Effect of DC Pre-Discharge on the Generation of Atmospheric Pulsed Microdischarges

Abstract

The effect of driving frequency in the range of 13-50 MHz on the electron energy probability function (EEPF) in capacitively coupled discharges is studied using particle-in-cell/Monte Carlo(PIC/MC) method. The EEPFs obtained from the simulation generally agree well with the measured data of Abdel-Fattah and Sugai. Detailed comparison shows, however, that the transition from the Druyvesteyn distribution to the bi-Maxiwellian type with increasing driving frequency is not as clear in the simulation as in the experiment.Capacitively coupled plasmas (CCPs) have been extensively used in material processing such as etching and sputtering. The use of very high frequency (VHF) CCPs is a recent trend. When the driving frequency increases from 13.56 MHz to VHF range (30-100 MHz), the characteristics of the plasma drastically change, e.g., plasma density increases and dc self-bias voltage decreases. Further advantages of the VHF CCPs are reported in ref. 1. Abdel-Fattah and Sugai 2) measured the effect of the driving frequency (13-50 MHz) on the electron energy probability function (EEPF) in capacitively coupled argon plasma.Wakayama and Nanbu 3) developed the one-dimensional (1D) simulation code for CCPs for arbitrary frequency. The most severe method of validating the plasma simulation code is by checking whether the EEPF obtained from experiments can be reproduced. The EEPF is one of the most important characteristics of processing plasmas because all types of rate constant are determined by the EEPF. In this work, we examine the effect of driving frequency on the EEPF and compare the simulation results with the experiment conducted at Nagoya University. 2) The 1D computational domain based on the experimental apparatus described in ref. 2 is shown in Fig. 1. In the experiment, 2) the capacitive discharge is formed between two stainless-steel electrodes with a diameter of 15 cm separated by H ¼ 6:5 cm. The lower electrode is powered by the RF-to-VHF (13.56-50 MHz) generator through a matching box, while the upper electrode is grounded. The spaces between each electrode and the chamber wall are filled with insulating material to prevent the plasma from expanding outward, and hence, make the self-bias as negligibly small as 1-2 V, thus making the discharge symmetrical.The discharge conditions are chosen to be equal to those of the experiment; the peak-to-peak voltage V pp is 80 V, the gas pressure is 100 mTorr, and the discharge gas is argon. As for eAr collisions, elastic collision, ionizing collision, and 25 exciting collisions are considered on the basis of a set of cross-section data obtained by Hayashi. 4) The collision judgment and the determination of collisional event are carried out using Nanbu's method. 5) Elastic collision and resonant charge-exchange are considered for Ar þ Ar collision using the simple and computationally efficient model. 6) Calculation is carried out for the driving frequencies of 13.56, 27, 37, 44, and 50 MHz by fixing the discharge voltage and gas pressure. Figure 2 shows the EEPFs at freque...