Introduction:
Improvements in Sputtering
Deposition systems have led to the use of pulsed plasmas. It is
believed that pulsing the plasma causes a temporary increase in
electron temperature and that as pulsing frequency increases higher
plasma potentials result. Higher plasma potentials cause higher
energy ion bombardment of films and thus greater surface mobility of
atoms and more thermodynamically favorable film growth. For the
systematic study the effects of frequency and duty cycle on pulsed
plasmas for sputtering systems, a spectroscopy system was designed
using a JY Intensified CCD detector system to study the effects of
pulsing parameters on the optical emission spectra. Experiment:
In sputtering deposition
systems, a pulsed plasma is generated in a vacuum chamber. Material
from a target inside the chamber is vaporized, and the vapour is
deposited on a substrate. The ratio of the gas mixtures in the plasma
can be controlled remotely, and affects the deposition rate. The
repetition rate and duty cycle of the pulsed plasma generator are
also used to adjust the deposition rate and more importantly the
quality of deposition.
The plasma is optically
monitored using an optical fibre inside the chamber and near to the
plasma. The fiber brings the emission light to the entrance of the
TRIAX 320 spectrometer for analysis. The spectrometer disperses the
light into its individual wavelength components and presents the
signal to the i-SpectrumONE ICCD detector. Together, the TRIAX 320
and i-SpectrumONE ICCD detector system can be used to collect either
steady-state or time based spectra and provide useful information
about the chemical processes inside the plasma chamber which would be
difficult to achieve by other measurement means.
Experimental Set-up:

Features
Benefits
Results:
The ICCD can be used in a
steady-state mode to provide Optical Emission Spectra for the plasma.
This provides a picture of the various chemical species inside the
deposition chamber and can be used to determine the effects of
different gas mixtures and target materials. In steady-state mode the
resulting spectra are the averaged emission over the full integration
time of the measurement. When operated in time gated mode, the ICCD
detector system can collect the signal over a specific time period
which is correlated to the time at which the plasma is triggered. The
start of the plasma pulse was indicated to the system by an
electronic trigger from the Plasma generator. The ICCD collects a
signal at a certain time period (delay) from the trigger and
collected signal only for the time slice (gate width) specified by
the experiment. The signal from the same time slice was repeatedly
collected over multiple pulses and the signal integrated together.
The time between the start of the pulse and the data collection is
automatically incremented by the software and the experiment repeated
for a different time slice. All of the time slices are then combined
into single dataset showing the time evolution of the optical
emission spectra.

From this data, the
optimum repetition rate and duty cycle of the pulsed plasma generator
can be determined for different materials and allows the user to
adjust and optimise the plasma discharge for better quality
deposition of thin films.
Conclusion:
Using a Jobin Yvon
spectroscopy system allows for the analysis and optimisation of
pulsed plasmas to provide higher quality deposition as well as
providing a steady-state Optical Emission Spectrum measurement.
Acknowledgements:
Developed in collaboration
with Deposition Sciences Inc.,
Santa Rosa, CA:
M. George, M. Cornett.
|