V. S. Speriosu

X‐ray rocking curve analysis of superlattices

We present detailed analyses of x-ray double-crystal rocking curve measurements of superlattices. The technique measures depth profiles of structure factor, and profiles of perpendicular and parallel strains relative to the underlying substrate. In addition to providing a detailed picture of the state of stress, the profiles are a direct measure of the composition modulation. The thickness of the period of modulation and the average strain are determined with a precision of ~1%. The detailed structure of the period is determined to ~5%. We obtain an expression relating the structure of the rocking curve to the structure of the period. This expression allows analytic determination of the structure without Fourier transformation or computer fitting. We show the influence of small random fluctuations in layer thicknesses and strains. The technique is applied to a 15-period GaAlAs/GaAs and a ten-period AlSb/GaSb superlattice grown on GaAs and GaSb substrates, respectively. In the former, the thickness of the period was 676 A and the perpendicular strain varied between zero for the GaAs layer and 0.249% for the layer with peak (93%) Al concentration. Transition regions, ~100 A thick, with continuously varying composition, were found between the GaAs and the Ga0.07 Al0.93As layers. Fluctuations in structural properties were less than 5% of the average. The AlSb/GaSb superlattice had a period of 610 A with sharp transition regions between the layers and negligible fluctuations from period to period. The perpendicular strains were −0.03% and 1.25%, respectively, for the GaSb and AlSb layers. A uniform parallel strain of 0.03% was found throughout the superlattice. Nonzero parallel strain indicates that a small fraction of the misfit between the superlattice and the substrate is plastically accommodated by net edge dislocations lying in a narrow region (a few hundred A thick) at the interface with the substrate. The net number of edge dislocations was calculated to be ~1×10^4/cm^2. The measured perpendicular strains were in excellent agreement with the values calculated from bulk lattice parameters, elastic properties, and the parallel strain. For both superlattices, the standard deviation of random atomic displacements away from perfect crystal sites was below 0.1 A, in agreement with reported ion channeling and electron diffraction measurements of superlattices. The rocking curve method is a major tool for quantitative analysis of superlattices.

X‐ray rocking curve and ferromagnetic resonance investigations of ion‐implanted magnetic garnet

Detailed analyses of x-ray rocking curves and ferromagnetic resonance spectra were used to characterize properties of -oriented Gd, Tm, Ga:YIG films implanted with Ne+, He+, and H + 2 . For each implanted species the range of doses begins with easily analyzed effects and ends with paramagnetism or amorphousness. Ion energies were chosen to produce implanted layer thicknesses of 3000 to 6000 A. Profiles of normal strain, lateral strain, and damage were obtained. The normal strain increases with dose and near amorphousness is 2.5%, 3.4%, and 3.9% for Ne+, He+, and H + 2 , respectively. Lateral strain is 0 for all values of normal strain, implying absence of plastic flow. Comparison of these results with the reported decrease in lateral stress implies either a large reduction in Youngs modulus or a transition to rhombohedral equilibrium unit cell. Damage is modelled by a spherically-symmetric Gaussian distribution of incoherent atomic displacements. Due to the use of (444), (888), and (880) reflections the sensitivity is greatest for the c sites occupied by Gd, Tm, and Y. The standard deviation of displacements increases linearly with strain with proportionality constant 0.25, 0.18, and 0.13 A/% for Ne+, He+, and H + 2 , respectively. For maximum strains up to 1.3% annealing in air reduces the strain without changing the shape of the profile. The behavior of the strain with annealing is nearly independent of implanted species or doses. After annealing at 600 °C the strain is 40% of the original value. Magnetic profiles obtained before and after annealing were compared with the strain profiles. The local change in anisotropy field DeltaHk with increasing strain shows an initially linear rise for both He+ and Ne+. The slope is −4.1 kOe/%, in agreement with the magnetostriction effect estimated from the composition. For strain values between 1 and 1.5%, DeltaHk saturates reaching peak values of −3.6 kOe for He+ and −2.8 kOe for Ne+. At strain values near 2.3% for He+ and 1.8% for Ne+, DeltaHk drops to nearly 0 and the material is paramagnetic. For peak strains greater than 1.3% for He+ and 1.1% for Ne+ the relation between uniaxial anisotropy and strain is not unique. The saturation magnetization 4piM, the ratio of exchange stiffness to magnetization (A/M) and the cubic anisotropy H1 decrease with strain reaching 0 at 2.3% and 1.8% for He+ and Ne+, respectively. At these strain values the damping coefficient alpha is 50% and 80% greater than bulk value for He+ and Ne+, respectively. For higher observed strains the material remains paramagnetic. Upon annealing of samples implanted with low doses of Ne+ and He+ the anisotropy field follows uniquely the behavior with strain for unannealed material. At 600 °C the magnetization returns to bulk value but the ratio A/M remains 20% low. For H + 2 implantation the total DeltaHk consists of a magnetostrictive contribution due to strain and of a comparable excess contribution associated with the local concentration of hydrogen. The profile of excess DeltaHk agrees with calculated LSS range. The presence of hydrogen results in a reduction of 4piM not attributable to strain or damage. For a peak strain of 0.60% and a peak total DeltaHk of −4.5 kOe, the magnetization is only 40% of bulk value. After annealing up to 350 °C the excess DeltaHk diminishes and redistributes itself to the regions neighboring the peak damage. At 400 °C the excess is nearly 0. For higher annealing temperatures the only component of DeltaHk is magnetostrictive. At 600 °C, the magnetization, the ratio A/M, and alpha return to bulk values.

