Zinc Diffusion Research Articles

Point Defect Properties from Metal Diffusion Experiments — What Does the Data Really Tell Us?

Point defect properties, including diffusivities and equilibrium concentrations for both interstitials and vacancies, are commonly extracted from metal diffusion experiments, and these values are widely used in process simulation software. However, in many cases, these parameter values were extracted using oversimplified models which ignore interactions between interstitial and vacancy diffusion mechanisms. Questions about the accuracy of these parameters have come from ab-initio defect calculations which conclude that vacancies diffuse faster than interstitials, in contrast with published reports on metal diffusion which find vacancies diffuse much more slowly than interstitials. We have reanalyzed published data for zinc and platinum diffusion and find that it is possible to match all of the data using fast vacancy diffusivity. The most direct evidence for slow vacancy diffusion (and a high equilibrium concentration) comes from platinum diffusion experiments. However, we are able to reproduce these results with fast V diffusion and carbon/interstitial clustering, using carbon concentrations typical of Czochralski and float zone silicon (1016cm−3). We evaluate the effectiveness of metal diffusion experiments in determining point defect parameters, and find that it is not possible to reliably determine both diffusivities and equilibrium concentrations for both interstitials and vacancies from metal diffusion results.

Features of radiative transitions of charge carriers in Al x Ga1−x As and GaAs x P1−x p-n-structures

The AlxGa l_xAs and GaAsxPi-x compounds have formed the basis for the development of highly efficient light-emitting diodes for the visible and IR regions of the spectrum. The position of the edge luminescence band in these materials is controlled by the quantity x without a noticeable change in the efficiency of electroluminescence (EL). The losses occurring in this case are caused not only by radiationless transitions of nonequilibrium carriers, but also by rec. mbination in an undesirable spectral regior,. B~I ..... we present results that reveal such a channel of recombination and indicate that under certain conditions its role in the occurring processes can be substantial. We also investigated the EL spectra with reverse-biased diode structures, which indicate the dominant role of hot carriers in the formation of such radiation. The objects of investigation were epitaxial AlxGat-xAs and and GaAsxPl-x p-n structures�9 The layers of AIxGal-xAs of the n- and p-types were obtained by the method of liquid epitaxy on GaAs substrates with a hole concentration of about 1018 cm -3. The width of the forbidden band Eg in the p-type AI0.1sGa0 8sAs layers was 1.66 eV at 300 K, and the hole concentration in it was equal to about 5.1017 cm -3. The n-type AIo4Ga0.6As layers h;ad a large fordidden band (1.95 eV, 300 K), while ~hc electron concentration was close to 2-1018 cm -3. Mesa diodes with an area of 0.4  0.4 = 0.16 mm 2 were formed by chemical etching in a 5H2SO4:2H202:IH2 0 solution at 75-80~ [1 ] using photolithography. Ohmic contacts with both n- and p-regions were created by chemical deposition of Au-Ni-Au with subsequent burning in a flow of dried hydrogen at 420-450~ The lateral and frontal surfaces of the mesa structures were protected by dielectric layers of A1203 and SiOz obtained by pyrolytic deposition in a vacuum chamber. The n-type GaAsxPI-x layers (x = 0.15, Eg = 2.18 eV at 300 K~, which were doped with tellurium up to 1015-1016 cm -3, were obtained by gas-phase epitaxy on n-type GaP substrates with a carrier concentration of 1016-1017 cm -3. The p-n junct,ons were produced by zinc diffusion (to a concentration of 10 )8 cm -3) into an epitaxial n-layer. The use of an Au-Bi (2~o) alloy for ohmic contacts made possible chemical etching of the diodes (in a 3H2SO4:iH202:IHz O solution at 60~ to remove the disturbed laver on the side surface of the p-n-junction after mechanical cutting of the wafer into diodes measuring 0.4  0.4 mm ]2 1. We performed spectral measurements on a setup consisting of a double quartz monochromator with a photomultiplier (1.7-4.0 eV) and a cooled PbS photoresistor r eV) as detectors. The setup was carefully calibrated for spectral sensitivity by a certified tungsten ribbon lamp. Figure 1 presents the spectra of injection EL for the p-n structures under investigation with forward bias. The characteristic feature of these spectra is the presence of wider-band radiation in the IR region (hv < 1.8 eV) along with a dominating shortwave band in the visible spectrum region. As we can see from Fig. 1, the contribution of the IR band to the total radiation of the AlxGa l_xAs p-n heterostructure changes substantially depending on

