Temperature-Dependent Hall Effect Measurements in Low – Compensated p-Type 4H-SiC

P-type 4H-SiC substrates grown by high temperature chemical vapour deposition with doping concentration in the range of 10 17 cm -3 to 10 19 cm -3 have been characterised by temperaturedependent Hall effect. The study has been performed in the temperature interval from 100 K to 850 K using the Van-der-Pauw method. Low resistivity of 3.57 Ωcm to 0.45 Ωcm, depending on the concentration, is obtained at 300 K. It decreases to 1.98 – 0.14 Ωcm, respectively at 600 K. The concentration NA and the ionisation energy ∆EA of the Al-acceptors, as well as the concentration of the compensating donors ND have been determined taking into account the Hall scattering factor temperature dependence. The compensation ratio ND/NA varies from 1 to 14 % for the different samples. The Al ionisation energy decreases slightly from 200 meV to 185 meV with increase of Al doping concentration. The highest value of the hole mobility, 217 cm 2 V -1 s -1 , is obtained at 130 K for the sample with Al-acceptor concentration of 2.50x10 17 cm -3 . Introduction High quality p-type substrates are needed for the fabrication of power devices such as MOSFETs, JEFETs, and vertically integrated BTs, [1]. While n-type material has been investigated in much detail, p-type 4H-SiC substrates have been less studied. The knowledge of the material parameters is however important in two aspects: for material evaluation and device simulation. Temperature-dependent Hall effect is the most widely used technique, which provides information about the main material characteristics such as resistivity, free carrier concentration and mobility. The doping and compensation concentration, the degree of compensation, the ionisation energy of the impurities and the scattering mechanisms can also be found. In this work, temperaturedependent Hall effect measurements have been done to characterise low-compensated p-type 4HSiC substrates. The results obtained for the impurity concentration NA have been compared with the Al concentration measured using SIMS analysis and the net doping concentration NA-ND determined from C-V measurements. Experiment Three types of substrates denoted S04, S05 and S07 have been investigated. They had Al concentration in the range of 10 17 cm -3 to 10 19 cm -3 . The substrates were grown by the high temperature chemical vapour deposition (HTCVD) technique, at temperatures above 2000 °C. Similar to conventional CVD; HTCVD uses purified gas precursors (silane and a hydrocarbon) as source materials. This enables the growth of semi-insulating SiC crystals with very low background doping of both shallow and deep impurities [2] but also doped Nor P-type material with very low compensation ratio have been grown. The HTCVD technique allows doping since dopants easily can be added in gaseous form. For p-type doping, the metal-organic (MO) source Materials Science Forum Online: 2004-06-15 ISSN: 1662-9752, Vols. 457-460, pp 677-680 doi:10.4028/www.scientific.net/MSF.457-460.677 © 2004 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-11/03/20,15:23:34) trimethylaluminum ((CH3)3Al) has been used. By adjusting the carrier gas flow through the MO source, and the MO source temperature, the Al concentration in the growth cell can be controlled to achieve different doping [3]. For the Hall effect measurements four circular ohmic contacts were formed by aluminium evaporation and subsequent anneal at 950 o C in argon ambient in a resistance furnace [4]. Hall effect measurements have been performed in the temperature range of 100 K to 850 K using the Van-der-Pauw method. The net doping concentration has been measured by capacitance-voltage method using a mercury probe and structures with gold Schottky and ohmic contacts. The SIMS analyses were performed using two independent set-ups, on two sets of samples (other than the Hall samples). In the first set-up, the depth profiles of B and Al were measured using 15 keV O2 + source and positive (+4.5keV) secondary ions, while the 10 keV Cs + source and negative (4.5keV) secondary ions were chosen to analyse N concentration. The detection limits were 1x10 15 cm -3 for Al, 1x10 15 cm -3 for B, and 1x10 17 cm -3 for N. The boron concentration was found to be below 4x10 15 cm -3 and the nitrogen level below the detection limit. In the second set-up, only the Al concentration was determined for all the samples. The additional analysis of the front side of the S07 Hall sample, performed in this set-up revealed the Al and N concentration of 2.8x10 17 cm -3 and 9x10 15 cm -3 respectively. Results and discussion Measurements of the resistivity (ρ=1/σ) and Hall coefficient (RH) have been performed in a large temperature range from 100 K to 850 K. The free hole Hall concentration (1/(eRH)) and the Hall mobility (σ×RH) have been deduced as a function of the temperature. The ratio RAB,CD/RBC,AD in the van-der-Pauw method [5] has been used as a criterion for the macroscopic homogeneity of the samples. This ratio showed no appreciable changes within the temperature interval of 100-850 K for all the samples investigated, which is an indication of good macroscopic homogeneity. The temperature dependence of the resistivity is one of the main characteristics when the material is intended for high power device applications. The results of our experiments showed exponential decrease of the resistivity when the temperature increases (Fig.1), the decrease being more pronounced at temperatures up to 250 K. In this range, the resistivity drops by five orders of magnitude, while above 250 K it changes by an order of magnitude only. It was found that at higher temperatures the material resistivity depends on the doping concentration stronger than on the temperature. Depending on the concentration, values of 3.57 Ωcm to 0.45 Ωcm were obtained at 300 K and they decreased to 1.98 – 0.14 Ωcm at 600 K. To the best of our knowledge, the resistivity values obtained at 600 K are among the lowest values ever reported for p-type SiC substrates. This result characterises the p-type 4H-SiC substrates under investigation as suitable for high power device applications. Fig. 1. Resistivity vs. reciprocal temperature for 4H-SiC samples with a different acceptor concentration. Fig. 2. Dependence of the free hole concentration on the reciprocal temperature for the different samples. 678 Silicon Carbide and Related Materials 2003