This article presents research results on the formation kinetics and structure of mesoporous silicon layers synthesized by electrochemical anodic treatment in an electrolyte based on a 12 % aqueous solution of hydrofluoric acid. The electrolyte consisted only of deionized water and hydrofluoric acid and contained no organic additives thus avoiding carbon contamination of the porous silicon during anodic treatment. Another distinguishing feature of the work is that all the experiments were conducted for whole silicon wafers 100 mm in diameter rather than for small size samples often used to save silicon. The initial substrates were single crystal silicon wafers brand IES −0,01 cut from Czochralski grown ingots. The thickness of the porous silicon layers, its growth rate and the bulk porosity of porous silicon were estimated as functions of anodic current density and anodic treatment time. The structure of the porous silicon layers and the size and the density of the pore channels investigated using SEM. We found optimum treatment modes allowing one to obtain homogeneous porous silicon layers for subsequent use as buffer layers for epitaxy.

The composition of diffusion silicon layers doped by rare earth erbium was investigated. The diffusion source was an erbium oxide layer on the surface of the test silicon wafer. The erbium and oxygen distribution profile in silicon was measured by SIMS. The concentration of electrically active erbium impurity in the diffusion layers on silicon was determined by measuring the surface resistance and carrier mobility during consecutive etching of layers. The erbium diffusion coefficient at 1240 °C was estimated to be 4.8 · 10−13 cm2 · s−1. A model of erbium and oxygen simultaneous diffusion was suggested. The model takes into account the association of erbium and oxygen into complexes. The results of numerical simulation and experimental data are in a good agreement for the near−surface region of the diffusion layer.

The properties of Chochralski grown [100] undoped GaSb crystals with diameters > 60 mm have been studied. We found that the dislocation density reduction in large undoped crystals can be achieved both by the well known technological approaches and by isovalent impurity (indium) doping. We show that the introduction of two additional annealing stages, one being close to the moment the crystal reaches the target diameter (the length of this stage is 1 hour, the temperature being close to the melting point) and the other being 3 h long post−growth annealing at 650 °C, reduces the dislocation density in ~60 mm diameter crystals to (3—5) · 103 cm−2. We found that dislocations in large GaSb crystals form in two distinct temperature ranges as evidenced by the difference in the morphology of the dislocation traces. The dislocation critical stresses were identified in the 420—690 °C temperature range. It is demonstrated that isovalent doping (In, concentration in the range (2—4) · 1018 cm−3 leads to a significant incr ase in dislocation critical stresses and, accordingly, to a drastic decrease in the average dislocation density to (4—5) · 102 cm−2. This opens new prospects for obtaining large low dislocation density GaSb crystals.

Voltage−current characteristics (VIC′s) were investigated and electric resistivity  and  Hall coefficient measurements  of high−ohmic cadmium telluride samples were made at room temperature. The samples were cut from ingots grown by traveling heater method and  doped with chlorine (2 • 1017 cm−3 in load). Indium and  gold were used as contact materials. Galvanomagnetic measurements  were made at  square−form samples using the  Van der  Pau  method in magnetic fields B ≤ 1.5 Tl. The near−contact regions were illuminated by white light of variable intensity; the  VIC′s were investigated also. In addition, the VIC′s were investigated when the sample was illu- minated by monochromatic 480 nm light through gold−covered sides of the sample. The experiments showed that  illumination  of the  near−contact region leads to a considerable decrease of sample resistance (by 2—3 orders of magnitude). The VIC′s had linear shapes and in most cases came through the  origin of coordinates. With an increase in the light intensity the angle between the X−axis and the VIC straight line increased. Similar results were obtained when monochromatic light passed through gold−covered sides of the sample. We show  that  illumination  of the  near−contact region allows measuring the electrical resistivity and the Hall coefficient of the sample which is impossible without illumination.