Zinc diffusion in tellurium doped gallium antimonide

Zinc diffusion into tellurium doped gallium antimonide, GaSb, has been carried out as a function of time, temperature and antimony over-pressure. Total zinc profiles as well as carrier concentration profiles have been measured. Results indicate an inverse dependence of the diffusivity on antimony over-pressure and favour an interstitial-substitutional vacancy [F.C. Frank and D. Turnbull, Phys. Rev. 104 (1956) 617] or kick-out [U. Gösele and F. Morehead, J. Appl. Phys. 52 (1981) 4617] mechanism. Furthermore, at high zinc concentrations, the profiles indicate an additional component associated with a non-electrically active zinc species which has a small, strongly temperature dependent diffusion coefficient.

Zinc diffusion in tellurium doped gallium antimonide

Zinc diffusion into tellurium doped gallium antimonide, GaSb, has been carried out as a function of time, temperature, and antimony over-pressure. Total zinc profiles as well as carrier concentration profiles have been measured. Results favor a substitutional-interstitial vacancy (Frank-Turnbull)1 or kick-out (Gosele-Morehead)2 mechanism, although there is insufficient evidence to conclusively distinguish between them. There is also an inverse dependence of the diffusivity on antimony over-pressure, this is discussed in terms of zinc diffusion superimposed on gallium vacancy diffusion. Tellurium doping seems to have little effect on the diffusion because of its low level in comparison to that of zinc. Furthermore, at high zinc concentrations, the profiles indicate an additional component associated with a non-electrically active zinc species which has a small, strongly temperature dependent diffusion coefficient.

Defect development and dopant location due to elevated temperature implantation of InP with MeV zinc ions

Zinc implantations in the MeV energy regime at temperatures of about 200°C within the dose range of 5 × 10 14–1 × 10 16 cm −2 were carried out. The investigations included RBS and PIXE measurements combined with ion channeling in the major crystallographic axes and additionally SNMS and XTEM. The implantation-induced damage is characterized by mobile point defects at the elevated implantation temperature. This results in damaged layers containing point-like defects, but far from amorphization. During rapid thermal annealing the surface up to about 0.6 R p recovered channeling-perfect, while in a depth of (1–2) R p a band of extrinsic dislocation loops was formed. Comparing the experimental with the calculated PIXE minimum yields, conclusions about the zinc positions were drawn: Due to the low ZnK aligned yield nearly all zinc atoms occupy substitutional lattice sites in the as-implanted samples. After annealing a remarkable diffusion of zinc combined with lattice site changes is observed.

Study of Zinc-Delta-Doped Strained Quantum Well InGaAs/GaAs p-Channel Heterostructure Field-Effect Transistors

Zinc-delta-doped strained quantum well InGaAs/GaAs p-channel heterostructure field-effect transistor (HFET) grown by low-pressure metalorganic chemical vapor deposition has been realized. In terms of using the zinc-delta-doping technique by interrupting the growth, high concentration is carried out. The influence of spacer layer thickness on two-dimensional hole gas concentration and hole mobility is also studied. The delta-doped p-type strained InGaAs/GaAs HFET with 8 nm spacer layer thickness exhibits extrinsic transconductances of 15 mS/mm at 300 K and 24 mS/mm at 77 K. Secondary-ion mass spectrometry (SIMS) is carried out to confirm the film quality, zinc diffusion profile and sharpness of the proposed structures. Due to the smaller mobility difference in InGaAs/GaAs than in using AlGaAs barriers, the gate voltage swing in the p-channel delta-doped InGaAs/GaAs HFET exhibits a value about 1.7 V. Consequently, a wider linear amplifier application is expected in the present device.

