This paper reports the results from a study conducted to determine the effect of electron energy on the dose-to-amorphization of Zr3Fe at 23-30 K. Zr3Fe samples were irradiated in the HVEM at Argonne National Laboratory, at energies ranging from 200 to 900 keV. Amorphization occurred at electron energies from 900 keV down to 250 keV. Three distinct regions were observed: between 900 and 700 keV amorphization occurred at a constant low dose of ~ 4 × 1021 e cm-2; a higher plateau at 1022 was observed between 600 and 400 keV, and finally there was a sharp increase in the dose-to-amorphization below 400 keV, so that at 250 keV the necessary dose was an order of magnitude higher than that at 900 keV. In the region below 400 keV there was evidence of a dependence of the dose-to-amorphization on the orientation of the sample with respect to the electron beam. The results can be analyzed in terms of a composite displacement cross section dominated at high energies by displacements of Zr and Fe atoms, by displacements of Fe atoms at intermediate energies and of secondary displacements of lattice atoms by recoil impurities at low energies.

Ion mixing was studied in polycrystalline Nb/V and single-crystal Mo/V superlattices and in bilayers of the same materials. Systematic variation in the modulation wavelength, Λ, revealed a factor of three decrease in mixing efficiency at small Λ for Nb/V. The decrease appears to coincide with a structural transition in which the Nb/V superlattices become coherent. The results are consistent with a diffusion-induced grain boundary migration interpretation of mixing during irradiation.

A dynamic Monte-Carlo simulation (MCS) program, containing not only collisional mixing but also sputtering effects, has been used to elucidate the dynamic mixing processes and the atomic transport of constituents in Al-Pd bilayer systems. MCS results reveal that the preferential inward displacement of the top layer element dominates, and that there is an enhancement of the inward displacement when the heavier element is in the top layer. The inward displacement is controlled by both of an anisotropic and isotropic atomic transport. The anisotropic term is caused by the primary recoil of atoms, which has a characteristic of φ(ion dose) dependence, and the isotropic term is associated with the random cascade motion, which has a dependence. However the outward displacement is governed by only the isotropic motion, thus the inward displacement always dominates over the outward motion, which leads to a preferential displacement of top layer element to the bottom layer.

In the Nb-rich Nb-Fe(Co,Ni) multilayered films, two metastable crystalline (MX) phases of hep and fee structures were formed by room temperature 200 keV xenon ion mixing. The phase formation was through a two-step transition, Le. a first shearing of bcc—>hcp and then a sliding of hep—>fcc. The free energy curves of the observed MX phases were calculated and included into the free energy diagram together with those of the terminal solid solutions, the amorphous phase and the equilibrium intermetaШc compounds. An independent comparative thermal annealing experiment was conducted with the corresponding as-deposited multilayers and the results certified the emerging sequence of the MX phases being exactly the same as the calculated diagram predicted. Moreover, a crystallographic model was figured out for the bec—>hcp—>fcc transition and the resultant lattice parameters agreed well with the experimental ones.

FexAg1-x (x=0.20-0.65) and Fey Cu1-y (y=0.50-0.75) nano-multilayers were prepared by alternative deposition of pure constituent metals. Ion mixing of the multilayers by 200 keV xenon ions at 77K resulted in the formation of a similar metastable phase with a big lattice of cubic structure, when the ion dose exceeded 1×1016 and 1×1015 Xe+/cm2 for Fe/Ag and Fe/Cu systems, respectively. The fraction of the new phase emerged in the alloy films increased with increasing ion dose in both cases.

The development of ion implantation for doping of Si-IC devices is reviewed with a special focus on some of the commercial as well as technological factors.

The study of doping of silicon nano-grains in semi-insulating polycrystalline silicon is of paramount importance since intense room-temperature luminescence at 1.54 µm has been demonstrated in this silicon-based semiconductor when doped with erbium ions. We have investigated the formation of p- and n-type layers of semi-insulating polycrystalline silicon by implantation and diffusion of B, P, As, and Er. The room-temperature resistivity can be changed by more than six orders of magnitude for both p- and n-type doping. A dramatic decrease of resistivity is observed for dopant concentrations above a threshold level; this effect is explained by assuming that the free-carrier motion is limited by grain boundary barriers and the electrical conduction is due to thermionic emission and tunneling of the carriers through the barriers. The prevalence of one mechanism over the other depends upon temperature, oxygen concentration and doping. In the undoped material the barrier height is large (≈ 0.5 eV), but for dopant concentrations above the threshold, it decreases with the doping level. Correspondingly, the conductivity increases by many orders of magnitude. The determination of the threshold value allows the evaluation of donor and acceptor grain boundary trap densities. Diodes have been fabricated by implantation and diffusion of boron and erbium. The I-V characteristics of these diodes are interpreted on the basis of the material modeling.

