ICMCTF2001 Session T2-1: Principles of Pulsed Plasmas

Monday, April 30, 2001 10:30 AM in Room Sunset

Monday Morning

Time Period MoPL Sessions | Abstract Timeline | Topic T2 Sessions | Time Periods | Topics | ICMCTF2001 Schedule

Start Invited? Item
10:30 AM T2-1-1 Understanding Pulsed DC Magnetron Plasmas Through Time Resolved Electrical Probe Measurements
J.W. Bradley (UMIST, United Kingdom); P.J. Kelly, R.D. Arnell (University of Salford, United Kingdom); H. Bäcker (UMIST, United Kingdom)
The use of pulsed DC magnetron plasma sources is now of great interest for the production of technologically relevant thin films and coatings. This technique is particularly important in the deposition of oxides and nitrides in reactive sputtering processes. However, to date there has been little published work in the area of the diagnosis and modelling of the time dependent plasma parameters in these discharges. We shall present preliminary time resolved Langmuirprobe measurements of the plasma density, electron temperature, plasma and floating potentials and electron energy distribution function at different positions in the plasma and for a range of pulsing frequencies (5 to 350 kHz). Through these results, we shall relate the time dependent plasma parameters to the applied voltage and current waveforms and then ultimately to the structure and properties of the deposited coatings.
11:10 AM T2-1-3 Degree of Ionization and Sputtering Efficiency for Different Cathode Materials in High Power Pulsed Sputtering
U. Helmersson, J Alami (Linköping University, Sweden); A. E. Ehiasarian, K.M. Macák (Sheffield-Hallam University, United Kingdom); W.-D. Münz (Sheffield Hallam University, United Kingdom)
The temporal evolution of the plasma composition in low frequency (50 Hz) high power pulsed magnetron discharge was studied by optical emission spectroscopy. Initially in the pulse, the emission near the cathode sheath is dominated by the signal from Ar transitions, while later metal emission signal dominates over Ar. The maximum in metal emission coincided with the peak in the target current pulse. This may indicate that a transition from Ar sputtering to self-sputtering occurs during the pulse. In view of this we have studied a range of different metals with different self-sputtering yields and measured the sputtering efficiency as compared with that of dc sputtering at similar conditions. We can observe a clear trend with high relative sputtering efficiency for metals with high self-sputtering yield such as Cu and Ag, and a relatively low relative sputtering efficiency for metals with low self-sputtering yield especially Ti, but also Ta and Al. The degree of ionization of the sputtered atoms reaches 30 to 70 % depending on the cathode materials and the process conditions (energy in the pulse).
11:30 AM T2-1-4 Characterization of Pulsed dc Magnetron Discharges for High-Rate Sputtering
J. Vlcek, J. Musil, P. Belský, A. Pajdaru, J. Lestina (University of West Bohemia, Czech Republic)

In several recent years, a great attention has been devoted to a reactive pulsed-dc magnetron sputtering. The main aim of this effort is to master a reproducible formation of insulating, e.g. oxide, films with a high deposition rate. Pulsed dc magnetron sputtering has a great potential also for many other applications. Its main advantage is that the target power loading in a pulse can be considerably higher than the maximum target power loading in dc sputtering (limited by target heating). This makes it possible to form films under new physical conditions, particularly (i) at higher deposition rates, (ii) at highly ionized fluxes of sputtered particles, and (iii) at simultaneous sputtering and evaporation from a partially molten target.

The article characterizes pulsed dc magnetron sputtering in a wide range of experimental conditions. The depositions were performed in a standard stainless steel vacuum chamber using an unbalanced planar circular magnetron operated with a Cu target, 100 mm in diameter. The substrate-to-target distance was 100 mm. The sputtering process was primarily controlled by the repetition frequency of pulses, fr, (1 - 50 kHz), the length of the voltage pulse, t1, (10 - 200 microsec) with the t1/T ratio, where T = 1/fr, in the range from 0.2 to 0.9, the argon pressure, pAr, (0.1 - 5 Pa) and the average pulse current, Ida, (2 - 60 A). The maximum pulse voltage and maximum pulse current of the used computer-controlled (see calculation of the Ida values) pulsed power supply were 1000 V and 120 A, respectively. The deposition rate of Cu films was higher than 2 µm/min at maximum average target power loading in a pulse Pda = 600 W/cm2.

The discharge current-voltage characteristics, i.e. dependences of the magnetron voltage Ud on pre-set values of Ida, were measured to explain the obtained dependences for the deposition rate of Cu films. Time resolved optical emission spectroscopy was used to study formation of high density metal plasma in front of the sputtered Cu target. Time resolved mass spectrometry with energy resolution was carried out near the substrate to characterize occurrence of the Cu+ and Ar+ ions and their kinetic energies.

11:50 AM T2-1-5 Diagnostic of a Pulsed CH4-H2 Plasma to Improve MWPA-CVD Process for Diamond Synthesis
L. De Poucques, G. Henrion, J. Bougdira, R. Hugon (CNRS UPRESA, France)

Owing to its physico-chemical properties, diamond is an excellent material for applications in optics and microelectronics where hard conditions (temperature, power) of use may often be encountered. To achieve high purity and high quality diamond thin film deposition, one of the best way consists in using microwave plasma assisted chemical vapour deposition (MWPACVD) with a modulation of the microwave power.

The aim of the present contribution is to investigate the influence of time-parameters of the pulsed discharge (namely discharge and afterglow duration) on the plasma reactivity in close connection with the properties, the quality and the growth rate of the deposited diamond layers.

Time- and space-resolved plasma characterization by means of optical emission spectroscopy and laser induced fluorescence point out that small size microwave plasmas are mainly governed by the diffusion of neutral and charged particles and the heating of the neutral gas as well. In particular, we show that these phenomena are responsible for the displacement of the discharge during the pulse period. A particular study of this displacement as a function of the plasma parameters (pressure, power) allows us to determine the limit process conditions and to show that these limits may be enlarged (as compared to a continuous regime) by using a suitable modulation of the microwave power, and thus to increase the plasma reactivity with respect to the substrate surface. Moreover, the variation of the species concentration with time over the pulse period exhibits different behaviour for reactive species such as H-atoms, CH- and C2- radicals.

Consequently, these results lead us to determine the best value for the on- and off-time of the pulsed discharge and to establish some relationships that describe the growth rate and the quality of diamond films as a function of the concentration of reactive species.

12:10 PM T2-1-6 Time-Resolved Investigation of Pulsed-dc Reactive Sputtering of Dielectrics
Z. Zhao, A. Belkind (Stevens Institute of Technology); D. Cater, G. McDonough, G. Roche, R. Scholl (Advanced Energy Industries, Inc.)
DC reactive sputter deposition of dielectric can be greatly effected by arcing. Observations have indicated that arcing is due to breakdown of the dielectric (oxide) film, which grows on the surface of the metal target as a result of positive charge accumulation. The use of pulsed-DC power has been employed to reduce or eliminate arcing. Using duty cycles, which could be varied between 10% and 50%, plasma dynamics were studied. The relationship with various deposition process parameters were studied using time-resolved electrical and optical measurement techniques and the results are discussed here-in.
Time Period MoPL Sessions | Abstract Timeline | Topic T2 Sessions | Time Periods | Topics | ICMCTF2001 Schedule