Shigeru Ono, Hayato Nishi, Shinriki Teii

Department of Electrical and Electronic Engineering
Musashi Institute of Technology
1-28-1 Tamazutsumi, Setagaya-ku, Tokyo, JAPAN
ono@ee.musashi-tech.ac.jp

Because high density active species are obtained easily, the plasma jet has been applied to various plasma processes. In general, because the gas heated to 10000 [K] or more in the arc discharge gushes to a reaction chamber through the nozzle in the reduced pressure plasma jet, the flow velocity of plasma reaches the supersonic speed. For instance, this flow velocity is the speed that the active species reaches the substrate in PCVD, and an important parameter which relates directly to the deposition rate. Because the thermal velocity of ions are lower than flow velocity in the down steam of the plasma jet, the ion current characteristic of the electrostatic probe greatly receives the influence of flow velocity. The flow velocity can be determined from the change of the ion current of the probe by applying the orbital motion limited probe theory.

The DC plasma jet was used in the experiment. Major experiment conditions are reaction chamber pressure: 1Torr, plasma gas flow rate 2l/min, and discharge current: 20A, respectively. Axial and radial flow velocity distribution was determined from the voltage current characteristic of two cylindrical probes. One probe was squarely set up for the flow, and one probe of the remainder was set up in parallel for the flow. These probes are made of tungsten, and stand some high temperature. Nitrogen and argon were used as a plasma gas.

Typical experimental results shows that the measured flow velocity has decreased while going from the upstream to the downstream of the plasma jet. The flow velocity reached 2800-3700m/sec in the position of 100mm downstream from the nozzle on the axis. Moreover, flow velocity decreased while parting from the axis though flow velocity was high on the axis. The range at high flow velocity was narrow in the upstream part, and it is wide in the downstream. The experiment result was examined from the viewpoint of continuousness axially of the volume of the gas which passed all sectional areas of the reaction chamber. Moreover, the experiment results were compared as a result of the numerical calculation. In addition, our results were compared with the result of obtaining by the flow velocity measurement with the pitot tube which was a quite independent method.