Ion Beam Synthesis of Silicon Carbide: Infra-Red and RBS Studies
- PDF / 305,545 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 98 Downloads / 146 Views
which contains oxygen and carbon impurities. We can see in this figure that at a temperature of 400*C a typical signature of SiC is obtained. When carbon implantation is performed at room temperature, Ishikawa et al [7] have shown that post annealing at 9000C was necessary to g~t the same result. Below 400°C in our case, a broad I.R peak, positionned at 725 ± 5 cm-, reveals the presence of amorphous SiC [1]. Above 400"C, the magnitude of the SiC infra-red response is improved, however the minimum transmittance value corresponding to monocrystalline SiC layer is not reached. RBS results on these samples did not reveal a significant change of crystalline quality. This has to be attributed to the fact that RBS is sensitive to long rang crystalline order, while I.R measurements probe the short range structure of the layer. 190
i
60
.............
.................
S70 "5500•
400
of-•":
:1,•,
...
70600
... ... 900 .......... 30 1000
900
800
700
600
500
400
Wave number (cnml) Fig. 1: Differential I.R transmittance of implanted samples at different temperatures. To go further into the correlation between the I.R analysis and the microscopic details of the synthesized SiC layers, we use the classical dispersion theory based on the following dynamic equation which describes the ionic vibration modes d2U
m dt- +my
dU
TO
02-0
TE(I
C+mOO eTm
(1)
Durupt [8] showed that y can be extracted from the experimental data when transformed according to the relation 1/T((o) - 1= f (w0 / O)TO)
(2)
where T(co) is the experimental transmittance. The function f(oVdhp) is of Lorentzian type with the full width at half maximum (FWHM) being equal to the damping factory. As it appears in the dynamic equation above, this factor describes the damping of the optical vibrational waves and should therefore depend on the crystallinity of the environnement. To illustrate this correlation an analysis in monocrystalline bulk 6H-SiC damaged by Ge implantation at different doses was performed [9]. y increases from 2x10-3 , in the control sample, up to 4x10"2 for a dose of 1015 cm-2 which is necessary to yield an RBS signature typical for an amourphous layer. In our 1-SiC layers the behavior of the damping factor is shown in Fig. 2. The data represented by open circles show a strong decrease of ywhen the synthesis temperature 232
increases, clearly establishing the improvement of the crystal quality. The same sets of samples have undergone a post-implantation anneal at 1200 *C for 30 min. The results represented by crosses show that the post-implantation anneal is efficient only for samples synthesized at low temperatures. As expected, the treatment seems to have a reduced effect when implantation is performed at high temperatures. 0.3 0.25 0.2
8 .5 0.15 2• a 0.1 0.05 0 0
200
400
600
800
1000
Temperature (°C) Fig.2: Impact of the implantation temperature and post-annealing on the damping factor.
49.88
CO 42.46 r 32.88 19.49
Surface
600
1200
1800
2400
Depth (A)
Fig. 3: Comparison between the profile of carbon imp
Data Loading...
