Diffusion Coefficient Of Ge Research Articles

Annealing studies of high Ge composition Si/SiGe multilayers

Abstract For the design flexibility of SiGe based quantum cascade lasers high Ge composition Si/SiGe (x = 80%) superlattices are important. We present an X-ray small angle scattering and high-resolution X-ray diffraction study on strain compensated Si/Si1– x Gex multiple quantum well structures with Ge compositions (x up to 80%), grown on Si0.5Ge0.5 pseudo-substrates. To test the temperature stability of such layers occurring in processing steps, in-situ annealing X-ray reflectivity measurements were performed for temperatures up to 810 °C. From the analysis of the reflectivity data, we obtain the layer thicknesses and the interface roughness of the superlattices during annealing. Using a one dimensional diffusion equation, the Ge diffusion coefficient for these conditions was obtained. Furthermore, the strain status and Ge composition in the superlattice structure and in the SiGe buffer before and after annealing were determined from the symmetrical and asymmetrical reciprocal space maps.

(A)thermal migration of Ge during junction formation in s-Si layers grown on thin SiGebuffer layers

ABSTRACTSolid phase epitaxial regrowth (SPER) has been proven to be highly advantageous for ultra shallow junction formation in advanced technologies. Application of SPER to strained Si/SiGe structures raises the concern that the Ge may out diffuse during the implantation and/or anneal steps and thus reduce the strain in the top silicon layer.In the present studies we expose 8-30 nm strained silicon layers grown on thin relaxed SiGe-buffers, to implant conditions and anneal cycles, characteristic for formation of the junctions by solid phase epitaxial regrowth and conventional spike activation. The resulting Geredistribution is measured using SIMS. Based on the outdiffused Ge-profiles the Ge-diffusion coefficient has been determined in the temperature range of 800-1100C from which an activation energy of ∼ 3.6 eV can be deduced. Up to 1050 C, 10 min, even a 30 nm strained film remains highly stable and shows only very moderate outdiffusion.We also have observed a far more efficient, athermal Ge-redistribution process linked to the implantation step itself. This was studied by analysing the Ge-redistribution following an Asimplant (2-15 keV, 5 1014 – 3 1015 at/cm2). It is shown that the energy of the implant species (or more specifically the position of the damage distribution function relative to the Ge-edge) plays a determining factor with respect to the Ge-migration. For implants whereby the damage distribution overlaps with the Ge-edge, a very efficient transport of the Ge is observed, even prior to any anneal cycle. The migration is entirely correlated with the collision cascade and the resulting (forward!) Ge-recoil distribution. The scaling with dose for a given energy links the observed Ge-profile with a broadening mechanism related to the number of atom displacements induced in the sample within the vicinity of the Si-SiGe-transition.

AES study of surface segregation of Ge in amorphous Si 1− xGe x thin film alloys

Surface segregation of Ge in amorphous Si 1− x Ge x thin film alloys was studied by Auger electron spectroscopy technique. The alloys (with Ge bulk concentrations in the range of 18–58 at.%) were prepared by DC magnetron sputtering and annealed in a UHV chamber in the temperature range of 653–673 K. The surface equilibrium data were evaluated using segregation kinetics and the bulk diffusion coefficient of Ge was estimated in the framework of the model developed by Lea and Seah. We show that the experimental data can be interpreted using the McLean–Langmuir isotherm. The estimated segregation energy is lower than the theoretical one calculated in crystalline SiGe alloys neglecting the allowing and the size effects. We also found that the Ge bulk diffusion coefficient shows similar concentration dependence, as observed in amorphous SiGe multilayers.

Microstructure and coarsening kinetics of Ni 3Ge precipitates in aged Ni [sbnd]Ge alloys

The decomposition of supersaturated Ni Ge alloys containing 14.54, 14.97 or 15.44wt.%Ge was investigated at 700 °C. Transmission electron microscopy was used to measure the average particle sizes and particle size distributions, and magnetic analysis was used to measure the reduction of supersaturation with aging time. Coherent precipitates of Ni 3Ge with the ordered L1 2 crystal structure were present from the earliest aging times (10 min) in all three alloys. The average particle size increases with aging time, t, as t 1/3 within the limits of experimental error, consistent with growth by diffusion-controlled coarsening. The particle size distributions are slightly broader than the theoretical distribution of the Lifshitz-Slyozov-Wagner theory. The kinetics of the depletion of supersaturation with aging time area at least consistent with the predicted t −1/3 time law, although the scatter in the data was quite large. There does not appear to be any effect of volume fraction on the coarsening behavior over the range of f′ γ investigated (0.18–0.27). The values of the interfacial free energy, σ, and diffusion coefficient of Ge in Ni, D, derived from the data were85.2 ± 37.5mJ m −2 and2.00 ± 0.48 × 10 −13cm 2s −1, respectively. The large standard deviations are due to uncertainties in the measurements of the kinetics of solute depletion of the matrix. The equilibrium solubility of Ge in the matrix at 700 °C is12.81 ± 0.07wt.% (10.62 ± 0.06 at.%, which is slightly smaller than previously reported values. The Ni 3Ge precipitates are spherical at small sizes and become increasingly cuboidal as they grow. Spatial correlations that lead to rafting in Ni-base alloys containing misfitting γ′ precipitates are evident in Ni Ge alloys as well. However, the spatial correlations do not appear to be as strong, and the cuboidal morphology not as pronounced, as might be expected solely on the basis of the relatively large lattice misfit (δ ≈ 0.0064) in this alloy. Possible reasons for this behavior are discussed.

