ON THE INFLUENCE OF NEAR-CATHODE SPACE CHARGE ON THE ELECTRIC STRENGTH OF SOLID DIELECTRICS

A. V. BARANOV, M. N. LISETSKAYA, I. S. PIKALOVA

(Presented by the scientific seminar of the Department of High-Voltage Engineering)

In the multi-avalanche-streamer mechanism of discharge formation in solid dielectrics, before the inception of a streamer at the anode leading to breakdown, several electron avalanches arrive in a small region of the anode [1, 2]. In this case the electric strength is inversely proportional to the emission-current density from the cathode [2], which is confirmed by experiments on the influence of ultraviolet irradiation of the cathode and of the cathode material on the electric strength of rock salt. Consequently, by reducing emission from the cathode, one can achieve an increase in electric strength.

In studying the pulsed electrical conductivity of alkali-halide crystals, regularities were obtained indicating the formation of a negative space charge in the cathode region due to the settling of electrons in traps [4], which limits the emission current from the cathode. The creation of a near-cathode space charge has been considered in a number of other works [5, 6]. In particular, [6] points to the good stability of the near-cathode space charge.

In the present work we present data on measurements of the electric strength of alkali-halide crystals that had previously been subjected to the action of a strong electric field.

Experiment

The objects of investigation were single crystals of rock salt (with thicknesses of 10 and 60 μm) and KBr (with a thickness of 10 μm). The specimens were prepared according to the method of [7]. As electrode material, a saturated solution of the corresponding salt (electrolyte —) and graphite were used.

Several rectangular pulses were first applied to the dielectric specimen under study, with a front duration of \(5 \cdot 10^{-6}\) sec and an amplitude 30% below the breakdown value. The amplitude was then increased, and the specimen broke down on the pulse front.

Figures 1–3 show the dependences of the probability of dielectric breakdown on the field strength, \(\varphi = f(E)\). From these curves it is seen that, when a small number of voltage pulses are preliminarily applied to the dielectric specimen under study, its electric strength increases. When graphite electrodes are used, the electric strength of rock salt (\(d = 60\) μm) upon application of 2–3 preliminary

pulses increases by 26%, at a 90% breakdown probability (Fig. 1, curve 1—2), in comparison with the electric strength of the same specimens but without prior application of voltage pulses (Fig. 1, curve 1—1).

When 10 preliminary pulses were applied, \(E\) did not change. Curve II—1, 2 in Fig. 1 shows analogous dependences for rock salt \(10\ \mu\) thick with graphite electrodes.

When 2–3 preliminary pulses are applied, at a 90% breakdown probability, the strengthening is 32%, which indicates an enhancement of this effect as the thickness decreases.

Fig. 1. Dependence of the breakdown field strength of rock salt on the breakdown probability with graphite electrodes: I — thickness \(d = 60\ \mu\); II — thickness \(d = 10\ \mu\), 1 — without preliminary voltage exposure, 2 — with preliminary voltage exposure, 3 — with preliminary exposure to pulses and weak illumination by ultraviolet light, 4 — with preliminary exposure to pulses and strong illumination, 5 — influence of a change in pulse polarity.

Fig. 2. Dependence of the breakdown field strength of KBr \(10\ \mu\) thick on the breakdown probability with electrolyte electrodes: 1 — without preliminary exposure to pulse voltage, 2 — with preliminary voltage exposure.

Fig. 3. Dependence of the breakdown field strength of rock salt \(10\ \mu\) thick on the breakdown probability with liquid electrodes: 1 — without preliminary voltage exposure, 2 — with preliminary voltage exposure.

For KBr crystals \(10\ \mu\) thick with electrolyte electrodes (Fig. 2), the strengthening is 30% at a 90% breakdown probability.

In the study of the electric strength of rock salt with liquid electrodes after the preliminary application of voltage, no strengthening was detected (Fig. 3).

Studies were carried out on the influence of irradiation with ultraviolet light on the magnitude of the strengthening.