The charge collection efficiency of a diode with a retrograde well was estimated using focused ion beam irradiation at 400 keV and 2 MeV. The retrograde well was found to effectively suppress a collection of charge carriers created by energetic particles. The charge collection efficiency of the diode with the retrograde well was ~ 25 % lower than that with the conventional well when 400 keV ~ 2 MeV protons were irradiated normal to diodes. This result was in good agreement with device simulation.

The activation kinetics of Si+ and Be+ ions implanted into InGaAsP (λ = 1.3 µm) grown by MOMBE were studied over a variety of doses (5 × 1012 - 5 × 1014 cm-2), annealing times (3-60 sec) and temperatures (575-750°C). Maximum doping concentrations of ~2 × 1019 cm-3 were obtained for both Si+ and Be+, with activation energies for electrical activation of 0.58 eV and 0.39 eV, respectively. Multiple energy F+ or H+ implants can be used to produce high resistance layers for isolation purposes - maximum sheet resistances of ~8 × 106 Ω/□or ~106 Ω/□ for initially p+or n+ InGaAsP, respectively, were obtained for F+ implants followed by annealing near 450 °C. Smooth, anisotropic dry etching of the InGaAsP was obtained with low pressure CH4/H2/Ar discharges with the etch rates being independent of conductivity type or doping levels.

The effect of the Ga dose on the activation of implanted carbon in GaAs is determined. The free hole concentration is found to depend on the depth of the amorphous layer created by the Ga co-implant. Initial results on C implantation in InP indicate the behavior of C is very different in InP when compared to GaAs. The role of precipitation in reducing the activation of C in both GaAs and InP is discussed.

Threshold voltages of Si-implanted gallium arsenide (GaAs) MESFETs depend on not only wafer crystallographic characteristics, but also MESFET fabrication processes. When the position of the orientation flat was changed to improve mechanical shock tolerance, it was found that Vth varied on the wafers due to a crystallographic orientation effect on the Si implantation. Thus, we investigated Vth distributions on wafers of several different Si implantation orientations. It was confirmed that Vth largely varies according to Si-implantation orientations.

Control of threshold voltage during gallium arsenide (GaAs) Metal Semiconductor Field Effect Transistor (MESFET) processing is critical. Channel formation typically is done using ion implantation of 29Si+ from a SiF4 source. The use of Si+ presents a variety of potential cross-contamination problems. 28Si+ and 30Si+ beams can become contaminated with 28N2+, 28CO+, and 30NO+. While 29Si+ is relatively pure, the abundance of 29Si+ in the mass spectrum is 4.67%, thus reducing the potential beam current. This study investigates the effects of varying the mass resolving power of an Eaton 6200AV implanter on the electrical parameters and defect formation. The mass resolving power was adjusted by changing the mean path size through the slit of the aperture opening and magnetic separator current. Electrical device characterization measured a small shift in saturated source-drain current (Idss) and break down voltage, while threshold voltage shifts of approximately 80 mV were observed, with the various mass resolution powers. Transmission Electron Microscopy (TEM) showed that there is minimal change in the extended defect density with changes in isotope and aperture opening. Secondary Ion Mass Spectrometry (SIMS) measured the amount of cross contamination and these results correlated well with the observed changes in device electrical properties.

Manganese ions implantation into ultrapure GaAs layers grown by molecular beam epitaxy were investigated by photoluminescence technique systematically in a wide range of manganese concentration up to 1×1020cm-3. Five shallow emission bands denoted by (Mn°, X), “G”, “G' “ “H” and (D, A)2 are formed in the implanted layers in addition to the well known Mn impurity related emission at ~880nm. With increasing manganese concentration to 1×1019cm-3, “G” exhibits no energy shift, suggesting that “G” is different from the behavior of [g-g] emission that is commonly formed in shallow acceptor (such as C) incorporated ultrapure GaAs. (Mn°, X), “G” and “G' “ present no energy shift with increasing excitation intensity, while “H” and (D, A)2 indicate peak energy shift greatly showing typical donor-acceptor pair characteristics. “G” and (Mn°, X) are found to hold similar radiative origin which is different from “G”. Temperature dependence measurement reveals that emission “G” has a thermal activation energy of 5.4meV.

