POST-RADIATION CHANGES IN CONDUCTIVITY IN IRRADIATED AMMONIUM PERCHLORATE

Yu. A. Zakharov, G. T. Shchekov

(Presented by the scientific seminar of the Department of Radiation Chemistry)

It is known that ammonium perchlorate is one of the most widely used oxidizers [1]. The areas of application of ammonium perchlorate require a comprehensive study of the action of various factors that affect its thermal and radiation stability.

The thermal decomposition of $\mathrm{NH_4ClO_4}$ is significantly affected by the composition and pressure of the gaseous atmosphere, as well as by the introduction of various donor–acceptor impurities into the ammonium perchlorate lattice [3].

The influence of preliminary irradiation of $\mathrm{NH_4ClO_4}$ with $\gamma$- and X-rays on thermal decomposition is described in works [2, 3, 4]. The proposed schemes for the thermal and radiation decomposition of pure and doped ammonium perchlorate are based on the premise of the important role of electronic defects in the development of the processes. At the same time, the literature contains no results on the study of the electrophysical properties of ammonium perchlorate, with the exception, perhaps, of the single work [5]. In this connection, the study of the electrophysical properties of $\mathrm{NH_4ClO_4}$ is of undoubted interest; by analogy with ionic compounds, these properties should be highly sensitive to changes in the defectiveness of the crystal structure upon introduction of ion-valent impurities into the lattice of the objects under investigation [18, 12] or upon their irradiation with various types of ionizing radiation [10, 11].

Knowledge of the mechanism of electrical conductivity and correlation of conductivity data with other electrophysical characteristics, as well as with experimental data on thermolysis and radiolysis, makes it possible in a number of cases to draw more definite conclusions about the processes of thermal decomposition [6, 7] and annealing of radiation defects [8, 10].

For this reason, we investigated changes in the electrical conductivity of $\mathrm{NH_4ClO_4}$ after irradiation of the preparations with $\gamma$-rays from $\mathrm{Co}^{60}$ and with protons of energy $4.5$ MeV. In formulating the present work, we were interested above all in the following questions.

1. Determination of the minimum doses of ionizing radiation that cause a post-radiation change in conductivity.

2. Clarification of the specificity of the action of different types of ionizing radiation on the electrical conductivity of $\mathrm{NH_4ClO_4}$.

3. Correlation of the obtained data with results on the thermal decomposition and radiolysis of ammonium perchlorate.

In addition, the thermoelectromotive force and dielectric losses in irradiated and unirradiated $\mathrm{NH_4ClO_4}$ were measured.

Irradiation of the Samples

The NH₄ClO₄ preparations used by us were irradiated with Co⁶⁰ γ-rays in an installation with an activity of 16,000 g-eq Ra. Irradiation was carried out at room temperature in air. The dose rate was 1,600 r/sec. The preparations were irradiated with doses of 0.25, 0.40, 0.80, 1.20, 2.40, 7.70, and 16.80 · 10²⁰ eV/g.

Irradiation of the samples with X-rays was carried out at \(t = 18^\circ\mathrm{C}\) in air on an RUP-200 apparatus operating at 200 kV and 20 mA. The absorbed dose of X-radiation was determined with the aid of a ferrous-sulfate dosimeter. Conversion of the dose absorbed in NH₄ClO₄ was performed using the mass coefficient of true absorption, equal to 0.145 for ammonium perchlorate. We also used NH₄ClO₄ preparations irradiated with protons with an energy of 4.5 MeV. Irradiation was carried out in air at \(t = 18^\circ\mathrm{C}\) on the cyclotron of the Research Institute of Nuclear Physics at the Tomsk Polytechnic Institute. The tablets were irradiated with protons at three dose rates: \(9.8 \cdot 10^{20}\) eV/g min; \(0.490 \cdot 10^{20}\) eV/g min; \(0.196 \cdot 10^{20}\) eV/g min.

The absorbed dose of proton radiation in N₄ClO₄ was calculated analytically under the assumption that, in the range of proton energies not exceeding several MeV, the principal type of proton energy loss is loss in elastic collisions with bound electrons of the atoms of the medium (ionization losses).

Experimental Part

Measurement of Electrical Conductivity

In this work, preparations of pure ammonium perchlorate without additional recrystallization were used. For the measurements, standard tablets 16 mm in diameter and 0.45 or 0.66 mm thick were pressed.

Pressing was carried out in a special die with polished and strictly parallel surfaces at a pressure of 4.5 t/cm².

