_{4}$-, and $V_{7}$-centers.

Since $V_{2}$-, $V_{4}$-, and $V_{7}$-centers do not include halides displaced into interstitial sites but apparently do contain a halide displaced into a cation vacancy, the first assumption seems more probable. To clarify the role of cation vacancies in the process of defect creation by ionizing radiation, the present work was undertaken. Comprehensive studies were carried out of the kinetics of accumulation of $F$-, $V_{2}$-, and $V_{3}$-centers and of the kinetics of the radiation-induced decrease in electrical conductivity, which made it possible to monitor the change in the concentrations of anion and cation vacancies in the crystal with irradiation time and to compare them.

For the investigation, crystals of NaCl, KCl, NaBr, and KBr with different contents of Ca and Sr impurities were selected. The crystals were grown from the melt by the Kyropoulos method; the impurity, in the form of a salt, was added to the charge of the host substance. The impurity content in the crystal was determined by the complexometric method using an FEK-56 nephelometer colorimeter. The crystals were irradiated on the TPI cyclotron, which accelerated protons to an energy of 4.5 MeV. The absorption spectra were measured on an SFD-2 spectrophotometer. Electrical-conductivity measurements were carried out on the setup described in [11]. Crystals with dimensions $12 \times 7 \times 1\ \mathrm{mm}^{3}$ were cleaved for measuring absorption spectra, and crystals of $10 \times 10 \times 0.5\ \mathrm{mm}^{3}$ for electrical conductivity, from the same single crystals. To measure electrical conductivity the crystals were irradiated from two sides in such a way that, between the colored parts of the crystal, an uncolored layer remained with a thickness of no more than 5% of the total thickness. Under these conditions the error in measuring electrical conductivity due to nonuniform coloring of the crystal through its thickness did not exceed 1%. This was achieved by selecting the sample thickness and the depth of the colored layer of the crystal.

As a result of the investigations carried out, the following results were obtained. The presence of alkaline-earth-metal impurities in all alkali-halide crystals leads to an increase in the number of accumulated $F$-centers at the first stage of accumulation in comparison with pure crystals (Fig. 1). The formation of $F$-centers at the first stage of accumulation occurs mainly at vacancies existing in the crystal before irradiation [8]. Apparently, complexes of vacancies enter the crystals during growth together with the impurity, in order to compensate elastic stresses arising from the mismatch between the sizes of the impurity and host-substance ions. This is confirmed by a number of experiments [9]. In KCl and KBr crystals the presence of impurity leads to a 2–5-fold increase in the rate of accumulation of $F$-centers in the region of slow linear accumulation (at the 2nd stage).

We have shown [4] that the increase in the rate of accumulation of $F$-centers in the linear region of accumulation of the kinetic curve is due mainly to an increase in the generation rate of anion vacancies. This effect can be explained by the inclusion of a new source of generation of anion vacancies, namely, the generation of anion vacancies due to displacement of halides into cation vacancies. Studies of the kinetics of accumulation of $F$-centers in the linear region in NaCl and NaBr crystals containing impurities confirm this assumption. Since in these crystals the difference between the sizes of the halides and the metal is considerably greater than in KCl and KBr crystals (Table 1), it follows—

makes it possible to expect that the process of displacement of halides into cation vacancies in NaCl and NaBr crystals is hindered, since the geometrical parameters of the lattice strongly affect the processes occurring in crystals [10]. We have shown experimentally that, in NaCl and NaBr crystals, no increase in the rate of accumulation of \(F\)-centers in the linear region (in the second stage) is observed upon addition of an impurity (Fig. 1).

Fig. 1. Kinetics of accumulation of \(F\)-centers in NaCl and KBr crystals with different contents of alkaline-earth-metal impurities: a) 1—pure NaCl; 2—NaCl + 0.0143 mol % CaCl\(_2\); 3—NaCl + 0.186 mol % CaCl\(_2\); 4—NaCl + 0.0303 mol % CaCl\(_2\); 5—NaCl + 0.0432 mol % CaCl\(_2\); b) 1—KBr; 2—KBr + 0.0026 mol % SrBr\(_2\); 3—KBr + 0.00938 mol % SrBr\(_2\); 4—KBr + 0.0115 mol % SrBr\(_2\).

