Comparison of Group IV and VI Doping by Implantation in GaAs

Empirical models have been developed for predicting the carrier distributions after ion implantation and annealing process steps used in normalGaAs device fabrication. The present models predict the carrier distributions for n‐type normalGaAs using Si and Se as the dopants. The simulation is done by using the atomic profile and activation efficiency models developed in this study. These models are based on SIMS and measured carrier concentration profile studies reported in the literature. The model for atomic distribution provides a gaussian profile for Si and a joined half‐gaussian one for Se using the projected range and standard deviation parameters obtained from SIMS studies. The effect of radiation enhanced diffusion is taken into account. Impurity diffusion during annealing for temperatures up to 920°C for Si and 1000°C for Se is neglected. Volumetric concentration has been properly chosen as a canonical parameter for obtaining activation efficiency. The universal activation efficiency plots for each dopant species give percentage activation efficiency as a function of atomic concentration (cm−3) at different annealing temperatures.

Section: Silicon Activation Efficiency--the Activation Efficiencymentioning

confidence: 99%

Section: Silicon Activation Efficiency--the Activation Efficiencymentioning

confidence: 99%

Process Modeling of n‐Type Doping in Gallium Arsenide

Dhiman

Wang

1984

“…Therefore it is necessary to encapsulate the sample with a suitable dielectric layer or to perform the annealing in a carefully controlled ambient. There have been numerous reports on the encapsulants of Si-implanted GaAs, that is, large doping efficiency using pyrolytic Si3N4 passivating film (7), superiority of SiO2 passivating film (8), and capless annealing without using dielectric encapsulant (9). However, few investigations have been reported on the effects of encapsulation on the carrier concentration profiles of Si-implanted GaAs.…”

mentioning

confidence: 99%

Study of Encapsulants for Annealing Si‐Implanted GaAs

Onuma

Hirao

Sugawa

1982

GATE dently make an estimate of the accuracy of the final solution.Computers.--Computations were performed on the Cray-1 at Bell Laboratories, Murray Hill. A Honeywell 6000 was used as the host computer to the Cray-1. A DEC Vax was used for file maintenance: files were sent from the Vax to the Cray-1 via the Honeywell; output files were returned in the same manner to the Vax. Calculations typically took between 150 and 300 sec of Cray-1 time. ABSTRACTThe effects of encapsulation on the carrier concentration profiles of Si implanted into GaAs have been studied. Large differences were found in the carrier concentration profiles among the materials used for the annealing cap. Diffusion of Si was enhanced by SiO2 encapsulation but was negligibly small with Si3N4 encapsulation and capless. ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 128.192.114.19 Downloaded on 2015-06-19 to IP

“…Implantation is considered to be an attractive alternative to diffusion for the creation of thin heavily doped n-type regions in GaAs (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15). We have utilized ion implantation to fabricate high-low READ GaAs IMPATT diodes (6) and state-of-the-art Ka-band GaAs Gunn effect diodes (7).…”

mentioning

confidence: 99%

“…These include sealing the GaAs both with and without a dielectric encapsulant in evacuated quartz ampuls prior to annealing (4,8). The encapsulants have included SiO2 (3,4,6,9), Si3N4 (8,9), A12Oa (11,12), and Ga-doped SiQ (13,14). However various problems associated with uncontrolled contamination and interdiffusion at the encapsulant-GaAs interface have been noted (15).…”

mentioning

confidence: 99%

Annealing of Ion‐Implanted GaAs in a Controlled Atmosphere

Malbon

Lee

Whelan

1976

Implantation is considered to be an attractive alternative to diffusion for the creation of thin heavily doped n-type regions in GaAs (1-15). We have utilized ion implantation to fabricate high-low READ GaAs IMPATT diodes (6) and state-of-the-art Ka-band GaAs Gunn effect diodes (7). However, the results reported to date indicate that, in general, the doping efficiencies of the n-type implanted species are low unless long annealing periods are used. Lengthy annealing times are unattractive because diffusion tends to excessively broaden the impurity doping gradients and thus impair device performance. Even for short anneal times within the range of commonly used annealing temperatures, 700~176it is necessary to take special precautions to minimize erosion of the GaAs surfaces. A variety of approaches have been used. These include sealing the GaAs both with and without a dielectric encapsulant in evacuated quartz ampuls prior to annealing (4, 8). The encapsulants have included SiO2 (3, 4, 6, 9), Si3N4 (8, 9), A12Oa (11, 12), and 14). However various problems associated with uncontrolled contamination and interdiffusion at the encapsulant-GaAs interface have been noted (15).This paper describes the results of substituting a controlled atmosphere for the dielectric encapsulant to protect the GaAs surface during high temperature anneals. Epitaxial materials technology has demonstrated that polished substrates can be maintained at elevated temperature with minimal surface degradation in a flowing high purity hydrogen atmosphere. Taking advantage of this experience, an anneal system was designed which utilizes a controlled atmosphere of high purity hydrogen (He) in conjunction with an arsenic (As) source to protect the GaAs surface during annealing. The system was used to anneal epitaxial GaAs films implanted with either silicon ions or sulfur ions to create thin, highly doped n-type layers. The epitaxial surfaces exhibited no degradation after anneals at 800~ for 20 min. The electrical results indicate that the apparent electrical conversion efficiencies achieved for the implanted layers were as high as 85% for the 800~ 20 min anneal. ExperimentalThe GaAs epitaxial material used in this investigation was grown by the He-Ga-AsCts vapor phase technique on heavily Te-doped n+-substrates oriented 3 ~ Off the <100> toward the <111>. The epitaxial layers were doped with sulfur and were electrically uniform in depth with net donor concentrations within the range 2-8 X 1015 cm -~. The epitaxial layers were implanted at room temperature with either 120 keV silicon ions or 140 keV sulfur ions. In the case of the silicon implants it was concluded after careful analysis of the separated ion spectrum that the implanted species was 28Si+ with less than 1% of other possible molecular components (e.g., N2+). To minimize channeling effects, the GaAs lattice was oriented to appear random to the incident ion beam. Under this condition, a first order approximation to the implanted ion distribution is Gaussian with a mean projected range Rp...