The electrical conductivity of the irradiated tablets was measured in a glass cell with a quartz window, in an apparatus described earlier in [9]. Gas evolution in the system was recorded simultaneously under its constant, uninterrupted evacuation. The use of a combination of an RVN-20 forevacuum pump and an SDN-1 (glass diffusion oil pump) made it possible to obtain a vacuum in the apparatus no worse than \(10^{-4}\) mm Hg.

Heating of the sample was carried out at a constant rate by means of a transformer equipped with a “Uorren” motor with a reducer. The heating rate could be varied; in our experiments it was ~2 deg/min. To obtain reproducible results, before being placed in the cell the samples were conditioned for 2–1.5 hours at \(t = 25^\circ\mathrm{C}\) in a vacuum of \(10^{-1}\) mm Hg, and then in the measuring cell for 3 hours at a pressure not exceeding \(10^{-4}\) mm Hg.

The resistance of the samples in the temperature range 20–170°C was measured with an E6-3 direct-current amplifier at a field strength of 70 V/cm. The use in the measuring circuit of high-quality coaxial cables with double shielding, exclusively in Teflon or polyethylene insulation, made it possible to reduce leakage currents to immeasurably small values. All leads and connections were also shielded, and the electric resistance furnace in which the sample was heated was grounded. Al-, Cu-, and Pt-electrodes were used in the work. Graphite contacts were applied to the ends of the tablet, obtained by evaporating an alcoholic suspension of graphite.

Measurement of Dielectric Losses

Measurement of dielectric losses in the temperature range 20–130°C in a vacuum of \(10^{-2}\) mm Hg was carried out using an unbalanced bridge [17].

Preliminary experiments showed that replacing the contacts obtained by evaporating an alcoholic suspension of graphite with electrodes made of polished nickel plates gives minimum values of the tangent of the dielectric loss angle. Therefore, in subsequent measurements we used only nickel plates without graphite contacts as electrodes.

Measurement of Thermoelectromotive Force

Measurement of thermoelectric emf was carried out in air in the temperature range 20–250°C. The apparatus was fully automated. Measurements of thermoelectric emf were made with a U1-2 type amplifier and recorded on an EPP-09. During the experiment the temperature of the cold electrode \(t_1\), the hot electrode \(t_2\), and the temperature difference between them \((\Delta t)\) were measured by means of a differential copper-constantan thermocouple; these readings were also recorded on an EPP-09. The heating rate was regulated in the same way as in the measurement of electrical conductivity, and was equal to \(\sim 2\) deg/min. The temperature difference \((\Delta t)\) between the cold and hot electrodes varied in different experiments from 12 to 20°C. The measurements were carried out on tablets whose conductivity had been measured previously.

Experimental Results

The temperature dependence of the logarithm of the electrical conductivity of \(\mathrm{NH_4ClO_4}\) irradiated with protons and \(\gamma\)-rays of \(\mathrm{Co}^{60}\) is shown in Fig. 1. In Fig. 1 one clearly observes an increase in the logarithm of conductivity with increasing absorbed dose of proton radiation (curves 1–3) and the simultaneous appearance of a bend region on the curve of the temperature dependence of conductivity.

In the course of the experiments it was also established that the appearance of a bend on the curve of the temperature dependence of the electrical conductivity of irradiated \(\mathrm{NH_4ClO_4}\) occurs simultaneously with the onset of intense gas evolution from the sample (Fig. 1, curves 5 and 6). The activation energies of electrical conductivity, calculated from the slope of the curves up to the first bend, are identical and equal to 0.85 eV. The activation energies from the temperature dependence of electrical conductivity obtained upon cooling the samples are also identical and equal to 1.1 eV, which corresponds to the activation energy of unirradiated \(\mathrm{NH_4ClO_4}\) and coincides with the value of the activation energy of electrical conductivity in work [5].

It should be noted that, at smaller absorbed doses of proton and \(\gamma\)-radiation of \(\mathrm{Co}^{60}\), three regions are not observed on the curve of the temperature dependence of the electrical conductivity of irradiated ammonium perchlorate.

