Study of  NH 3 Plasma Damage on GaAs Schottky Diode in Inductively Coupled Plasma System

Abstract

GaAs-based devices are currently widely used in wireless applications owing to the advantages of high electron mobility and the availability of the semi-insulating substrate. Many efforts have been put into epitaxial growth and developments of device structures to improve the performance of GaAs based devices. 1,2 To further improve device performance, stability, as well as long-term reliability, device passivation becomes one of the most critical issues. Currently, plasma enhanced chemical vapor deposition (PECVD) with SiH 4 /NH 3 plasma chemistries is used for III-V device passivation. This process involves relatively high ion energies and low plasma density, and operates at a relatively high temperature (250-300ЊC). 3,4 The high energy ion bombardment leads to semiconductor damage. Furthermore, significant amounts of hydrogen and oxygen can be incorporated in the dielectric film and thus cause device instability.There has been extensive interest in the use of high density plasma sources as a replacement for conventional PECVD. Most of the work to date has been performed with electron cyclotron resonance (ECR) sources, which provide high ion density and low ion energy. 5,6 Therefore low temperature deposition can be realized. While these systems work very well, there is some reluctance to introduce them into manufacturing lines because the automatic plasma tuning networks for microwave discharges are less mature than their radio frequency (rf) counterparts, and the uniformity across large diameter wafers has not been fully realized. Recently inductively coupled plasma (ICP) sources powered with radio frequency have been successfully used for etching III-V based semiconductors. 7,8 However, depositions with ICP sources have not been widely investigated. 9,10 In this paper we report on the results of a parametric study of the influences of rf chuck power, ICP source power, chuck temperature, and operating pressure on the GaAs Schottky diodes. The NH 3 discharges were used for the damage study. There was no actual deposition during the experiments, therefore the results shown here can be extremes of the ion bombardment damage and H-passivation in the real deposition. ExperimentalThe layer structure of the Schottky study consists of a 6000 Å n-type Si-doped GaAs layer (2 ϫ 10 17 cm Ϫ3 ) epitaxially grown on n ϩ -GaAs substrates with metall organic molecular beam epitaxy (MOMBE). An ohmic contact was placed on the back side of the wafer using electron-beam evaporation. The contact metallization comprised of Ge/Ni/Au-Ge/Ag/Au (50 Å/50 Å/240 Å-168 Å/500 Å/ 2000 Å) and the metallization was then alloyed at 400ЊC for 20 s. Prior to the plasma treatment, the GaAs native oxides were removed with a solution of HCl:H 2 O (1:20 mixture). The samples were then exposed to the NH 3 discharges in a Plasma Therm 790 inductively coupled plasma (ICP) system. The ICP source operates at 2 MHz with powers of 0-950 W. The sample chuck was biased from 0-100 W at 13.56 MHz. Substrate temperature was also varied from 0-260 ЊC and pressure...

E‐ and H‐mode transition in a low pressure inductively coupled ammonia plasma

“…Many of the experimentalists used RF inductively (ICP) and capacitively (CCP) coupled systems for generating the discharge. Details about the experimental setup are rarely described in the cited literature; often only external parameters, such as the gas pressure and discharge power, at which the experiment was performed, are given.…”