The Schottky barrier height of 4H and 6H SiC/Metal (Pt, Mo, Ti) contacts depe n s on the crystallographic face of the SiC epilayer. The breakdown fields of Sc h ttky contacts without edge terminations are also discussed. Introduction At ICSCRM’2001 we reported the initial results [1] of our detailed investigation of Schottky barriers on four different faces of 6H and 4H SiC homoepitaxial layers. At tha t time, we had available only data for Pt contacts. Also, the n-type doping concentration of the 6H SiC epitaxial layers was about 10 18 cm, which is far too high for our purposes. We have extended our work to include three metals: Pt, Mo and Ti; four faces on both 4H and 6H SiC epilayers: (0001) Si , ) 1 000 ( C, ) 00 1 1 ( and ) 10 2 1 ( ; and three measurement techniques: Current-Voltage (I-V), Capacitance-Voltage (C-V) and Internal Photoemission (IPE) at room tem perature. Our latest Schottky barriers show improved ideality factors (closer to unity) and reduce d leakage current. Experiment The preparation of the 4H and 6H SiC boule pieces was described previously [1]. The substr ate doping concentration was approximately 5×10 -1×10 cm for both 6H and 4H SiC. N-type homoepitaxial layers about 10 m thick were grown by cold-wall CVD at 1520oC. The doping concentrations are of order 10 15 cm for both polytypes of SiC. The surface morphology of the epilayers is: (0001) Si – smooth, small pits, H 2 etched pits, surface pit density 200-500 cm ; ) 1 000 ( C – smooth, occasional 3C hillocks, H 2 etched pits, hillock density 10-30 cm ; ) 00 1 1 ( elongated defects, scratches, hillocks, surface defect density 500-1000 cm ; ) 10 2 1 ( smooth, occasional screw-induced hillocks, hillock density 10-30 cm . On all faces the rms surface roughness measured by AFM is about 3-9 Å in areas without large scale surface defects. C-V and low temperature photoluminescence measurements indicate that donor concentrati ons a e in the range 1-3×10 15 cm, the samples are slightly doped with aluminum, but the compensation is low. Ni ohmic contacts were fabricated on the back sides of the samples by e-beam eva poration followed by annealing at 1000oC for ten minutes. Pt, Mo and Ti Schottky contacts were fa bricated by sputtering with Ar (0.9-2.0×10 -8 torr base pressure). Before deposition the samples were RCA cleaned, etched in HF for five minutes, rinsed in deionized water and dried in N 2 gas. The circular contacts are about 100 Å thick and 0.12 mm and 0.5 mm in diameter for Current-Voltage (I-V) an d Capacitance-Voltage (C-V) measurements, and 1.0 mm in diameter for interna l photoemission (IPE). The setups for I-V, C-V and IPE are discussed elsewhere [1, 2]. Materials Science Forum Online: 2003-09-15 ISSN: 1662-9752, Vols. 433-436, pp 705-708 doi:10.4028/www.scientific.net/MSF.433-436.705 © 2003 Trans Tech Publications Ltd, Switzerland All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications Ltd, www.scientific.net. (Semanticscholar.org-12/03/20,10:57:00) To obtain the reverse high voltage breakdown data, all devices were measured in the dark while immersed in Flourinert FC-77 fluid. I-V curves below 1100 V were taken using a Keithley 237 High-Voltage Source-Measure Unit, while I-V curves above 1100 V were taken on a Tektronix Model 371A digitizing curve tracer. The 237 unit measures DC steps and can resolve much lower reverse leakage current than the 371A. The digitizer records at least a couple of 60 Hz voltage sweeps. The data were collected prior to catastrophic failure. Because the work function of a metal can be anisotropic [3], the morphology and crystallographic orientation of the metal films were investigated using X-ray diffraction. Both symmetrical (angle of incidence = angle of reflection) and glancing angle (angle of incidence = 0.5o) 2θ/ω scans were obtained using a Philips X’pert with a Cu source ( λ = 1.54 Å) operated at 45 kV, 40 mA. An X-ray monochromator was used to eliminate K lines. A symmetrical angular scan shows the metal planes parallel to the sample surface, while a glancing angle scan reveals metal microcrystallites with other orientations. Pt and Mo metal films tend to show a dominant (111) and (110) line, respectively, for the symmetrical angle scan and multiple l ines for the glancing angle scan for all four SiC faces. These results suggest that P and Mo thin films form (111) and (110) textured structures, respectively, for all SiC surface orientat ions. We were not able to obtain satisfactory Ti X-ray data due to its low Z. Based on these results we neglect the role of differences in the metal work function in the analysis. Results and Discussion The standard approaches to extract the Schottky barrier height (SBH), assum ed to be uniform over the interface, from the data are the thermionic emission model for I-V and the photoe missi n current model for IPE [4]. Inhomogeneity [5] can be taken into account by integrat ing over an assumed Gaussian distribution (GD) of noninteracting local barriers [6], specif ied by a mean barrier height and a width. It is essential to include series resistance when applying the GD approach to I-V data in order to obtain values of the ideality factor n greater than unity. The obtained SBH is the mean value. The GD method enables a comparison of a set of contacts on the same sample having different ideality factors. Fig. 1 shows that the mean SBH obtained using the GD method is: 1) independent of the value 6H-SiC I-V Work-function [eV] 4.0 4.4 4.8 5.2 5.6 6.0 S ch ot tk y ba rr ie r he ig ht [e V ] 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 0001 Si 000-1 C 1-100 1-210 Ti (polycrystal) Mo (110) Pt (111) 4H-SiC I-V Work-function [eV] 4.0 4.4 4.8 5.2 5.6 6.0 S ch ot tk y ba rr ie r he ig ht [e V ] 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 0001 Si 000-1 C 1-100 1-210 Ti (polycrystal) Mo (110) Pt (111) Fig. 3. SBH versus metal work function for four faces of 4H SiC obtained by forward I-V. Fig. 2. SBH versus metal work function for four faces of 6H SiC obtained by forward I-V. Ideality factor n 1.00 1.05 1.10 1.15 1.20 1.25 1.30 φ S [e V ] 0.95 1.00 1.05 1.10 1.15 1.20
[1]
E. Burian,et al.
C–V dependence of inhomogeneous Schottky diodes
,
1998
.
[2]
E. Dobročka,et al.
INFLUENCE OF BARRIER HEIGHT DISTRIBUTION ON THE PARAMETERS OF SCHOTTKY DIODES
,
1994
.
[3]
Tung,et al.
Electron transport at metal-semiconductor interfaces: General theory.
,
1992,
Physical review. B, Condensed matter.
[4]
H. Michaelson.
The work function of the elements and its periodicity
,
1977
.
[5]
Wolfgang J. Choyke,et al.
Static Dielectric Constant of SiC
,
1970
.
[6]
W. Mönch.
Semiconductor Surfaces and Interfaces
,
1993
.