The activation energy of the electrical conductivity of \(\mathrm{NH_4ClO_4}\) irradiated with \(\gamma\)-rays of \(\mathrm{Co}^{60}\), calculated from the slope of the curve obtained upon cooling the sample, coincides with the activation energy for proton-irradiated material and is equal to 1.1 eV. The dependences of the logarithm of the electrical conductivity of ammonium perchlorate on the dose of proton and \(\gamma\)-radiation of \(\mathrm{Co}^{60}\) at temperatures of 73 and 157°C are presented in Figs. 2, 3, and 4. The magnitude of the electrical conductivity is noticeably affected by the dose rate of proton radiation.

of irradiation (Fig. 2, curves 1 and 4). Unfortunately, we were unable to study the dependence of $\lg \sigma$ on dose for specimens irradiated with protons in the dose interval $4.9 \cdot 10^{20}$—$9.8 \cdot 10^{20}$ eV/g at a dose rate of $0.1 \cdot 96 \cdot 10^{20}$ eV/g min, owing to the destruction of the pellets at the beginning of the experiment. It should be noted that the form of the curves for the electrical conductivity of $\mathrm{NH_4ClO_4}$ irradiated with protons and with $\gamma$-rays from $\mathrm{Co}^{60}$ is the same. However, the minimum values

Fig. 1. Temperature dependence of the electrical conductivity of $\mathrm{NH_4ClO_4}$ irradiated with protons at a dose rate of $0.49 \cdot 10^{20}$ eV/g min.
Curves 1, 2, 3 were obtained on heating $\mathrm{NH_4ClO_4}$ irradiated with protons; the doses are respectively $3.43$; $4.9$; $9.8 \cdot 10^{20}$ eV/g.
Curve 4 was obtained on cooling $\mathrm{NH_4ClO_4}$ irradiated with protons; dose $9.8 \cdot 10^{20}$ eV/g.
5, 6—gas-evolution curves on heating and cooling the specimen (corresponds to curve 2).

of conductivity in $\mathrm{NH_4ClO_4}$ irradiated with $\gamma$-rays from $\mathrm{Co}^{60}$ are observed at higher absorbed doses than in perchlorate irradiated with protons. This is probably associated with the higher dose rates in the case of irradiation of the preparations with protons. The minimum values of the electrical conductivity of irradiated $\mathrm{NH_4ClO_4}$ at temperatures of 73 and 157°C are approximately equal and are an order of magnitude lower than the electrical conductivity in unirradiated perchlorate, although one would expect a relatively greater change in conductivity at low temperatures than at high temperatures. The curves shown in Figs. 2, 3, and 4, obtained on cooling the specimens, indicate the relative stability of the disturbances induced by

by irradiation. The dependence of the electrical conductivity for samples irradiated with protons at a dose rate of \(9.8 \cdot 10^{20}\ \text{eV/g min}\), after annealing of defects for four months under natural conditions \(t = 20^\circ\text{C}\), \(P = 760\ \text{mm Hg}\), is shown by curve 2 in Fig. 2 and also indicates the stability of the disturbances introduced by radiation.

In the course of the experiments it was established that the electrode material (we used Al, Cu, Pt) does not exert a noticeable influence on the magnitude of the conductivity.

The measurements of dielectric losses that were carried out showed the presence of a maximum in the dependence of the tangent of the dielectric loss angle on the dose of proton irradiation at a dose of \(9.8 \cdot 10^{20}\ \text{eV/g}\) (Fig. 5). It is interesting to note that, at this same dose of proton irradiation, a maximum of conductivity is also observed in the irradiated perchlorate. To determine the sign of the current carriers in irradiated and unirradiated perchlorate, we measured the thermoe.m.f. The results obtained are presented in Fig. 6. The temperature difference (\(\Delta t^\circ\text{C}\)) between the cold and hot electrodes during the experiments was \(\sim 16^\circ\text{C}\). In the measurements, a negative sign of the current carriers was observed both in irradiated and in unirradiated samples.

Fig. 2. Dependence of the electrical conductivity of \(\mathrm{NH_4ClO_4}\) on the dose of proton and \(\gamma\)-irradiation at \(t = 73^\circ\text{C}\).

Curves 1, 2, 3 — \(\mathrm{NH_4ClO_4}\) irradiated with protons; dose rate — \(9.8 \cdot 10^{20}\ \text{eV/g min}\).

Curve 4 — \(\mathrm{NH_4ClO_4}\) irradiated with protons; dose rate — \(0.49 \cdot 10^{20}\ \text{eV/g min}\).

Curve 5 — \(\mathrm{NH_4ClO_4}\) irradiated with \(\gamma\)-rays of \(\mathrm{Co}^{60}\); dose rate — \(0.055 \cdot 10^{19}\ \text{eV/g min}\).

Curves 1, 4 were taken immediately after irradiation.

Curves 2, 3 were taken after 120 days of annealing.