Open tube zinc diffusion into GaAs0.8P0.2 using AlN and SiNx cap films

The open-tube diffusion of zinc in GaAs0.8P0.2 from a zinc-doped silica film was investigated in detail. Aluminum nitride (AlN) and silicon nitride (SiNx) films were used as the anneal caps. The dependence of diffusion depth on the thickness of an AlN-cap was found to differ from its dependence on SiNx-cap thickness. The selective masked diffusion of zinc using an AlN diffusion-mask was also systematically studied. The diffusion depth in selective masked diffusion was found to depend on both AlN-cap thickness and AlN-diffusion-mask thickness. The experimental results suggest that diffusion depth is not necessarily governed by either cap thickness or diffusion-mask thickness. From this standpoint, the role of film stress on diffusion depth was then quantitatively investigated. It was found that diffusion depth can be scaled well with total film stress in the measured film-thickness range. In this sense, it can be concluded that total film stress is the primary factor that determines the diffusion depth under the measured diffusion conditions.

Electroluminescence from forward-biased Er-doped GaP p-n junctions at room temperature

The electroluminescent properties of Er-doped GaP light emitting diodes are reported. Diodes were prepared by metalorganic vapor phase epitaxy and zinc diffusion. Strong characteristic Er3+ emission at 0.80 eV (1.54 μm) under forward bias is detected at room temperature. Minimal thermal quenching of the electroluminescence was observed. The luminescence intensity depended linearly on current density for low applied current densities, and saturated at a current density of 17 A/ cm2

High-resolution transmission electron microscopy of structural changes at and near surfaces of ZnTe crystals

Structural changes on and near surfaces of ZnTe crystals were observed by the surface profile imaging method in a high-resolution transmission electron microscope (HRTEM) with 200 kV accelerating voltage. It was found that ZnO particles form under electron irradiation on surfaces originally covered with a thin amorphous layer. For amorphous coverages thicker than about 7 nm, ZnO particles were not observed. Under the ZnO particles, 2×1 and 2×2 superstructures (about 10 nm in size) were often found in the ZnTe matrix. These superstructures can be described as a metastable Zn2Te3 having a structure similar to the ordered Ga2Te3 structure. This change can be accomplished by considering the diffusion of zinc from the ZnTe matrix to the surface to form the oxide and evaporation of tellurium out of the surface, leaving behind a matrix with an altered stoichiometry. The HRTEM computer-simulated images for the proposed Zn2Te3 superstructure matched very well the experimental images. The 2×1 and 2×2 superstructures corresponded to the HRTEM images of the new Zn2Te3 structure viewed along its [211] and [100] directions, respectively.

Properties of intrinsic point defects in silicon determined by zinc diffusion experiments under nonequilibrium conditions.

The present paper deals with self- and foreign-atom diffusion processes in Si. We focus on the foreign-atom Zn whose diffusion behavior is shown to be influenced by intrinsic point defects like Si self-interstitials (I) and vacancies (V). Diffusion experiments with Zn in dislocation-free Si were carried out between 1208 and 870 \ifmmode^\circ\else\textdegree\fi{}C applying a special method to perform isothermal diffusion anneals as short as a few seconds. Concentration-depth profiles measured with the help of spreading-resistance analysis are completely described by simultaneous diffusion via the kickout and dissociative mechanism. The evolution of Zn diffusion with time is characterized by short-, intermediate-, and long-time diffusion regimes. Profiles belonging to the long-time regime are suitable to extract transport capacities of intrinsic defects given by the product of thermal equilibrium concentration and diffusion coefficient. Zn profiles after intermediate diffusion times are shown to be sensitive to the prevailing thermal equilibrium concentration of Si self-interstitials. Results are presented for the thermal equilibrium concentrations of self-interstitials and vacancies as well as their transport properties. From the temperature dependence of these quantities the formation and migration enthalpy of I and V are obtained. These data are compared to previous experimental results as well as theoretical calculations.