Piezorcsistive properties have important sensor applications. In this study we find that several ion implanted polymers exhibit piezorcsistive properties. Piezoresistive gauge factors of these films are found to be significantly greater than those of metals, however smaller than those of semiconductors. Potential advantages of using the piezorcsistive properties of these films are also discussed.

Low energy ion bombardment has been utilized to fabricate rectifying contacts on aluminum nitride grown on single crystal silicon substrates. Bombardment of aluminum nitride with methane was followed by sputter deposition of gold contacts. To our knowledge, this is the first report of rectifying contact formation on aluminum nitride. Scanning electron micrographs show that the initially ordered aluminum nitride surface is significantly altered with low energy methane ion beam exposure. Electrical measurements made on samples which had been partially masked during implantation indicate that rectification is a result of the ion bombardment.

Boron depth distributions are reported for MeV implants into silicon through a variety of masking materials. Silicon is implanted with boron through a 0.1-µm-thick layer of thermally grown silicon dioxide. Secondary ion mass spectrometry (SIMS) shows the projected ranges agree within 10% with data reported in the literature and with results from the computer program TRIM. Silicon dioxide, photoresist, and metal layers are used to mask the high-energy boron implants. The SIMS results indicate that TRIM overestimates the energy loss of MeV boron ions as they pass through photoresist and/or silicon dioxide.

Fe ions were implanted into yttria stabilized zirconia with doses of 5×1015 -2×1017 ions/cm2 at an energy of 140 KeV. The effects of Fe-implanted zirconia and post-implantation thermal annealing on electrical properties and structure of an implanted layer were studies by electric measurement, X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD). It was found that the resistivity for the as-implanted sample decreased greatly with increasing ion dose, and at high dose the resistivity decreasing gradually. After annealing at 1200°C in N2 environment, the resistivity relative to that for the as-implanted sample had clearly marked decreases. Results of the XRD and XPS measurements showed that the implanted Fe ion was mainly in a charge state of Fe2+ and the sections of Fe2+ were incorporated into the ZrO2 lattice, forming a FeZrO3 compound. Finally, the conductivity mechanisms were discussed briefly.

The capability of discharge-pumped vacuum ultraviolet F2 molecular laser for laser annealing of heavily ion implanted semiconductor was demonstrated for the first time using Se+ heavily ion implanted GaAs. Cr-doped semi-insulationg GaAs wafers were used as the substrates, and the Se+ implantation energy and dose were controlled to 100 keV and 1× 1015 cm-2, respectively. Samples were annealed using a F2 molecular laser ( wavelength = 157 nm ) with a single pulse ( width ~ 20 ns ) in the energy density range from 200 to 800 mJ/cm2 in a nitrogen atmosphere. In addition, furnace annealing was done on separate samples at 850 ºC for 20 minutes in a purified hydrogen atmosphere for comparison. Characterization of the samples was carried out using Raman scattering and ellipsometry. The laser annealed samples exhibited intense Raman scattering LO phonon peaks whose intensity increased with increasing laser power density, whereas the furnace annealed samples exhibited a very weak LO phonon peak. It was demonstrated for the first time that VUV photons can be very effective in annealing ion implantation damage as compared with conventional furnace annealing. The behavior of Raman scattering spectra as a function of laser energy density was explained quantitatively by a “spatial correlation” model. The model made it possible to estimate the average size of the recovered crystal regions in samples for any given laser energy density.

Oxygen and nitrogen ion implantation have been applied, for the first time, to C-doped Al0.35Ga0.65As layers produce high resistivity regions (ps ≥ 1×1010 Ω/□) that are stable after annealing at 900 ºC. A dose threshold for stable compensation for both O and N ions was found above 8×1013 cm-2 for samples doped at 2×1018 cm-3. Although O implantation has been reported to form stable compensation in Si-doped and Be-doped AlGaAs, the ability of nitrogen implantation to produce thermally stable compensation has not been previously reported and may be due to a C-N complex. The existence of this C-N complex is supported by results for O- and N-implants into C-doped GaAs where N formed thermally stable compensation but O did not. Sheet resistance data versus anneal temperature and estimates of the depth of the defect levels are reported. This result will have application to heterojunction bipolar transistors and complementary heterostructure field effect transistor technologies that employ C-doped AlGaAs or GaAs layers along with high temperature post-isolation processing.

Erbium is a rare-earth ion which, when incorporated in a solid in the trivalent state, shows characteristic intra-4f emission at 15 µm, an important telecommunication wavelength. A review is given of recent work in the area ofEr ion implantation into soda-lime silicate and AI2O3 optical waveguide films, LiNbO3 and silicon. Structural and optical characteristics are discussed and future challenges for this exciting new field are described.