Halides displaced into cation vacancies form, according to the models currently accepted, \(V_2\)- or \(V_3\)-centers. We have carried out studies of absorption spectra in the region of hole bands and of the accumulation of \(V_2\)- and \(V_3\)-centers.

At room temperatures, under the action of radiation, intense \(V_2\)- and \(V_3\)-absorption bands are formed in the crystals studied. In doped KCl and KBr crystals, predominantly \(V_2\)-...

centers, whereas the number of induced centers is approximately equal to the number of \(V_3\)-centers induced at the same doses in pure crystals. In NaCl crystals, both doped and pure, predominantly

Table 1

Ratios of anion radii \((r_a)\) to cation radii \((r_k)\) in some alkali-halide crystals

\(V_3\)-centers are formed. These results make it possible to assume that, with an increase in the rate of accumulation of \(F\)-centers in KCl and KBr upon introduction of an impurity into the crystal, the formation of \(V_2\)-, but not \(V_3\)-centers is involved. A study of the kinetics of accumulation of \(V_2\)-centers in crystals with different impurity contents confirms this (Fig. 2). In crystals containing an alkaline-earth-metal impurity, the number of \(V_2\)-centers accumulated in the initial

Fig. 2. Kinetics of accumulation of \(V_2\)-centers in crystals: 1—KBr; 2—KBr + 0.0026 mol % SrBr\(_2\); 3—KBr + 0.0073 mol % SrBr\(_2\); 4—KBr + 0.00938 mol % SrBr\(_2\); 5—KBr + 0.0135 mol % SrBr\(_2\).

period and the rate of their accumulation in the second stage increase. Consequently, the rate of displacement of halides into cation vacancies increases.

To clarify the nature of the change in the concentration of single cation vacancies, we studied the dependence of the electrical conductivity of the crystals on the irradiation dose. It is known that the electrical conductivity of alkali-halide crystals is due to free single cation vacancies. Under the action of radiation, a decrease in electrical conductivity is observed, i.e., a decrease in the number of cation vacancies in the crystal occurs. One of the processes leading to a decrease in their number may be the process of displacement of halides into cation vacancies.

When an impurity of alkaline-earth metals is introduced into the crystals, the content of isolated cation vacancies increases considerably [11]. Therefore, the electrical conductivity of doped crystals in the structure-sensitive region [12] is higher than that of pure NaCl and KCl (Fig. 3). From the results presented in Fig. 3 it is seen that, under the action of ionizing radiation, the electrical conductivity $\sigma$ decreases both in pure and in doped crystals. A comparison of $\lg \sigma$ at identical temperatures makes it possible to say that the radiation-induced change

Fig. 3. Temperature dependence of the electrical conductivity in pure and impurity-containing crystals before and after irradiation with protons for 10 min:
a) 1—unirradiated NaCl; 2—irradiated NaCl; 3—NaCl + 0.0303 mol % CaCl$_2$, unirradiated; 4—the same, irradiated.
b) 1—unirradiated KCl; 2—irradiated KCl; 3—KCl + 0.0085 mol % SrCl$_2$, unirradiated; 4—the same, irradiated.

in electrical conductivity in a doped crystal is greater than the radiation-induced change in electrical conductivity in a pure one. This means that, in doped crystals, a considerably larger number of cation vacancies than in pure crystals drops out of the current-transport process upon irradiation. However, the concentration of free cation vacancies is always higher in doped crystals, even irradiated ones, than in pure crystals. This fact permits the conclusion that not all cation vacancies are bound under the action of radiation.

A study of the kinetics of radiation-induced change in electrical conductivity at a temperature at which \(V_2\)- and \(V_3\)-centers are stable in pure and doped KCl and NaCl crystals led to the following results (Fig. 4). At small irradiation doses, a sharp decrease in electrical conductivity is observed in all crystals. Then, with increasing irradiation dose, the decrease in electrical conductivity reaches a certain value and does not change within the range of doses investigated by us. In a pure KCl crystal, the electrical conductivity could not be measured after irradiation at the temperature chosen by us, \(60^\circ\)C, since the current was below the sensitivity limit of the apparatus. Therefore, Fig. 4 gives the results of electrical-conductivity measurements at a temperature of \(125^\circ\)C. The continuous slow decrease in electrical conductivity with irradiation dose in the KCl crystal is apparently due to the change in the ratio of \(V_2\)- and \(V_3\)-centers and to their different stability.