X-ray reflectometry study of interdiffusion in Si/Ge heterostructures

Interdiffusion in molecular-beam-epitaxy-grown Si-Ge heterostructures has been investigated by grazing incidence x-ray reflectometry. Measurements on Si1−xGex/Si superlattices reveal that the thickness of the Si1−xGex layers increases when intermixing is induced by high-temperature annealing. This results from a strong composition dependence of the Ge diffusion coefficient in Si1−xGex. Experiments on coherently strained superlattices show that the diffusion rate is faster in the early stage of annealing, before the strain is relieved. This suggests that strain enhances interdiffusion at the interfaces. Similar experiments performed on (SimGen)p atomic layer superlattices show that considerable intermixing occurs in these heterostructures even after moderate annealing treatments (i.e., 20 s at 700 °C). Due to the very low diffusion rate of Ge in Si-rich Si1−xGex alloys, interdiffusion is more pronounced in structures of smaller periodicity or in which the interfaces are initially intermixed.

Comments on the diffusion of Ni and Ge in nickel

The self-diffusion coefficient of Ni varies by eight orders of magnitude in the temperature range 879-1670.4 K. This variation can be well described by the elastic and expansivity data of the bulk material in the framework of a model suggested recently by Varotsos and Alexopoulos (1986). The same model can account for the diffusion coefficient of Ge in Ni which varies by six and a half orders of magnitude in the temperature range 940-1676 K.

Diffusion of Ge in GaAs at SiO2-encapsulated Ge–GaAs interfaces

The diffusion of Ge in semi-insulating GaAs was investigated for annealing temperatures in the range 650–800 °C. The samples consisted of electron-beam or metalorganic chemical-vapour deposition deposited Ge–GaAs interfaces encapsulated with SiO2. Ge diffusion profiles, measured with secondary-ion mass spectroscopy (SIMS), had a characteristic plateau and steep slope at the diffusion front, similar to those obtained for Si and Sn diffusion at SiO2-encapsulated GaAs. A saturated-surface concentration of 2 × 1019 atoms∙cm−3 was observed for anneals above 750 °C, likely the solid solubility of Ge in GaAs at these temperatures. Based on the Greiner–Gibbons substitutional-pair diffusion model (Appl. Phys. Lett. 44, 750 (1984)), the Ge-pair diffusion coefficients were calculated from the SIMS data. They ranged from 1 × 10−13 to 3 × 10−12 cm2∙s−1 but varied depending upon the annealing time and sample structure. A new model that eliminates the necessity for paired-Ge diffusion was introduced. It proposed that dissociation and out-diffusion of GaAs into the SiO2 produced a nonequilibrium flux of vacancies into the GaAs. Calculated profiles using a position-dependent Ge diffusion coefficient were shown to fit the SIMS data independent of the type of Ge diffusing species, whether single ions or pairs.

Growth of lead‐germanium‐telluride thin film structures by molecular beam epitaxy

The development of Pb1−xGexTe devices would extend the total wavelength range that can be covered by tunable diode lasers. Pb1−xGexTe was grown from PbTe, GeTe, and Te sources at 2 μm/h and in vacuums <10−8 Torr. The Ge concentration as a function of depth in the grown layers was measured with Auger electron spectroscopy in combination with argon ion sputtering. GeTe segregates on the surface for substrate growth temperatures of 220 °C and below, and preferentially evaporates at 330 °C. Heat treatment of PbTe/Pb1−xGexTe superlattice structures with 200 A periodicity yields a Ge diffusion coefficient of 2×10−18 cm2/s at 270 °C. The effect of lattice mismatch on crystal quality was determined from etch pit density studies. PbTe/ Pb1−xGexTe/PbTe structures with thin intermediate layers had much lower etch pit densities than thicker films, in good agreement with theory.

Interface properties of oxidized germanium-doped silicon

Czochralski-grown Si-Ge single crystals, with several atomic per cent of Ge, were investigated from the aspects of their oxide interface properties and their suitability for MOS device making. It was found that the oxide formed is almost pure SiO2, but the C(V) curve of both n- and p-type MOS capacitors is shifted in the positive direction by an amount dependent on the oxidation rate and the Ge diffusion coefficient in Si, thereby showing the existence of a negative interface charge. The surface state density function was measured using the high-frequency and the quasistatic capacitance methods and was found to have more peaks, located near the middle of the gap, than the normal Si function. The states were mainly of the acceptor type. Their number, established by the variable temperature method or by the length of the flat region (`ledge') of the C(V) curve which forms at liquid nitrogen temperature, was one to two orders of magnitude larger than in Si but could be much reduced by annealing. The capture cross section, measured by the conductance method, was larger than in Si and varied by jumps along the band gap. MOS transistors of the n-channel enhancement type were made and were found to have a relatively low mobility.