slide No. 13 / 15

Power dependence of the lines at several gas pressures. These are the results at the observation point of 5 mm, the one nearer the target. In these figures, horizontal axis is the discharge power. Again the both axis is written in log scale.

Also in these figures, Ar lines don't show a strong dependence on gas pressure. And at each pressure, their intensities are roughly proportional to the discharge power.

Cu lines are seem to be proportional to the square of the discharge power. It is suggested that the copper flux sputtered from the target is roughly proportional to the power, and the excitation probability by the electron impact should also proportional to the electron density, hence the power. The emission intensity might be given by the product of these two effects, so is proportional to the square of the power.

Finally, please note that the 510 of Cu line at high pressure of 20 Pa increase again more rapidly than 324 Cu. We think that this difference and the difference in pressure dependence come from the same origin, and would like to explain it in next slide.