Enhanced activation of Zn-implanted GaAs

Abstract

Limitations on high level doping have been investigated for implanted Zn in GaAs. Fast diffusive redistribution during the annealing of heavy dose Zn implants generally leads to broader doped layers of lesser concentrations. Though such a redistribution can be prevented by short duration annealing of ∼1 s, this alone is not sufficient to increase the peak concentration. Significantly better activation can be obtained if an excess of As is also provided. It is found that coimplanting As with Zn in addition to short duration annealing provides layers with peak doping concentrations increased to levels approaching 1020 cm−3. Doping enhancement related to encapsulation and the outdiffusion of Ga into SiO2 has also been observed.

Annealing Behavior of GaAs Ion Implanted with p‐Type Dopants

“…Severe film growth suppression and inhomogeneity have been observed for arsine and phosphine at very low concentrations, indicative of a strong interaction of these dopant species with the chemistry of silane pyrolysis (1-4, 6, 7). Similar anomalous results have been found in the production of semi-insulating polysilicon (SIPOS), where oxygen may play a similar role to arsine and phosphine in the in situ doping of polycrystalline silicon (8)(9)(10)(11)(12).…”

“…Previous studies (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11) on annealing behavior of gallium arsenide (GaAs) wafers ion implanted with p-type dopants such as Be, Mg, Zn, and Cd have generally shown that the diffusive redistribution of the p-type dopants during annealing is quite complex and that this redistribution depends on a number of parameters. Some of these parameters, such as annealing temperature and time, annealing environment, dopant concentration, ion dose, which controls the lattice damage, and the type of GaAs, are well known and easily characterized, while others, such as the role of the encapsulant in case of capped annealing, dislocation density of GaAs, and the rate of temperature rise during the annealing cycle, are not so obvious.…”

“…Recent studies (3)(4)(5) of annealing of Zn-implanted GaAs encapsulated in silicon nitride show that significant in-diffusion resulted in considerable broadening of the dopant profile compared to the as-implanted profile during annealing at or above 850~ This profile broadening was observed to be as much as three to four times the as-implanted profile, i.e., up to 0.8-1.0 ~m deep into GaAs. Some limited recent investigations (6)(7)(8) of capless rapid thermal annealing (RTA) of Zn-implanted GaAs reported negligible in-* Electrochemical Society Active Member.…”

Co-implantation of Al+, P+, and S+ with Si+ implants into In0.53Ga0.47As

“…11 Specifically, in the case of group VI dopants, it was proposed that the introduction of excess group III species would promote the creation of more group V vacancies to enhance group VI donor solubility. 12,[30][31][32][33] This same idea was applied to amphoteric dopant species with the thought that Si or Ge could be made more n-type in GaAs or InP by the implantation of As or P to promote creation of group III vacancies to increase occupation of group III sites with Si resulting in donors. [13][14][15][16]18,20,21,[34][35][36] The addition of group V species in the case of co-implantation with amphoteric dopants was also anticipated to reduce the propensity of group IV to occupy group V sites or the formation of IV-IV next nearest neighbor pairs and limit electrical activation.…”

High doping level by rapid thermal annealing of Mg-implanted GaAs/GaAlAs for heterojunction bipolar transistors

“…Other studies have reported on Zn and Mg distributions in GaAs [3]- [6]. We have investigated Mg implantation because of its low diffusion coefficient and high solubility in GaAs.…”