Investigation of Electrical Breakdown and Detonation Initiation in Thallium and Copper(II) Azides

D. A. Zakharov, D. N. Sukhushin

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

A comparatively small number of works has been devoted to the study of detonation initiation during the breakdown of explosives [1–3], although this process is of definite scientific interest [1, 4].

This paper reports some results of a study of the electrical strength of thallium azide and divalent copper azide under voltage pulses with an exposure time of about \(10^{-7}\) sec, with the aim of elucidating the mechanism of breakdown and its connection with the mechanism of detonation initiation in the indicated objects. The samples were tablets 0.2 mm thick and of density close to that of a single crystal, pressed onto a metallic electrode. The electrode geometry—“sphere against plane”—provided a field close to uniform, and placing the sample in a protective medium (silicone oil) excluded the possibility of ignition of the high explosive by the corona of the electrodes. The voltage and the current through the sample during breakdown were recorded oscillographically.

Essential for elucidating the mechanism of the electrical form of breakdown is the question of the source of nonequilibrium charge carriers which, multiplying by the mechanism of impact ionization, lead to a catastrophic increase in current and to destruction of the sample (in our case, to an explosion). Nonequilibrium charge carriers may appear as a result of electrostatic excitation of electrons (impurity electrons located-

…located at trapping levels, traps), and also as a result of injection of electrons from the cathode, or holes from the anode [5, 6, 7, 8, 9]. To determine the source of nonequilibrium carriers during breakdown of $\mathrm{TlN_3}$ and $\mathrm{Cu(N_3)_2}$, we investigated the effect of the material and polarity of the electrodes on their electrical strength.

The results are presented in Figs. 1–2 in the coordinates: cumulative breakdown frequency—electric-field strength. The results were processed by the method described in [10], using data obtained on 80–120 specimens. The material of the flat electrode was varied. It was found that, during breakdown of $\mathrm{Cu(N_3)_2}$, the material of the anode has an exceptional influence; moreover, a sequence of anode-material influence is established with increasing thermodynamic work function, in which the electrical strength increases $(\mathrm{Zn}-\mathrm{Cu}-\mathrm{W})$ [11]. During breakdown of $\mathrm{TlN_3}$, both the anode material and the cathode material have an influence (the latter somewhat stronger), with a sequence of metals opposite to that for $\mathrm{Cu(N_3)_2}$ being established.

The fact that the electrode material has an influence permits one to speak of injection as a source of nonequilibrium carriers during breakdown of $\mathrm{Cu(N_3)_2}$ and $\mathrm{TlN_3}$; for $\mathrm{Cu(N_3)_2}$ these are holes, and for $\mathrm{TlN_3}$ both holes and electrons (double injection [7, 9]). The change in electrical strength as a function of the metal work function cannot be explained by a difference in the height of the potential barrier for injection and extraction of electrons, since in that case different sequences should be observed for $\mathrm{TlN_3}$ for the cathode and the anode (according to the known concepts of the theory of contact phenomena [12]). The results obtained are naturally explained on the basis of the theory, developed by one of the authors, of transient processes in a metal–semiconductor–metal system, according to which, under certain conditions (the presence of impurity levels,

Because of the low mobility of the carriers, volume charges present at the metal–semiconductor contact even before the voltage is applied [12] play a major role in creating pre-breakdown conditions. In an external field, a nonuniform potential distribution arises; near the electrodes a region of increased field strength is formed, which is transformed in time at a finite rate determined by the mobility of the charges. In a number of cases, the initial nonuniform distribution creates conditions sufficient for carrier injection and subsequent impact ionization. Such conditions may also be created in the process of charge redistribution. In our case, because of the extremely short duration of the applied voltage and the low mobility of the carriers, the transformation processes of the volume charges may be neglected. Then, under the condition \(\varphi_{p/n} > \varphi_m\), the contact field strengths will have the form:

\[ E_k(t)= \frac{V_{\mathrm{ext}}(t)} {2\ell_D+(\ell-2\ell_D)\dfrac{\Delta \tilde{\varphi}\cdot \Delta(t)} {(t+1)\left(e^{\Delta \tilde{\varphi}}-1\right)}} - \frac{\Delta \varphi_k}{\ell_D\,\Delta(t)} \tag{1} \]

\[ E_a(t)= \frac{V_{\mathrm{ext}}(t)} {2\ell_D+(\ell-2\ell_D)\dfrac{\Delta \tilde{\varphi}\cdot \Delta(t)} {(t+1)\left(e^{\Delta \tilde{\varphi}}-1\right)}} + \frac{\Delta \varphi_k}{\ell_D\,\Delta(t)} \tag{2} \]

Here the sign “\(+\)” applies to a semiconductor with hole conductivity \((\mathrm{Cu}(\mathrm{N}_3)_2)\), and “\(-\)” to one with electronic conductivity \((\mathrm{TlN}_3)\). Since

\[ \Delta \tilde{\varphi}=\frac{\Delta \varphi_k}{kT}, \qquad \Delta \varphi_k=\varphi_n-\varphi_m, \]

the dependences obtained (Figs. 1, 2) are qualitatively described in detail by formulas (1) and (2) (the relative value of the electric strength, the sign of the influence of electrode polarity, and the series of metals).

