Annealing effects on aSiC:H and aSiC:HF thin - PowerPoint PPT Presentation

1 / 20
About This Presentation
Title:

Annealing effects on aSiC:H and aSiC:HF thin

Description:

of the exposure time, an experimental indication of a gas diffusion ... between the two curves. Assuming spherical microvoids, we calculate their radii from ... – PowerPoint PPT presentation

Number of Views:30
Avg rating:3.0/5.0
Slides: 21
Provided by: tflabS
Category:
Tags: asic | annealing | effects | ray | thin

less

Transcript and Presenter's Notes

Title: Annealing effects on aSiC:H and aSiC:HF thin


1
Annealing effects on a-SiCH and a-SiCH(F)
thin films deposited by PECVD at room temperature
Dong S. Kim and Young H. Lee, Drexel
University Thin Solid Films (1995)
Seminar 24
2
Introduction
Khakani et al. reported that the annealing of
a-SiCH resulted in relaxation of the compressive
stress due to the dissociation of hydrogenated
bonds (Si-H and C-H).
In this work, the annealing behaviors of
fluorinated and non-fluorinated a-SiCH films
deposited at room temperature under heavy ion
bombardment were studied. The effect of annealing
on the stress, bonding structure, hydrogen
content and optical properties were investigated.
3
Experimental
4
Ramp rate 10oC When a sample is quoted or
labeled as annealed at 600oC, it was actually
subjected to the cumulative effects of 200, 300,
400, 500 and 600oC annealing, each of 1h duration.
5
For diamond-like amorphous silicon carbide, the
fraction of microvoid is relatively
small. However, as the annealing temperature
is increased the void fraction may increase. Note
that as the annealing temperature is increased
from 200 to 500oC, the peak Height increases,
indicating that there is a certain degree of
cross-linking and/or oxidation. However, when the
annealing temperature is 600oC, the peaks
broaden somewhat and an additional peak
appears. In particular, the strong peak at around
1040cm-1 (corresponding to Si-O stretching)
indicates that oxidation has occurred.
6
Si-O stretching mode show up.
7
desorption
There is little change as the annealing temperatur
e is increased up to 400oC, but after annealing
at 500oC, decreases are observed. This could be
due to thermally induced hydrogen desorption.
8
(No Transcript)
9
Direct evidence of porosity in carbon-rich hydroge
nated amorphous silicon carbide films
R. Arce, R.R. Koropecki, R. H. Buitrago,
Argentina F. Alvarez and I. Chambouleyron,
Brazil J. Appl. Phys. (1989)
10
Introduction
a-Si1-x Cx H The number of silicon-carbon
bonds reaches a maximum for x between 0.6 and
0.7. Menna et al. concluded that the
incorporation of carbon into the network results
in a very large void fraction. Mui et al. found
that the fitting of optical data with the complex
dielectric function gives a maximum in the void
fraction for x?0.4. Beyer et al. found two peaks
in the hydrogen effusion curve occurring at low
and high temperatures, respectively. The hydrogen
evolved at low temperatures is associated with
the presence of voids producing a material with
poor electronic properties. We report a study on
the evolution of IR spectra of samples with
different carbon contents. It has been found that
samples possessing a low carbon content present
stable IR spectra, even after a long exposure to
oxidizing atmospheres. On the contrary, samples
having a high carbon content incorporate oxygen
when exposed to air. This experimental finding is
interpreted in terms of a carbon-induced open
porous network. This oxdation kinetics of a-Si1-x
Cx H samples, isothermally annealed at 200oC in
air, has been determined for carbon-rich material.
11
Experimental
SiH4 CH4 substrate Si and Corning 7059
glass pressure 50Pa substrate temp. 260oC rf
power 2.5W total gas flow rate 10sccm film
thickness 500nm gaseous mixture y
CH4/(SiH4 CH4) between 0 and 0.87 for
different carbon. The carbon content f the
