About the Formation of a Barrier Discharge in a KrCl Excilamp

Abstract

the space discharge was observed. With increasing pulse repetition frequency, diffusion channels of small diameter were formed, and at pulse repetition frequencies of ~1 kHz and higher, the discharge represented diffusion microdischarges shaped as two cones with joint apexes. The number of such microdischarges increased when the excitation power increases, for example, due to increased pulse repetition frequency, and they occupied the entire excilamp volume (cone bases covered the entire surface of quartz tubes). In this case, the maximum radiation power was obtained for the discharge consisting of diffusion microdischarges, and the maximum radiation efficiency was recorded at intermediate pulse repetition frequencies (~1 kHz), when the first diffusion cone-shaped microdischarges were formed. The present work studies the dynamics of the formation of the stationary barrier discharge stage for which the average radiation power of KrCl* molecules reaches its maximum. In our experiments, we used a coaxial excilamp described in [1, 2]. It was manufactured from two quartz tubes. The external diameter of the outer tube with walls having a thickness of 2.5 mm was equal to 65 mm, and the external diameter of the inner tube with walls having a thickness of 1.5 mm was equal to 43 mm. The discharge gap was 8.5 mm. The excilamp length was 57 cm. The length of the external grid electrode was 10 cm. The internal electrode was made from an aluminum foil. The inner quartz tube together with the electrode was cooled with running water. Unipolar excitation pulses had amplitudes up to 8 kV, duration of ~1.5 µs, and repetition frequency of 75 kHz. The given excitation pulse repetition frequency ensured comparatively high average radiation power and efficiency of KrCl* molecules (λ ~ 222 nm) equal to 35 mW/cm 2 and 11%, respectively. To study the dynamics of the barrier discharge ignition, excitation pulses were applied in series with variable number of pulses in each train. Pulse trains were applied with a frequency of 1 Hz. The pulse train duration increased continuously from 10 µs to 0.5 s, which allowed us to trace successive changes in the discharge shape after switching on the excilamp. During the experiment, we registered the voltage pulses on the excilamp, the discharge current, and radiation pulses of KrCl* molecules in the UV range of the spectrum. In addition, we took photographs of the discharge glow at various durations of the excitation pulse trains. Before the experiments, we optimized the pressure and composition of the working mixture. The best results were obtained at a pressure of 200 Torr for the Kr : Cl2 mixture in the ratio 200 : (1–0.5). The main results of the experiments were the following. The excilamp started to operate in the stationary mode 1 min after switching on, and four characteristic discharge stages were identified during the first second of its operation. In the first stage (0–0.02 ms), the space discharge was formed in the interelectrode gap. In the second stage (0.02–0.1 ms), filamentary channels were registered against the background of the space glow; then (0.1–400 ms) the filamentary channels were converted into a comparatively small number of bright branching channels whose radiation efficiency was low. In the fourth stationary stage typically formed in ∼1 s, only cone-shaped microdischarges were observed, and the efficiency of