For this purpose, rock-salt specimens with graphite electrodes, after preliminary exposure to voltage, were irradiated with a spark discharge lamp and then punctured.

It was found that the strengthening effect of the dielectric is removed under prolonged irradiation (20 pulses of duration 20 μsec, Fig. 1, curve II—4). Under short-term irradiation by 2–5 illuminating pulses there is only a slight decrease in the electric strength (Fig. 1, curve II—3).

No strengthening effect was found in rock-salt specimens with graphite electrodes if, after the preliminary action of a pulsed voltage, a pulse of opposite polarity was applied to the specimen under study (Fig. 1, curve I—5).

Discussion of the results

When graphite is used as the cathode material, the emission of electrons caused by the acting field \(E\) increases owing to irregularities on the emitting surface as a result of the local increase in the field. In addition, it has been found experimentally [3] that the electric strength of a solid dielectric with graphite electrodes is lower than when liquid electrodes are used. In this case it may be assumed that electron emission from the cathode begins at a weaker average field \(\bar{E}_{\text{av}}\) at the anode under the multichannel streamer mechanism of breakdown. The emitted electrons partly settle in traps and form a negative space charge in the near-cathode region, which screens the emission from irregularities of the cathode surface.

This impedes the accumulation of charges at the anode and the inception of a streamer after the preliminary action of the voltage. In order for breakdown to occur in this case, an increase in the field strength is required.

Under the action of ultraviolet irradiation, the electrons are partly released from the traps and partly scattered. As a result, the screening action of the negative space charge is reduced and the strengthening effect is somewhat diminished. Under prolonged irradiation, the negative charge near the emission points on the cathode is scattered to a considerable extent and the strengthening is completely removed.

Under prolonged exposure to voltage (a large number of preliminary pulses), no strengthening effect is observed. This is apparently connected with the fact that the density of the negative space charge formed is very large and the possibility arises of releasing electrons from traps owing to the field \(E\), which leads to a decrease in the electric strength.

When liquid electrodes are used, as measurements of electrical conductivity show [4], a negative space charge is also created near the cathode. It appears that emission from a liquid cathode occurs owing to the presence in the crystal, near the cathode, of negative vacancies. Electrons emerging from the cathode partially settle in negative vacancies, which leads to a decrease in the emission current. Preliminary voltage pulses may lead to the formation of a negative space charge at the cathode through the capture of electrons by vacancies, which may bring about an increase in the electric strength.

In the investigation of pulsed electrical conductivity, significant electron currents were detected in KBr and KCl crystals.

In rock salt, when liquid electrodes were used, no currents were detected within the sensitivity limits of the circuit \(\left(5 \cdot 10^{-8}\ a/\text{mm}\right.\) of oscilloscope screen). It therefore becomes clear why hardening is not observed in rock salt after preliminary application of voltage.

Conclusions

1. An increase in electrical strength has been found in crystals of rock salt and KBr after preliminary exposure to pulsed voltage.

2. The hardening effect is removed after irradiation of the dielectric with ultraviolet light.

3. The hardening of a solid dielectric is explained by the formation of space charge in the near-cathode region.

Literature

1. G. A. Vorob'ev. Publ. Acad. Sci. USSR, Dept. of Technical Sciences. Energy and Transport, No. 2, 1963.

2. A. A. Vorob'ev, G. A. Vorob'ev. Radio Engineering and Electronics, vol. VII, issue 9, 1962.

3. G. A. Vorob'ev, T. Kocherbaev. Radio Engineering and Electronics, vol. 10, issue 3, p. 397, 1964.

4. A. V. Baranov. Collection of papers of the interuniversity conference on breakdown of dielectrics and semiconductors, GEI, 1964.

5. I. M. Yashukova. Proceedings of Higher Educational Institutions—Physics, No. 5, 18, 1962.

6. Ya. N. Pershits. Scientific Notes of the Pskov State Pedagogical Institute, issue III, 1955.

7. G. A. Vorob'ev, V. A. Kostrygin, L. T. Murashko, PTE, No. 5, 1961.