In addition, the described concepts correspond to re-

The results of investigating the effect of temperature on the electrical strength of \( \mathrm{TlN_3} \) and \( \mathrm{Cu(N_3)_2} \) are an increase in electrical strength with temperature for \( \mathrm{Cu(N_3)_2} \), and a decrease for \( \mathrm{TlN_3} \).

For the purpose of directly determining experimentally the sign of the carriers during breakdown of \( \mathrm{TlN_3} \) and \( \mathrm{Cu(N_3)_2} \), experiments were carried out to study the Hall effect in the prebreakdown region under direct current. The results obtained confirm that breakdown in \( \mathrm{TlN_3} \) is effected by electrons, and in \( \mathrm{Cu(N_3)_2} \) by holes.

Thus, the following picture of the occurrence of breakdown in \( \mathrm{TlN_3} \) and \( \mathrm{Cu(N_3)_2} \) may be assumed: breakdown is effected in \( \mathrm{TlN_3} \) by electrons injected from the cathode, and in \( \mathrm{Cu(N_3)_2} \) by holes injected from the anode; both processes are governed by the nonuniform field distribution in the near-contact regions and develop into impact ionization. Schematically this can be written as follows:

hole injection:
\[ N_3^{-} \longrightarrow N_3^{0} + e \tag{3} \]

impact ionization by electrons:
\[ e + N_3^{-} \longrightarrow N_3^{0} + 2e \tag{4} \]

impact ionization by holes:
\[ N_3^{0} + N_3^{-} \longrightarrow 3N_2^{0} + e \tag{5} \]

The further fate of the holes is analogous to the processes of thermal decomposition, photolysis, and radiolysis of heavy-metal azides:

\[ N_3^{0} + N_3^{0} \longrightarrow 3N_2^{0} + h\nu \tag{6} \]

Turning to the mechanism of detonation excitation during breakdown of \( \mathrm{TlN_3} \) and \( \mathrm{Cu(N_3)_2} \), three possibilities should be noted:

1). Excitation due to thermal phenomena in the discharge channel;

2). Excitation due to the action of the shock wave in the discharge channel;

3). Excitation as the final stage in the development of breakdown, since the breakdown mechanism described by equations (3)–(6) coincides with the mechanism of excitation of the chemical reaction during detonation in heavy-metal azides, described in [1].

Special experiments have shown that for copper(II) azide mechanism III can operate, since we did not find such a minimum discharge energy that would produce breakdown in \(Cu(N_3)_2\) without initiating detonation, up to capacitances equal to the sample capacitance (1–2 pF). Cases of breakdown of \(TlN_3\) without detonation were observed; however, the character of the current oscillogram during breakdown shows that detonation initiation occurs not as a result of the action of heat in the discharge channel (the absence of an arc stage of the discharge), but as a result of the action of the shock wave.

Thus, the phenomena studied during breakdown of thallium azide and divalent copper azide make it possible to specify the electronic mechanism of breakdown of these objects, and also to draw a conclusion about the mechanism of detonation initiation during breakdown.

LITERATURE

1. F. Bowden, A. Yoffe. Fast Reactions in Solids. Izd. IL, 1963.

2. Nakano J., Tanaka J., Kusakabe M., J. of Ind. Expl. Soc. Japan, 24, No. 6, 45 (1963); 24, No. 6, 351 (1963); 26, No. 5, 265 (1965).

3. Schall R. C., 232, 706 (1951).

4. J. N. Maycock, Science, 152, No. 3721, 508 (1966).

5. Franz V., Breakdown of Dielectrics. Izd. IL, M., 1963.

6. P. H. Parmenter, W. Ruppel. J. Appl. Phys., 30, 1548 (1959).

7. M. A. Lampert, RCA Rev., 20, 682 (1959).

8. G. A. Marlor, J. Woods, Proc. Phys. Soc., 81, 1013 (1963).

9. A. Rose, J. Appl. Phys., 35, 2664 (1964).

10. V. V. Smirnov, L. V. Dunin-Barkovsky, A Course in Probability Theory and Mathematical Statistics. Nauka, M., 1965.

11. Chemist’s Handbook, Vol. I, p. 333, Goskhimizdat, M.–L., 1963.

12. A. I. Gusanov, Theory of the Rectifying Action of Semiconductors. GITTL, M., 1956.

Fig. 1. Effect of electrode material and polarity on the breakdown of \(\mathrm{Cu(N_3)_2}\).

Fig. 2. Effect of electrode material and polarity on the breakdown of \(\mathrm{TlN_3}\).