Use of zinc diffusion into GaAs for determining properties of gallium interstitials.

The fast diffusion of Zn into GaAs has recently been attributed to a minor fraction of Zn interstitials changing over to Ga sites thereby producing interstitial Ga (${\mathit{I}}_{\mathrm{Ga}}$). This kick-out reaction provides the possibility to determine ${\mathit{I}}_{\mathrm{Ga}}$ transport properties from Zn diffusion experiments in virtually perfect GaAs but previous attempts were frustrated by diffusion-induced generation of microstructural defects acting as ${\mathit{I}}_{\mathrm{Ga}}$ sinks. The present study prevents such defect formation by utilizing Zn-doped GaAs powder as the diffusion source. Measured two-stage profiles show that under these conditions Zn diffusion at 906 \ifmmode^\circ\else\textdegree\fi{}C is controlled by ${\mathit{I}}_{\mathrm{Ga}}^{3+}$ in addition to ${\mathit{I}}_{\mathrm{Ga}}^{2+}$. Analysis of the profiles yield quantitative data on Ga- and Zn-related diffusivities, concentrations of ${\mathit{I}}_{\mathrm{Ga}}$ as well as the corresponding electronic transition energy.

Transverse junction waveguide InGaAs/InP photodiode

A novel transverse InP/InGaAs/InP pn junction waveguide photodiode is demonstrated using in situ zinc diffusion before a regrowth step inside an organometallic vapour phase epitaxy reactor. The detector exhibits high speed, low leakage current, low capacitance and high breakdown voltage characteristics.

Preparation and chemical analysis of high purity iron-zinc alloys

We have produced and chemically analyzed a series of high purity iron-zinc alloys with iron concentrations in the range 4-27 wt. %. The technique involved slow diffusion of high purity zinc with small particle iron powder. We produced alloys within the four main iron-zinc phases, Zeta, Delta, Gamma-1, and Gamma to aid in the identification of the intermetallic phases formed in commercially produced galvanneal steel coatings. The diffusion technique produces iron-zinc alloys which are suitable as instrument calibration standards for galvanneal producers and users. A chemical titration technique which measures iron concentration to within 0.5 wt. % was also refined. In determining the bulk iron concentrations of the samples, we have compared the accuracy of the wet chemical technique with Induction Coupled Plasma spectroscopy. Homogeneity was measured using an electron microprobe and a scanning transmission electron microscope. The data show that samples are homogeneous to greater than 98% of the mean iron concentration.

Accumulation of arsenic and zinc in the rhizosphere of wetland plants

Oxidation of the rhizosphere by wetland plants leads to precipitation of iron oxyhydroxides (ironplaque) in the rhizosphere and on the root surface of the plants. Arsenic and zinc have a high binding affinity for iron oxyhydroxides and were found to accumulate in ironplaque on roots of Aster tripolium L. It was argued that rhizosphere oxidation and formation of an ironplaque would favor the accumulation of iron, arsenic and zinc in the rhizosphere. Oxidation of ferrous iron to its ferric form would lead to precipitation of iron oxyhydroxides in the rhizosphere, which in turn would lead to a decreasing concentration gradient of dissolved iron towards the plant roots. The iron oxyhydroxides in turn would bind arsenic and zinc, again creating a decreasing concentration gradient of both elements towards the roots. These gradients would lead to the diffusion of iron, arsenic and zinc in the direction of the roots. Assuming that uptake of the elements by the roots is slower than supply through diffusion, an increase in concentrations of all three elements would be expected to occur in the solid phase of the rhizosphere. The plant-induced processes in wetland soils may not only affect the availability of zine and arsenic to plantsmore » but also may play an important role in the sink function of wetlands for metals and metalloids. This paper reports the results of a pilot field study which compared concentrations of the metals in ironplaque, rhizosphere soil and bulk soil under Spartina anglica and Halimione portulacoides. 20 refs., 3 figs., 2 tabs.« less

