Effect of hydrostatic pressure on the characteristic parameters of Au/n-GaAs Schottky-barrier diodes G. C ¸ ankaya, N. Uc ¸ar, E. Ayyildiz, H. Efeog ˘ lu, A. Tu ¨ ru ¨ t,* S. Tu ¨ zemen, and Y. K. Yog ˘ urtc ¸u Physics Department, Faculty of Arts and Sciences, Atatu ¨rk University, Erzurum, Turkey Received 16 November 1998; revised manuscript received 10 May 1999 Au/ n -GaAs Schottky barrier diodes SBDs have been fabricated. Schottky diode parameters such as the ideality factor, the series resistance, and the Schottky barrier height SBH,  b , have been measured as a function of hydrostatic pressure using the current-voltage ( I - V ) technique. We have seen that the SBH has a linear pressure coefficient of 11.21 meV/kbar =112.1 meV/GPa. Also, the series resistance value increases with increasing pressure. We have concluded that the variation of the barrier height due to the applied pressure should follow precisely the variation of the semiconductor band gap, accepting that the Fermi level is a reference level which is pinned to the valance-band maximum as a function of the pressure. That is, we have concluded that the experimental results is in agreement with the model that the pressure coefficient is caused by the pressure coefficient of the direct midgap level. S0163-18299907643-2 INTRODUCTION The recent developments in the field of Schottky contacts have opened up interesting new possibilities for the study and application of these materials. In particular, Schottky barriers SB’s appear to hold considerable promise for pho- tovoltaic solar energy conversion and other device applications. 1–3 For many years, pressure has been employed as a tool in the investigation of the properties of semicon- ductor materials. 4 In recent years, various studies have been made on the hydrostatic press effects in Schottky diodes SD’s, and the effects were explained by the introduction of stress-sensitive generation-recombination centers, the change of band structure, and the change of minority-carrier lifetime. 5–9 In some studies, 10,11 it has been found that the Schottky barrier height SBH is almost independent of the surface orientation of the substrate, preparation of surface, and the type of metal being used. So, the SBH has been measured for ideal metal/GaAs contacts by means of current-voltage ( I - V ) and capacitance-voltage ( C - V ) techniques and has been pos- tulated models for Schottky barrier formation. 12 Recently, hydrostatic pressure has been used to explain the electronic structure of semiconductors by means of SB height measurements. 13 According to this matter, Peanasky and Drickamer 14 have showed that a pressure study of barrier height decouples the measurements of the conduction-band minimum E c movement from the valence-band maximum E v and creates a second frame of reference for interpreting the energy-band movement with pressure. The pressure depen- dence of the SB height at the Pt/GaAs interface has been determined by measuring the forward I - V characteristics of Schottky diodes using a diamond anvil cell DAC by Shan et al. 15 They have found that the SB height shifts to higher energy with a linear pressure coefficient of 11 meV/kbar and have shown that defect states are responsible for pinning the Fermi level in Schottky barriers under pressure. However, Mo ¨ nch 16 and Chen et al. 17 have concluded that the Fermi level is pinned relative to the valance-band maximum VBM to explain both the temperature dependence 16,17 and externally applied pressure dependence 16 of the SBH in the Schottky contacts. Furthermore, Werner and Gu ¨ ttler 18 have found that the pressure coefficients of the barrier heights at type-A and type-B NiSi 2 / n -Si111 interfaces vary between values of about -2.2 and -1.1 meV/GPa. Although an enormous amount of information on metal- semiconductor Schottky diodes has been gained, little is known about the effect of pressure on the diode parameters. The purpose of this study is to investigate the hydrostatic pressure dependence of Schottky diode parameters such as the ideality factor, barrier height, and series resistance by using the I - V characteristics. EXPERIMENTAL METHOD The Au/ n -GaAs Schottky diodes used in this study were fabricated using n-type liquid-phase epitaxy LPE GaAs wa- fers Te doped with 100 orientation and 0.01- cm resis- tivity. The wafer was rinsed by ultrasonic vibration in deion- ized water and was dried by high purity nitrogen. After this procedure, Ohmic contact was made evaporating Au-Ge 12% Ge and annealing 425 °C for 3 min. The Schottky contacts were formed by evaporating Au as dots with a di- ameter of about 1 mm onto the mirror surfaces. All evapo- ration processes were carried out in a turbo molecular fitted vacuum coating unit at about 10 -6 mbar. The pressure was created by a piston and cylinder-type chamber apparatus as sketched in Fig. 1. A special trans- former oil was used to transmit the pressure. Au/ n -GaAs Schottky diodes were located in a pressure cell with a special designed sample holder and I - V measurements under 0–6 kbar hydrostatic pressure were made by electrical connec- tions from cell to measuring device. The value of pressure in the pressure cell was measured with the resistance changes of the manganin wire. I - V characteristics of Schottky diodes under the pressure were performed around an HP4140B pi- coampermeter at room temperature. The high-pressure equipment used in this experiment has been described elsewhere. 13,14 The effect of hydrostatic pressure up to 6 kbar PHYSICAL REVIEW B 15 DECEMBER 1999-I VOLUME 60, NUMBER 23 PRB 60 0163-1829/99/6023/159444/$15.00 15 944 ©1999 The American Physical Society