Curve 5 was obtained after 15 days of annealing.

Curves 1, 2, 4, 5 were obtained upon heating of the samples; curve 3 — upon cooling.

During the measurement of the electrical conductivity of irradiated and unirradiated \(\mathrm{NH_4ClO_4}\), experiments were performed to determine the photocurrent in perchlorate at various temperatures when the tablet was illuminated with light of \(\lambda \simeq 2537\ \text{\AA}\). The absence of photocurrent in \(\mathrm{NH_4ClO_4}\) established in these experiments may be connected, in particular, with an insufficient energy of the light quanta to excite electrons into the conduction band of \(\mathrm{NH_4ClO_4}\), as was established for alkali-metal azides in work [16].

Discussion of the Results

In the experiments carried out by us, a high sensitivity of the electrical conductivity of \(\mathrm{NH_4ClO_4}\) to the action of various types of ionizing radiation was established. The minimum absorbed dose of \(\gamma\)-irradiation from \(\mathrm{Co}^{60}\) selected in the present work, equal to \(0.250 \cdot 10^{20}\ \text{eV/g}\), produces approximately a tenfold change in the electrical conductivity of ammonium perchlorate (Figs. 2–3). At the same time, the minimum absorbed dose of proton irradiation, equal to \(0.250 \cdot 10^{20}\ \text{eV/g}\), produces significantly smaller post-radiation changes in electrical conductiv-

Fig. 3. Dependence of the electrical conductivity of \( \mathrm{NH_4ClO_4} \), irradiated with protons (curves 1, 3) and \(\gamma\)-rays of \( \mathrm{Co}^{60} \) (curve 2), on dose at \(t = 73^\circ\mathrm{C}\).

Curves 1 and 3 were obtained for samples irradiated at a dose rate of \(0.49 \cdot 10^{20}\ \mathrm{eV}/\mathrm{g\,min}\). Curve 2 was obtained for samples irradiated at a dose rate of \(0.665 \cdot 10^{19}\ \mathrm{eV}/\mathrm{g\,min}\); \(x\!-\!x\!-\!x\) indicates samples irradiated with protons at a dose rate of \(0.196 \cdot 10^{20}\ \mathrm{eV}/\mathrm{g\,min}\). Curves 1 and 2 were obtained upon heating the samples; curve 3, upon cooling.

Fig. 4. Dependence of the electrical conductivity of \( \mathrm{NH_4ClO_4} \), irradiated with protons (curves 1, 3) and \(\gamma\)-rays (curve 2), on dose at \(157^\circ\mathrm{C}\). The remaining notation is the same as in Fig. 3.

Fig. 5. Dependence of the dielectric-loss tangent on dose for \( \mathrm{NH_4ClO_4} \) irradiated with protons. Dose rate: \(9.8 \cdot 10^{20}\) eV/g·min.
The curves were obtained at frequencies: 1—\(f = 100\) Hz; 2—\(f = 500\) Hz; 3—\(f = 1000\) Hz; 4—\(f = 50\) Hz; 5—\(f = 5000\) Hz.

Fig. 6. Thermoelectromotive force in \( \mathrm{NH_4ClO_4} \) irradiated with \( \gamma \)-rays from \( \mathrm{Co}^{60} \) (curve 2, dose \(0.25 \cdot 10^{20}\) eV/g), and in unirradiated \( \mathrm{NH_4ClO_4} \) (curve 1).
The heating rates of the cold electrode for unirradiated and irradiated \( \mathrm{NH_4ClO_4} \) are 15 and 6 deg/min, respectively (curves 3 and 4).

of \(NH_4ClO_4\). We shall discuss the possible reasons for such a difference below.

When solid compounds are irradiated with \(Co^{60}\) \(\gamma\)-rays, X-rays, and protons with an energy of \(4.5\) MeV, the radiation mainly causes only excitation of electrons bound to the ions—elements of the crystal lattice—up to the formation of free electrons. Various structural defects are thereby formed. Displacements of atoms of the initial substance under the action of the above-mentioned types of radiation are relatively unlikely. On this basis one may suppose that the dependence of the electrical conductivity of the object under study on dose, when irradiated with different types of ionizing radiation, should have a qualitatively similar form. As is seen from Figs. 2, 3, and 4, the shape of the curves \(\lg \sigma\)—dose for ammonium perchlorate irradiated with protons and with \(Co^{60}\) \(\gamma\)-rays is the same. When solids are irradiated with neutral particles (neutrons), neutrons interact with atomic nuclei, since neutrons do not interact with electrons and are not affected by the Coulomb forces of the nucleus. Therefore one may expect more significant quantitative and qualitative changes in the electrical conductivity of \(NH_4ClO_4\) after its irradiation with neutrons.