samples grown with a gaseous mixture y0.87 was
estimated to be x0.450.05, which corresponds to
the highest carbon content among the
studied samples.
12
Results and Discussion
The samples were measured immediately after
deposition and after six months of exposure to
air at room temperature.
The major changes in absorption
y 0.87
Sample having a smaller carbon content (y lt
0.7) do not present any particular feature in
their differential spectra, an indication of the
material stability. Under 200oC annealing no
aging effects were detected in the low carbon
content samples either.
13
Si-O-Si Si-O-C
The differential absorbance spectra
are interpreted as due to the formation
of Si-O-Si and Si-O-C complexes in detriment of
the Si-C bonds.
()
420cm-1
1270cm-1
Si-O-Si (b)
Si-CH3
Si-C
650cm-1
(-)
14
Within the studied time interval, the process of
oxygen incorporation in the carbon-rich samples
does not saturate. The experimental points fall
on a straight line with a slope of approximately
0.6. This value is close to the slope of 0.5
expected for a diffusion-limited process in
solids.
In the low carbon content samples, conversely, a
thin oxide barrier is probably formed at the
surface preventing any further oxidation.
15
Conclusion
We observed that samples having a low carbon
content ( x lt 0.4) do not display the Si-O-Si
absorption band, even when annealed at 200oC in
air. On the contrary, samples with a higher
carbon content (x ? 0.45) oxidize spontaneously
in air at room temperature. Similar effects are
found when these samples are annealed in air at
200oC. The oxygen incorporation has been shown to
be nearly proportional to the square root of the
exposure time, an experimental indication of a
gas diffusion limited process in solids. We
attribute these effects to the existence of a
connected void network.
16
Microvoids in amorphous Si1-xCx H
alloys studied by small-angle x-ray scattering
D. L. Williamson, A. H. Mahan, B. P. Nelson, and
R. S. Crandall, Colorado Appl. Phys. Lett. (1989)
17
Introduction
The size, shape, or number density of microvoids
was obtained. Samples In a series of a-Si1-xCx
H sample with 0 x 0.3 Mass density
measurement IR study gas mixture CH4 /
(CH4SiH4) substrate temperature 250oC pressure
600mTorr total gas flow rate 80sccm rf power
70mW / cm2 Substrate 1) 10?m ? 2.5cm ? 2.5cm
iron-free Al foil for the SAXS measurements
After deposition the Al foil was cut into eight
strips about 2-3mm wide and stacked to
produce an a-SiCH sample of sufficient
thickness 2) Crystalline Si for the IR
measurements
18
Experimental
The line-collimated x-ray beam was 150?m ? 2.2cm
at sample. The settings allow detection of
particles or density fluctuations with a
maximum size of about 26nm Electron microprobe
(EPMA) analysis using a stoichiometric SiC
crystal as a reference standard determined
x. Magnitude of the scattering vector h
2?(2?)/? A scan of the scattering angle 2? from
0.15o to 8.4o.
19
Results and Discussion
A radius of gyration Rg I(h)
I(0)exp(-Rg2h2/3) characterizes the size of the
electron density fluctuations This is evident
for the a-SiH sample(x0) and the two alloy
samples of lowest C content (x0.024,
0.046), even well beyond the limit where the
Guinier approximation is known to be accurate (Rg
h 1)
20
microvoid number density n ?f / V
To examine the shape o the microvoids, a SAXS
scan was made with the x0.3 sample tilted at an
angle of 45o and was compared to the nontilted
data. No difference was seen between the two
curves. Assuming spherical microvoids, we
calculate their radii from Rg and list them in
Table I. A large fraction of the microvoid
surface sites is not hydrogenated. Finally, the
dangling bond density in these materials is three
to four orders of magnitude less than the
microvoid number densities so that considerable
surface reconstruction on the microvoid
surfaces must occur.
Write a Comment
User Comments (0)
About PowerShow.com