Title: MOCVD%20growth%20of%20GaN%20on%20SiC%20at%20VCU
1MOCVD growth of GaN on SiC at VCU
- Projects completed
- Study of the temperature dependence of the GaN
buffer layer on SiC. - GaN HT epilayer overgrowth on nano-network of HT
GaN buffer layer - and comparison with growth on LT continuous
GaN buffer layer. - Effect of deposition of SiN on the GaN epilayer
both on HT and LT - buffer layer (scope of future work)
2Temperature dependence of the GaN buffer layer on
SiC
- GaN buffer layer deposited on 6H-SiC, 3.50 miscut
samples at four different temperatures - 5650 C (temp. used for buffer layer
deposition on sapphire, reference (a) - 9000 C
- 9500 C
- 9700 C
(temp. range recently reported in reference (a)
below)
- Thin solid films, 289 (1996), 256
- J. Cryst. Growth, 248 (2003), 533
3Temperature dependence of the GaN buffer layer on
SiC
900 0 C, 7 min
950 0 C, 4.5 min
- Buffer layer at 5650 C showed featureless
- continuous surface in SEM.
970 0 C, 4.5 min
4Temperature dependence of the GaN buffer layer on
SiC
- XRD data of the buffer layer on SiC.
Sample Temp.(time) 0C (mins) XRD FWHM (arcminutes) XRD FWHM (arcminutes)
Sample Temp.(time) 0C (mins) Sym. Asym.
SiC_36 5650 C (4.5) Broad broad
SiC_31 9000 C (7) 6.6 19.2
SiC_30 9500 C (4.5) 5 13.2
SiC_32 9700 C (4.5) 7.8 13.2
- Conclusion HT buffer layer at 9500 C is the
optimum for epilayer growth.
5HT epilayer growth on LT (5650 C) and HT (9500C)
buffer layer
400 nm of HT (10500) epilayer GaN was deposited
on 100 nm of both LT (5650 C) and HT(5650 C)
buffer Layer to study the effect of the buffer
layer on the quality on the overgrown film.
Sample Buffer layer Temp. Buffer layer XRD (arcminutes) Buffer layer XRD (arcminutes) Epi layer XRD (arcminutes) Epi layer XRD (arcminutes)
Sample Buffer layer Temp. Sym Asym Sym Asym
SiC_37 5650 C Broad Broad 14.7 29.6
SiC_35 9500 C 5 13.2 2.4 9.9
- Conclusion Based upon the xrd data, there is a
significant improvement on the epi-layer - quality grown on HT buffer layer. This may be
due to two reasons - The starting buffer layer quality is better for
HT buffer layer. - Lateral overgrowth taking place in the film grown
on the nano-network of the HT buffer - layer, as opposed to the growth on a continous
LT GaN buffer layer, as shown in the SEM - images on the next two slides.
6HT epilayer growth on LT (5650 C) and HT (9500C)
buffer layer (continued from the previous slide)
- SEM
- - HT epilayer on the continuous LT buffer layer
showed a continuous featureless - epilayer.
- - HT epilayer on the nano-network buffer layer
at 9500 C showed a continuous - epilayer but with pinholes. These pinholes
result from the partial coalescence of the - GaN epilayer after nucleating at nano network
the buffer layer. -
100 nm buffer layer at 9500C
400 nm epi layer at 10500 C on 9500C buffer layer.
7HT epilayer growth on LT (5650 C) and HT (9500C)
buffer layer (continued from the previous slide)
To study the evolution these pinholes, two HT
epilayers of different thickness (200 nm and 400
nm) were grown on the HT nano-network buffer
layer. The following SEM image shows that the
density of these pinholes decreased significantly
with the thickness of the epilayer. In addition,
these pinholes are hexagonal in shape (more
visible in the 400 nm thick layer) whose walls
are in the a-direction, the direction of the
lateral overgrowth.
100 nm buffer layer at 9500C
200 nm epi-layer at 10500C
400 nm epi-layer at 10500C
8Effect of deposition of SiN on the GaN epilayer
both on HT and LT buffer layer
- Two series of experiments done
- 1. SiN deposition on HT buffer layer and then 2
?m epilayer growth. - 2. SiN deposition on LT buffer layer and then
400 nm epilayer growth. - Experiment 1. Growth detail
9Effect of deposition of SiN on the GaN epilayer
both on HT and LT buffer layer (continued from
the previous slide)
Experiment 1. SiN deposition on HT buffer and
then 2?m epilayer growth at 10500 C.
Comments 1. Substrate 6H SiC, on-axis,
h-etched. 2. The surface of the
epilayer shows pinholes due to the partial
coalescence.
10Effect of deposition of SiN on the GaN epilayer
both on HT and LT buffer layer (continued from
the previous slide)
Experiment 2. SiN deposition on LT buffer and
then 400 nm epilayer growth at 10500 C.
Growth detail
11Effect of deposition of SiN on the GaN epilayer
both on HT and LT buffer layer (continued from
the previous slide)
Experiment 2. SiN deposition on LT buffer and
then 400 nm epilayer growth at 10500 C.
SEM SEM shows flat featureless GaN surface.
XRD results
asym
sym
Comments 1. Substrate 6H SiC, 3.50 miscut,
h-etched. 2. XRD value does not change
as much as in the case of the growth on the HT
buffer layer. 3. The XRD value of the
films grown on the HT buffer layer (with or
without the SiN deposition is far better than
the film on LT buffer layer with SiN
deposition. Sym 2 arcmins for film on
HT buffer layer 10 arcmins LT buffer layer
Asym 8 arcmins for film on HT buffer
layer (3 min SiN deposition) 13 arcmins for
film on LT buffer layer (5 min.
SiN deposition)
12Sample Time (minutes) Temp. (0C) XRD(arcmin) XRD(arcmin) Comments
Sample Time (minutes) Temp. (0C) sym asym Comments
SiC_30 4.5 950 5 13.2 HT buffer layer
SiC_31 7 900 6.6 19.2 HT buffer layer
SiC_32 4.5 970 7.8 13.2 HT buffer layer
SiC_33 6.75 950 5.5 14 HT buffer layer
SiC_34 4.59 9501050 5.4 14.5 HT buffer layer200 nm epi layer
SiC_35 4.518 9501050 2.4 9.9 HT buffer layer 400 nm epi layer
SiC_36 4.5 565 broad broad LT buffer layer
SiC_37 4.518 5651050 14.7 29.6 LT buffer layer 400 nm epi layer
SiC_38 4.590 5651050 9.6 23 LT buffer layer 1?m HT GaN (reference sample for SiN experiment)
SiC_39 4.5(5) 90 565(1020)1050 10 22.5 1st SiN (5 min) deposition on LT buffer layer and then HT GaN growth , 1 mu. (90min)
SiC_40 4.5(10)90 565(1020)1050 8.4 13.2 2nd SiN (10 min) deposition on LT buffer layer and then HT GaN growth , 1 mu. (90min)
SiC_41 4.5(15)90 565(1035)1050 9.6 13.2 3rd SiN (15 min) deposition on LT buffer layer and then HT GaN growth , 1 mu. (90min)
SiC_42 4.5(10)590 565(1035)10501050 9.6 13.2 4th SiN (10 min) deposition on LT buffer layer and then 5 min. annealing and then HT GaN growth , 1 mu. (90min)
Summary of the growths done at VCU in weeks 3