An asymptotic analysis of the kick-out diffusion mechanism

The kick-out model for impurity diffusion in semiconductors is studied. The kick-out mechanism is thought to play an important role in a number of applications, including the diffusion of zinc and chromium in gallium arsenide. Asymptotic solutions are derived for both one- and two-dimensional surface source problems. In the one-dimensional case, a mechanism for the destruction of self-interstitials is also incorporated. The calculated diffusion profiles have shapes which are typical of diffusion systems in which the kick-out mechanism is believed to be active. For the two-dimensional problem, contours of constant impurity concentration are calculated and some are found to have the ‘bird's beak’ shape which is frequently observed in experiments.

Unintentional zinc diffusion in inp pn-homojunctions

Unintentional zinc diffusion into uniformly Si-doped InP layers has been studied. The sharp non-error function Zn concentration profiles and inflections are shown to be consistent with the substitutional-interstitial mechanism when the influence of the electrostatic field of the pn-junction on diffusion is taken into consideration.

Studies on the diffusion of zinc and iodine into CdTe

Studies on the diffusion of iodine and zinc into CdTe are reported. Each iodine profile was divided up into four distinct regions and described mathematically by a function consisting of the sum of four complementary error functions. When plotted on an Arrhenius graph, the diffusivities gave four straight line relationships with similar slopes and the Arrhenius parameters for the fastest component of D01 = (7 ±3). 10-11 cm2 s-1 and E1 = (0.21 ±0.05) eV. When extrapolated down to 20°C this gave a diffusivity of 10-14 cm2 s-1 indicating that when iodine is diffused from the vapor it is not suitable as a long term stable dopant in devices where sharp impurity profiles are required. In the case of the zinc diffusions, each profile can be divided into two parts and was fitted satisfactorily by the sum of two complementary error functions giving two values of the diffusivities: Dslow due to zinc diffusion into the slice from the vapor and Dfast due to interdiffusion between a surface layer of Znx Cd1-x Te formed on the slice and the remaining CdTe.

Zinc diffusion in InP from spin-on films of various zinc concentrations

SiO2 spin-on films of various concentrations of 1, 2, 6, 22 and 36 at.% zinc have been fabricated and used for the p diffusion into undoped n-type InP. An increasing zinc concentration (N0) of the spin-on film leads to a higher atomic zinc concentration (NZn) and diffusion depth in InP. From the experimental results a distribution coefficient NZn/N0 of 0.012 has been determined. A higher atomic zinc concentration in InP results in a lower acceptor concentration. With additional heat treatment the electrical activity of zinc can be increased significantly.

Preparation of p-type GaAs layers for ohmic contact

The open-tube diffusion of zinc is used for the preparation of p-type layers in gallium arsenide. The process provides the capability of forming reproducible deep diffusions (0.4–2 Μm) over short diffusion times and in the temperature range 600–650 ‡C. TiPdAu contacts are prepared on these layers. Using cw laser annealing we have obtained specific contact resistance 6.5 × 10−7Ωcm2.

Point Defect Supersaturation During Zinc Diffusion into InP

AbstractFormation of defects during Zn diffusion into undoped and Fe-doped InP single crystals at 700°C has been observed by transmission electron microscopy for various diffusion conditions. The observations are correlated with Zn concentration profiles obtained by electron microprobe measurements and secondary-ion mass spectrometry. The results allow the conclusion that indiffusing interstitial Zn can occupy In sublattice sites via a kick-out reaction. Under appropriate diffusion conditions supersaturations of In self-interstitial atoms result leading to defect formation. Observations in Fe-doped InP suggest that Zn also replaces Fe on In sublattice sites leading to redistribution and to precipitation of Fe.