Under the action of radiation on the object under study, essentially two types of processes may occur:

1 — generation of current carriers in the specimen and, as a consequence, an increase in electrical conductivity;

2 — formation of free electrons and holes, their localization at microdefects or in vacant lattice sites, and the resulting decrease in electrical conductivity.

The sign of the effect of the change in electrical conductivity after irradiation and the magnitude of the change are determined by the predominance of one process over the other. It is not possible to predict in advance in which direction the magnitude of the electrical conductivity will change after irradiation without knowing the mechanism of electrical conductivity.

The decrease in conductivity in our experiments in the region of small absorbed doses of proton and \(Co^{60}\) \(\gamma\)-irradiation can tentatively be explained by taking into account Seitz’s ideas [14], proposed for alkali halides, if one accepts, following Freeman and Tsirkin [5], that the conductivity in \(NH_4ClO_4\) is more likely effected by cation vacancies. According to Seitz, electrical conductivity should decrease upon irradiation of crystals due to localization of holes (in our case, radicals having an effective positive charge), which may be the formations \(ClO_2^{\circ}\), \(ClO_3^{\circ}\), etc., on single carriers—vacancies of the positive ion.

The considerable polarization observed by us—the decay of current with time after application of a constant voltage—for specimens irradiated with protons and \(Co^{60}\) \(\gamma\)-rays in the dose range \(0.25 \div 2 \cdot 10^{20}\) eV/g, gives grounds for assuming the formation of complexes which, becoming oriented in the field, reduce the initial value of the current and may constitute a vacancy–hole (radical) pair.

With increasing temperature, the electrical conductivity in this dose interval should decrease substantially, as shown for \(NH_4ClO_4\) irradiated with \(Co^{60}\) \(\gamma\)-rays in Figs. 3 and 4, owing to the increased probability of interaction between holes and vacancies [13].

In order to elucidate the mechanism of conductivity of unirradiated \(NH_4ClO_4\), Freeman and Tsirkin [5] investigated the electrical conductivity of ammonium perchlorate in an atmosphere of argon and oxygen.

The significant effect of oxygen and argon on the value of the electrical conductivity of NH\(_4\)ClO\(_4\) in their experiments suggests that the radiolysis products formed during irradiation of NH\(_4\)ClO\(_4\), among which there are very effective oxidizing agents, will also affect the post-radiation change in electrical conductivity.

We compared the data on the change in electrical conductivity of NH\(_4\)ClO\(_4\) irradiated with protons and Co\(^{60}\) \(\gamma\)-rays with the results of analysis of the radiolysis products of ammonium perchlorate obtained in our laboratory [18].

When NH\(_4\)ClO\(_4\) is irradiated with Co\(^{60}\) \(\gamma\)-rays, the yields of the radiolysis products Cl\(^{-}\), ClO\(_3^{-}\), ClO\(_2^{-}\), and others considerably exceed the yields of these same products of ammonium perchlorate irradiated with protons. Moreover, with increasing dose rate the yield of the ClO\(_3^{-}\) ion decreases, while the conductivity of perchlorate, as shown in Fig. 2, increases.

Taking the above into account, it becomes possible to relate the increase in the conductivity of NH\(_4\)ClO\(_4\) also to the accumulation in the irradiated ammonium perchlorate of the decomposition product—chlorate—and thereby to explain the somewhat greater electrical conductivity of NH\(_4\)ClO\(_4\) irradiated with protons in comparison with the electrical conductivity of NH\(_4\)ClO\(_4\) irradiated with Co\(^{60}\) \(\gamma\)-rays.

The subsequent increase in electrical conductivity, in our view, is a consequence of the dissociation of a cation-vacancy—hole pair under the action of radiation as the absorbed dose increases. The considerable decrease in polarization in this region confirms the destruction of cation-vacancy—hole complexes. The holes then migrate to the surface of the crystal or into strongly disturbed regions, where, interacting with one another, they give a gaseous reaction product. This process must be accompanied by the accumulation of gaseous products in the bulk of the substance and by intense gas evolution in the system when the sample is heated, by cracking of the crystals, and by the presence of a break in the electrical-conductivity curve, as occurs for solid solutions when one of the components separates into a separate phase [12] after irradiation of the system with ionizing radiation. In our experiments, for these irradiation doses, a considerable decrease in the mechanical strength of the pellets was observed, and at temperatures of 120°C—their destruction.