Strain in GaAs by low‐dose ion implantation

The production of strain in (100) GaAs by low-dose ion implantation has been investigated. Implantations were conducted at room temperature with ions of He, B, C, Ne, Si, P, and Te. Energies were between 100 and 500 keV, and each species was implanted over a range of doses sufficient to create perpendicular strain below 0.3%. The perpendicular strains epsilon [perpendicular] were measured by x-ray double-crystal diffractometry about the (400) Bragg condition. Detailed depth profiles of epsilon[perpendicular] were obtained by fitting the resulting rocking curves with a kinematic model for the diffraction. For all implantations the maximum in the epsilon[perpendicular] distribution was found approximately from the separation of the lowest-angle prominent oscillation from the substrate peak. The depth profiles of perpendicular strain had the same shape as the calculated profiles of energy deposited per ion by nuclear collisions, FD. The maximum perpendicular strains scaled linearly with the dose phi of the implanted ions for all ion species. Also the ratio of maximum strain to dose was found to vary linearly with FD over more than 2 orders of magnitude in FD. We therefore conclude that epsilon[perpendicular]=KphiFD at all depths, where K is a constant. The value of K was found to be (5±1)×10^−2 A^3/eV. Our results suggest that this holds for any ion species in the mass range 4–128 amu, with energy in the hundreds of keV, implanted into (100) GaAs at room temperature, provided the maximum strain is less than 0.3%.

X‐ray study of low‐temperature annealed arsenic‐implanted silicon

Low-temperature anneals (500–650 °C) of 2, 4, and 8×10^15 cm^−2 As+ implanted in silicon at 50 keV were studied by x-ray double crystal diffraction. The rocking curves were analyzed by a kinematical model. Two regions of strain were found in the solid-phase epitaxially regrown layer. One layer was uniform and positively strained. The other was nonuniform and negatively strained. By comparing rocking curves of repeatedly etched layers it was found that the surface layer is negatively strained, corresponding largely to the substitutional As in the regrown layer. The positively strained region lies at the interface between the implanted layer and the undamaged silicon substrate.

Nonlinear strain effects in ion‐implanted GaAs

The nonlinear production of strain in (100) GaAs by room-temperature ion implantation has been studied. Ions of Ne, Si, and Te were used, with energies of 300, 300, and 500 keV, respectively. Doses ranged up to those required for amorphization. Strains were monitored by x-ray double-crystal diffractometry. Rocking curves were recorded about the (400) Bragg condition and detailed depth profiles of strain perpendicular to the sample surface, epsilon[perpendicular](x), found by fitting the rocking curves with a kinematic model. These were compared with calculated profiles of the density of energy deposited in nuclear interactions, rhoE(x). Rocking curves were also recorded about the (422) Bragg condition for selected samples, to monitor strain in the directionparallel with their surfaces. At low doses, epsilon[perpendicular](x) is a linear function of rhoE(x). At doses sufficient to create strains exceeding about 0.3%, strong nonlinearities are evident and strain profiles depart significantly from the rhoE(x) curves. For the Ne and Si implantations, the profiles tend to saturate at 0.4%–0.5% over a depth of ~4000 A. At higher doses a narrow (~2000 A), sharply peaked region develops, with strains up to 1.5%. At still higher doses this region becomes amorphous. The Te-implanted samples do not experience appreciable saturation; rather a sharply peaked profile develops, and grows with dose to amorphicity. Curves of epsilon[perpendicular] vs rhoE were extracted by comparison of epsilon[perpendicular](x) and rhoE(x) profiles. These demonstrated that for each ion species epsilon[perpendicular] is a unique function of rhoE at all depths. Although this function has the same general form for all three implantations, the curves differ from species to species. Above epsilon[perpendicular]=0.3%, epsilon[perpendicular] increases sublinearly with rhoE for all three implanted ions. For Ne and Si, epsilon[perpendicular] becomes almost constant at 0.4%, beginning at rhoE~0.15 eV/A^3. The strain epsilon[perpendicular] starts increasing again with rhoE at about 0.7 eV/A^3 for Ne and 0.3 eV/A^3 for Si, until the GaAs goes amorphous. The curve for Te shows only a slight inflection at epsilon[perpendicular]~0.3%, continuing to increase with rhoE to amorphicity. Parallel strains in the Si-implanted samples were not more than 0.02% at all values of rhoE.

