We report an analysis of the dynamics of a pulsed power, inductively-coupled chlorine plasma operated with continuous radio frequency (rf) bias applied to the substrate stage. A comparison of pulsed plasmas operated with and without 12.5 MHz rf bias is performed through an investigation of the time dependences of electron ( n e ) and positive ion ( n i + ) densities and electron temperatures ( T e ), measured with a Langmuir probe in a 10 mTorr Cl 2 plasma. There is no significant difference in the plasma characteristics with or without bias during the on portion of the power modulation of the source. Once the source power is turned off, T e initially decreases rapidly, and n e and n i + decay slowly, independent of the presence of the rf bias. About 25 µs into this decay, when T e is about 0.5 eV (and would continue to decay in the absence of the rf bias), the presence of the rf bias (70-100 W) causes T e to increase rapidly and reach values higher than those recorded at the end of the on portion. Meanwhile, n e and n i + continue to decay, independent of substrate bias. Only much later in the afterglow are charge densities affected by bias. About 50 µs into the off period, n e with added substrate bias is lower than with no bias. If the off period is sufficiently long, the plasma in the presence of the rf bias transforms into a so-called reactive ion etching (RIE) mode that is generated and sustained solely by the capacitively-coupled bias power. This occurs rather abruptly when n i + decays to the level of positive ion density for RIE operation (i.e. with the stage powered and the source power off). Since n e at this instant has decayed well below the electron density during RIE operation, it rapidly increases when this transition occurs. The behaviour of T e in the off portion of the pulsed plasma indicates that the presence of rf bias prevents the sheath collapse that occurs without bias. Since this will prevent negative ions (Cl - ) from reaching the wafer, the reduced plasma-induced damage (with respect to continuous source operation) reported for pulsed plasma operation with high-frequency (e.g. ~13 MHz) bias cannot be caused by the discharging of the wafer by negative ions.
No takes yet. Share an insight, caveat, or question.
Malyshev et al. (2000) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: