GaAs MESFET可靠性及快速评价新方法的研究

提出了一种快速评价GaAsFET可靠性寿命的新方法.利用GaAsFET失效敏感参数的温度特性和在一定电应力下的退化特性,及温度斜坡法在线快速提取器件失效敏感参数的退化量与温度的关系,从而进一步求出器件的失效激活能等相关的可靠性物理参数.     

GaAs器件及MMIC的可靠性研究进展

介绍了国外ＧａＡｓ微波器件及ＭＭＩＣ的可靠性研究进展情况，给出ＧａＡｓＭＥＳＦＥＴ、ＨＥＭＴ和ＭＭＩＣ的主要失效模式和失效机理以及在典型沟道温度下的平均寿命代表值。     

GaAs MESFET直流解析模型

根据FET沟道区电场分布的特点,用电子速场特性的分段近似提出了一种新的GaAs MESFET直流解析模型。本文提出的模型可用以计算高夹断电压和低夹断电压GaAs MESFET的直流特性,比完全速度饱和模型及平方律模型更为精确。     

GaAs MESFET的背栅效应研究

本文着重研究了分立GaAc耗尽型MESFET背栅效应的光敏性和尺寸效应。本文还采用一种实验方法——器件在光照下,观察栅压的调制作用,解释了背栅效应机理。不同栅压下,背栅效应的大小不同,零栅压下的背栅效应比负栅偏压下的背栅效应更小。发现衬底峰值电流与背栅效应有密切关系。     

GaAs MESFET的压力敏感特性

本文研究了ＧａＡｓ ＭＥＳＦＥＴ对应力的敏感特性，分析了敏感原理。对ＧａＡｓ ＭＥＳＦＥＴ用作力学量传感器的可能性进行了讨论。     

一种快速精确的GaAs MESFET寄生参数提取方法

提出了一种快速精确地提取GaAs MESFET寄生参数方法.这种方法以两组S参数为基础,应用解析表达式直接确定GaAs MESFET的九个寄生参数.该方法适用范围广.以一个0.3μm×280μm商用器件的一组寄生参数为基础,对该方法进行了验证.     

一种快速精确的GaAs MESFET寄生参数提取方法

提出了一种快速精确地提取GaAs MESFET寄生参数方法．这种方法以两组S参数为基础,应用解析表达式直接确定GaAs MESFET的九个寄生参数．该方法适用范围广．以一个0.3μｍ×280μｍ商用器件的一组寄生参数为基础,对该方法进行了验证．     

GaAs场效应晶体管不同极性ESD损伤机理

对GaAs场效应晶体管(FET)进行3个正向和3个负向脉冲(3“+”3“-”)、3个负脉冲(3“-”)、3个正脉冲(3“+”)3种极性静电放电(ESD)实验,不同极性ESD实验下器件的失效阈值不同.以栅源端对为例对实验结果进行分析,在3“+”3“-”和3“-”极性下,器件失效模式为栅源短路,在3“+”极性下器件电参数退化.运用热模型对ESD正负脉冲电压产生的温升进行了计算,器件的损伤机理为,在正向脉冲下为栅金属纵向电迁移导致肖特基势垒退化;在ESD负向脉冲下为高电场引起栅源端对击穿.     

用于光通信的GaAs集成电路

介绍了用于 1 0 Gb/s光传输系统前端模块的超细栅 Ga As FET器件的制作技术以及制作的数 /模 IC芯片组     

Reliability Study of GaAs MESFET's

Failure modes have been studied phenomenologically on a small-signal GaAs MESFET with a 1mu m aluminum gate. Three major failure modes have been revealed, i.e., gradual degradation due to source and drain contact degradation, catastrophic damage due to surge pulse, and instability or reversible drift of electrical characteristics during operation. To confirm the product quality and to assure the device reliability, a quality assurance program has been designed and incorporated in a production line. A cost-effective lifetime prediction method is presented that utilizes correlations between RF parameters and dc parameters calculated using an equivalent circuit model. Mean time to failure (MTTF) value of over 10/sup 8/ h has been obtained for the GaAs MESFET for an operating channel temperature of 100/spl deg/C.     

C波段功率GaAs MESFET的可靠性

本文通过功率GaAs MESFET的筛选与环境、寿命试验阐述了器件的早期失效模式模型,估计了可靠性水平,指出了主要失效因素,并用实验方法讨论了突变烧毁失效,分析了失效原因,探讨了提高可靠性的途径。     

New modeling of GaAs MESFET's

A first theoretical analysis is given based on a new model of GaAs MESFET's which considers the inherent effects of a free-surface depletion layer between source and gate as well as between gate and drain. Change of surface potential according to the input gate voltage causes variable series resistance and variable gate capacitance to be added to the intrinsic FET. The parasitic effects are now quantitatively estimated and an improved guideline for the design and the fabrication process is given. Detailed calculation of the effects of device parameters for recessed gate structure and some comments on the optimization of n⁺-layers in self-aligned structure are included. The effects of the interfacial depletion layer between active layer and substrate is also estimated in terms of drain voltage and the ratio of total deep levels density in the substrate to donor density in the active layer.     

GaAs dual-gate MESFET's

Performance of GaAs dual-gate MESFET, including high-frequency noise behavior, was analyzed on the basis of Statz's model. Under the design considerations developed from the analysis, fabrication and characterization of a prototype device were carried out. The present analysis was confirmed to reproduce satisfactorily the performance observed. Minimum noise figure and associated gain observed in the device with two 1-µm gates were; 1.2 dB and 16.7 dB at 4 GHz, 2.2 dB and 16.3 dB at 8 GHz, and 3.2 dB and 12.6 dB at 12 GHz, respectively. More than 35-dB gain controllability was also obtained at 8 GHz.     

Backgating in GaAs MESFET's

The phenomenon of backgating in GaAs depletion mode MESFET devices is investigated. The origin of this effect is electron trapping on the Cr²⁺and EL(2) levels at the semi-insulating substrate-channel region interface. A model describing backgating, based on DLTS and spectral measurements, is presented. Calculations based on this model predict that closely compensated substrate material will minimize backgating. Preliminary experimental data support this prediction.     

Backdating in GaAs MESFET's

The phenomenon of backgating in GaAs depletion mode MESFET devices is investigated. The origin of this effect is electron trapping on the Cr/sup 2+/ and EL(2) levels at the semi-insulating substrate-channel region interface. A model describing backdating, based on DLTS and spectral measurements, is presented. Calculations based on this model predict that closely compensated substrate material will minimize backgating. Preliminary experimental data support this prediction.     

Degradation mechanism of GaAs MESFET's

The degradation mechanism of X-band low-noise GaAs MESFET's is examined to obtain meaningful information on a common mode of failure. The devices tested have a half-micrometer gate (Au/ Mo) and source and drain ohmic contacts (Au/Ni/Au-Ge). Zero bias drain conductanceg_{D0}is considered as a representative parameter for degradation during aging. The major failure mode is an increase in series resistance of the ohmic contacts. The amount of degradation, decrease ing_{D0}, is proportional to the square root of aging time, and accompanied by an increase in minimum noise figureF_{min}. A degradation model based on the formation of a high-resistance layer between the ohmic metals and GaAs crystal by a diffusion reaction mechanism is proposed, resulting in excellent agreement between calculated and experimental results. Using ion-microspectroscopy analysis (IMA), diffusion of Ni into GaAs crystal is revealed. Mean time to failure (MTTF) is estimated to be 10⁷-10⁸h at channel temperature of 80°C with an increase inF_{min}of 0.5 dB as failure criterion.     

Analytical model of GaAs MESFET's

A simple analytical model of GaAs MESFET's is proposed. The model is based on the assumption that the current saturation in GaAs MESFET's is related to the stationary Gunn domain formation at the drain side of the gate rather than to a pinchoff of the conducting channel under the gate. The saturation current, channel conductance, transconductance, charge under the gate, gate-to-source and drain-togate capacitances, cutoff frequency, characteristic switching time, power-delay product, and breakdown voltage are calculated in the frame of this model. The results are verified by two-dimensional computer calculations. They agree well with the results of the computer analysis and experimental data for a 1-µm gate GaAs MESFET. It is shown that a stray gate-to-drain and gate-to-source capacitance sets up a limitation of a gate length which must be larger than or about 0.1 µm for a GaAs MESFET.     

Optical Control of GaAs MESFET's

Theoretical and experimental work for the performance of GaAs MESFET's under illumination from light of photon energy greater than the bandgap of the semiconductor is described. A simple model to estimate the effects of light on the dc and RF properties of MESFET'S is presented. Photoconductive and photovoltaic effects in the active channel and substrate are considered to predict the change in the dc equivalent circuit parameters of the FET, and from these the new Y- and S-parameters under illumination are calculated. Comparisons with the measured S-parameter's without and under illumination show very close agreement. Optical techniques can he used to control the gain of an FET amplifier and the frequency of an FET oscillator. Experimental results are presented showing that the gain of amplifiers can be varied up to around 20 dB and that the frequency of oscillators can be varied (tuning) around 10 percent when the optical absorbed power in the active region of the FET is varied by a few microwatts. When the laser beam is amplitude-modulated to a frequency close to the free-running FET oscillation frequency, optical injection locking can occur. An analytical expression to estimate the locking range is presented. This shows a fair agreement with the experiments. Some suggestions to improve the optical locking range are presented.