Interfacial strain in AlxGa1−xAs layers on GaAs

Detailed analysis of x‐ray rocking curves was used to determine the depth profile of strain and composition in a 2500‐A‐thick layer of AlxGa1−xAs grown by metalorganic chemical vapor deposition on 〈100〉 GaAs. The x value and layer thickness were in good agreement with the values expected from growth parameters. The presence of a transition region, 280 A thick, was detected by the rocking curve. In this region, the Al concentration varies smoothly from 0 to 0.87. Measurement and control of the sharpness of such interfaces has important implications for heterojunction devices.

Analyses of metalorganic chemical‐vapor‐deposition‐grown AlxGa1−xAs/GaAs strained superlattice structures by backscattering spectrometry and x‐ray rocking curves

Backscattering spectrometry with channeling and x-ray rocking curves have been employed to analyze metalorganic chemical-vapor-deposition-grown AlxGa1−xAs/GaAs strained superlattice structures in significant detail. Both techniques complement each other in the precise determination of composition, thickness, and strain in the individual layers of the superlattices. In addition, the sensitivity of the two techniques allows quantitative measurements of transition regions at the interfaces of various layers. Such fine probing into thin layered superlattice structures provides essential feedback in controlling their growth.

Annealing behavior of neon‐implanted magnetic garnet

Ferromagnetic resonance spectra and x-ray rocking curves were used to measure the change in magnetic properties and strain with annealing temperature in the surface layer of (111)-oriented Gd, Tm, Ga substituted yttrium iron garnet films implanted with Ne+ at 190 keV. For doses below about 4×10^14 ions/cm^2, the entire implanted layer remains crystalline and magnetic. The implantation-induced strain decreases monotonically with increasing annealing temperature, falling to zero at a temperature of 1100 °C. The implantation-induced magnetic anisotropy varies with strain in the same manner as for unannealed material until the annealing temperature reaches 800 °C. For higher temperatures, the anisotropy has a value larger than that expected for unannealed material. At a higher dose, 5×10^14 ions/cm^2, the center of the implanted region is both amorphous and nonferrimagnetic. Single-crystal order and ferrimagnetism return with annealing near 500 °C. The magnetization and exchange constant decrease with increasing dose, and annealing at 1100 °C restores them to bulk values.

Annealing behavior of hydrogen-implanted magnetic garnet

Ferromagnetic resonance spectra and x-ray rocking curves were used to measure magnetic and strain profiles of Gd, Tm, Ga substituted yttrium iron garnet films implanted with H 2 + at 120 keV and at doses in the range (3–80)×10^15 ions/cm^2. The maximum strain occurred at a depth of 3600 A, reaching a value of 2.9% at 40×10^15 ions/cm^2. At the highest dose, the garnet was amorphous in that region which had highest strain at lower dose. The strain has two components, one due to damage and one due to the presence of the hydrogen atoms. The second component disappears for annealing temperatures above 400 °C, at which temperature the hydrogen has been reported to be largely desorbed. The reduction of the first component with annealing follows the same pattern as other implant elements. The magnetic anisotropy exhibits a large anomalous nonlinear increase with dose. The excess over other implantation elements disappears for annealing temperatures above 400 °C. There is no significant change in gyromagnetic ratio with dose or annealing temperatures up to 600 °C.

Abstract: Comparison of magnetic and crystalline profiles in He+-implanted Gd,Tm,Ga:YIG

Profiles of magnetic and crystalline properties in He+-implanted -oriented Gd,Tm,Ga:YIG were obtained using ferromagnetic resonance and x-ray diffraction techniques, respectively. Implantation was done at room temperature several degrees off axis. One series of samples was implanted with 140 keV He+ at 1.5, 3.0, 4.5, and 6.0×10^15 at. cm^−2. Another series consisted of three profiles obtained with single, double, and triple energies and doses in the range 30–140 keV and 9.0×10^14–3.0×10^15 at. cm^−2. For the highest dose the maximum changes in magnetic properties were: −3200 Oe in uniaxial anistropy Hk, a 55% decrease in magnetization M, a 65% decrease in exchange constant A, a 20% increase in damping parameter alpha, and no change in gyromagnetic ratio gamma. The maximum perpendicular strain epsilon[perpendicular] was 1.26% with a corresponding maximum rms random atomic displacement of 0.23 A. Lateral strain epsilon|| was zero. No evidence was found for creation of extended defects. The highest dose was, at most, half of that required to produce amorphousness. Both DeltaHk and epsilon[perpendicular] had a significantly sublinear dependence on dose. The ratio of maximum DeltaHk for the 6.0 and 1.5×10^15 at. cm^−2 was 1.8 instead of 4.0. The corresponding ratio for strain was 2.6. The difference in sublinearity indicates that DeltaHk and epsilon[perpendicular] are not linearly related through magnetostriction. The equation relating DeltaHk, M, and epsilon[perpendicular] can be preserved only by invoking implantation-induced changes in Youngs modulus, Poissons ratio, or the magnetostriction.