Figure 1 presents typical curves constructed for NH\(_4\)ClO\(_4\) irradiated in the dose range \(1.2 \div 9.8 \times 10^{20}\) eV/g. The difference in the activation energies of electrical conductivity, calculated from the slope of the curves before the bend, equal to 0.85 eV, and after the bend, 1.1 eV, corresponds to a different concentration of current carriers [15].

The maxima of dielectric losses (Fig. 6), observed at different frequencies, coincide with the maximum of electrical conductivity for samples irradiated with a dose of \(9.8 \cdot 10^{20}\) eV/g. The losses in this case are probably due to losses arising from conductivity. The decrease in conductivity after the maximum at an absorbed dose of \(\simeq 15 \cdot 10^{20}\) eV/g can probably be attributed to the destruction of crystals and the radiolytic evolution of gases.

On the basis of our measurements of the thermoelectromotive force, which showed a negative sign of the current carriers (Fig. 6), and the absence of photocurrent in irradiated and unirradiated ammonium perchlorate, it may be assumed that NH\(_4\)ClO\(_4\) is a Schottky-defective material with predominantly mobile cation vacancies. However, to definitively clarify the nature of the current carriers in NH\(_4\)ClO\(_4\), it is necessary to carry out experiments, in particular with preparations doped with divalent cations, and to compare the data thereby obtained

with the results available in the literature on the thermal decomposition of NH₄ClO₄ doped with the same donor–acceptor impurities. In addition, determining the composition and concentration of the gaseous products evolved during heating of the samples in the temperature range corresponding to the break in the temperature-dependence curve of electrical conductivity would, in our view, help bring us closer to resolving the question of the nature of the current carriers in ammonium perchlorate.

Conclusions

1. The high sensitivity of the electrical conductivity of NH₄ClO₄ to the action of various types of ionizing radiation has been demonstrated at doses of \(0.065 \div 9.8 \cdot 10^{20}\) eV/g.

2. The dependence of electrical conductivity on dose rate has been established.

3. It has been shown that the sign of the current carriers in ammonium perchlorate is negative.

4. The nature of the disorder in NH₄ClO₄ is discussed.

References

1. A. I. Silyant’ev. Solid Rocket Propellants. Military Publishing House of the Ministry of Defense of the USSR, Moscow, 1964.

2. Yu. A. Zakharov, G. G. Savel’ev. Kinetics and Catalysis, 8, 1, 1966.

3. F. S. Freeman, D. A. Anderson, I. J. Campise. J. Phys. Chem., 64, 11, 1727, 1960.

4. I. C. Petriciani, S. E. Wibelay, W. H. Bauer, T. W. Ceapper. J. Phys. Chem., 64, 1309, 1960.

5. F. S. Freeman. Zirking, Nature, 199, 4900, 1281, 1963.

6. Yu. A. Zakharov, A. A. Kabanov. ZhFKh, 38, 2882, 1964.

7. Yu. A. Zakharov, G. G. Savel’ev, V. K. Zhuravlev, V. V. Boldyrev. Kinetics and Catalysis, 6, 415, 1965.

8. S. B. Mohanty. I. Sci. Res. Banar. Lindn. Univ., 12, No. 2, 1961–1962.

9. G. G. Savel’ev, Yu. A. Zakharov, G. T. Shechkov. ZhFKh, 41, 1691, 1967.

10. M. I. Ignat’eva. Candidate dissertation, Tomsk, 1964.

11. Interuniversity Collection of Proceedings. Issue 2. Effect of Radiation on Materials, Novosibirsk, 1963.

12. A. Lidiard. Ionic Conductivity of Crystals. Foreign Literature Publishing House, Moscow, 1962.

13. K. K. Aglintsev. Dosimetry of Ionizing Radiations, Moscow, Gostekhizdat, 1957.

14. F. Seitz. Rev. Mod. Phys., 26, 7, 1954.

15. R. W. Dreyfus, A. S. Nowick. Phys. Rev., 126, 1367, 1962.

16. S. K. Deb. Trans. Faraday Soc., 59, No. 486, 1415, 1963.

17. B. I. Vorozhtsov. PTE, No. 1, 116, 1959.

18. V. A. Nevostruev, G. N. Bazhenova, Yu. A. Zakharov. Proceedings of the Conference of Young Scientists of Tomsk, 1966, Tomsk State University Press, 1970 (in press).