Fig. 4. Kinetics of the radiation-induced change in electrical conductivity with irradiation dose in crystals:
a) 1—NaCl; 2—NaCl + 0.0303 mol % \(\mathrm{CaCl_2}\);
b) 3—KCl; 4—KCl + 0.0085 mol % \(\mathrm{SrCl_2}\).

From the results presented it follows that the decrease in electrical conductivity occurs identically in KCl and NaCl crystals. Comparing the kinetic curves for the accumulation of \(F\)-centers and the curves of the radiation-induced change in electrical conductivity, one can see that the rapid stage of the change in electrical conductivity ends considerably earlier than the first stage of accumulation of \(F\)-centers. Consequently, the process of binding single cation vacancies occurs in the initial period of irradiation. It may be assumed that halogens are displaced into single cation vacancies. This is in good agreement with the increase in the first stage of accumulation of \(F\)- and \(V_2\)-centers in the KCl crystal. In the NaCl crystal, however, displacement into cation vacancies is unlikely. The decrease in electrical conductivity in the initial period of irradiation is apparently associated with the displacement of halides into vacancy complexes.

From a comparison of the kinetics of accumulation of \(F\)- and \(V_2\)-centers it follows that the generation rate and the rate of displacement of halides into cation vacancies are higher in doped crystals than in pure ones. A study of the kinetics of the radiation-induced change in electrical conductivity leads to the conclusion that the number of single cation vacancies remains constant during prolonged irradiation. Consequently, the increase in the rate of generation of anion vacancies is not due to single cation vacancies, but to complexes: \(\mathrm{Me}^{++}\)-vacancy.

If continuous displacement of halides into the cation vacancies of complexes occurs, then cation vacancies must also continuously enter the crystal from outside in order to restore the complexes. Such a process under the action of ionizing radiation is quite possible [12].

The process of vacancy diffusion from the crystal surface may also proceed along dislocations [13]. Indeed, in deformed-

the number of dislocations in such a sample is sharply increased, which facilitates the diffusion of cation vacancies. This promotes intensification of the linear stage of accumulation of \(F\)-centers in the KCl·Sr crystal [14].

Thus, the experiments carried out make it possible to confirm the proposition concerning the major role of the defectiveness of the cation sublattice in the processes of formation of color centers.

REFERENCES

1. U. H. Crawford, Jr., C. M. Nelson. Phys. Rev. Letters. 5, 314, 1960.
2. W. Hayes, C. M. Nichols. Phys. Rev. 117, 998, 1960.
3. E. E. Schneider. Diss. Farad. Soc. 31, 176, 1961.
4. I. Ya. Melik-Gaikazyan, V. M. Lisitsyn. FTT, 7, 1994, 1965.
5. E. Sonder, W. A. Sibley. Phys. Rev. 140, 2A, 539, 1965.
6. H. Jtoh, B. R. H. Royce, R. Smoluchowski. Phys. Rev. 138, A17, 66, 1965.
7. B. J. Faraday, W. D. Compton. Phys. Rev. 138, A893, 1965.
8. P. Mitchell, D. Wiegand, R. Smoluchowski. Phys. Rev. 121, 484, 1961. M. A. Elango. Tr. IFA AN ESSR, 17, 135, 1961.
9. J. E. Caffin, B. K. Ridley. Proc. Phys. Soc. 77, 153, 1961. V. V. Lemanov, M. Kornfeld. Zh. ETF, 4, 1, 1954, 1961.
10. H. Rabin, C. C. Klick. Phys. Rev. 117, 4, 1005, 1960.
11. I. Ya. Melik-Gaikazyan, E. K. Zavadovskaya, M. I. Ignat’eva. FTT, 6, 1243, 1964.
12. A. E. Kiv, F. T. Umarova. Proceedings of the First Conference on Radiation Physics, Kiev, 1965.
13. V. N. Rozhanskii, A. A. Predvoditelev. DAN SSSR, 158, 835, 1964.
14. L. I. Roshchina, V. P. Klyukvina. FTT, 7, 2262, 1965.