We show that the magnetoresistive properties of n−Si/SiO2/Ni nanostructures containing nanogranular nickel pillars in verticals pores of the SiO2 layer differ considerably from those properties of previously studied nanogranular Ni films electrodeposited onto n−Si wafers. The electrophysical properties of these nanostructures are similar to those of a system consisting of two opposite−connected Si/Ni Shottky diodes. We studied the magnetoresistance of these structures in the 2—300 K temperature range and in magnetic fields of up to 8 Tl. The studies suggest that at 17—27 K the structures have a posi- tive magnetoresistive effect the magnitude of which depends on the transverse bias applied to the structure and increases with a decrease in the longitudinal current (along the pillars). At 100 nA current, the relative magnetoresistance in a 8 Tl field increased by 500 to 35,000% as the transverse bias varies from 0 to −2 V. The magnetoresistive effect observed in the structures is likely to be related to the effect of the magnetic field on the impact ionization of the impurities causing an avalanche breakdown of the Si/Ni Shottky diode. We prove the possibility of controlling the magnetoresistive effect in n−Si/SiO2/Ni template structures by applying an additional (transverse) electric field to the nanostructure between the silicon substrate (functioning as the third electrode) and the nickel nanopillars.

 

We show that the optical characteristics of an imperfect photonic crystal can vary significantly due to the transformation of the pho- ton mode spectrum caused by the presence of impurity layers. The photonic mode spectrum has been studied using the model of imperfect superlattice of a one−dimensional crystal with two elements (layers) in the unit cell: the first layer is silicon, and the second one is the liquid crystal. Peculiarities of the dependence of the lowest band gap on the concentration of randomly embedded admixture layers (including plasma layers) in that system have been studied. The theory developed on the basis of virtual crystal approximation allows carrying out numerical calculations of the concentration dependence of the corresponding optical characteristics. This latter advantage significantly expands the possibilities for simulation of similar composite materials with predetermined properties.

 

The operating experience of hydrogenated amorphous silicon (a−Si : H) based solar cells has shown that besides their low efficiency this type of photovoltaics degrade much faster compared to single crystal based solar cells. As far as the processes deter- mining the degradation of amorphous materials based solar cells are not well studied, and the degradation of similar cells without light exposure has also been reported, we conducted an experiment to compare the temporal change characteristics of main solar cell parameters in darkness and under natural light. The demonstration of short circuit current reduction in darkness aged solar cells should be considered as one of the most interesting results of the work. Moreover we have shown that the change of this parameter is on average the same for the illuminated cells, while for some cells short circuit current reduction is substantially higher. This is indicative of the fact that the observed effect is not related to the Staebler—Wronski effect.

The characteristics of low−power and high−power thyristors basen of dislocation−free single crystal silicon doped with ger- manium to the concentration range NGe ~ (0.05—1.5) • 1020 cm−3 have been investigated. The criterial parameters of thyristors exposed to radiation and high temperature gradients have been estimated using experimental data processing methods in the STATISTICA and MathCAD environments. We show the appropriateness of using germanium doped silicon for increasing the thermal stability and radiation strength of the devices exposed to γ−radiation in the range of doses of up to 2.94 • 106 mSv.

A significant dependence of the strain state of GaAs film lattice grown by molecular−beam epitaxy (MBE) on the nucleation method of early GaP buffer layers (50 nm) on the vicinal substrate Si(001) 4° around the <011> axis was discovered. GaP growth started layer−by−layer with a gallium or a phosphorus sublayer. If GaP nucleated with a gallium sublayer, the GaAs film has a significant lattice rotation around the <011> axis. If the buffer starts forming with a phosphorus layer the GaAs film evidently rotates around the <001> axis. The film relaxation degree ex- ceeds 100%, and the film is in a laterally strained state. Analysis was carried out using the triclinic distortion model. A reciprocal space scattering map was obtained using X−ray diffraction in a three−axis low resolution setup. The map clearly shows that the GaAs film lattice is rotated.

A detailed study of Si quantum dots/SiOx film structures synthesized using a new hydrofluoric technology of forming silicon nanoparticles in porous silicon oxide matrices has been performed. A physical mechanism of the effect of chemical treatment in HF vapors in air on the structural and luminescent properties of the film porous systems with nanosized silicon has been suggested. We show that the passivation of the broken bonds on the surface of Si nanoinclusions as a result of the treatment occurs with the participation of oxygen, fluorine and hydrogen atoms, and this effect depletes the nonradiative recombination channel by two orders of magnitude. We suggest a model explaining the blue shift of the photoluminescence spectra as a result of the treatment due to a decrease in the sizes of the Si−QD during the oxidation of their surface layers.

 

A new method of measuring the parameters of shallow and medium−depth levels in semiconductor band gaps has been presented. The method is based on temperature scanning, hardware integration and subsequent differentiation by the duration of the relaxation charge excitation pulse of the energy levels during the application of a small amplitude displacement meander to the barrier structure. Experimental results of research AlGaN/ InGaN/GaN of the structures are resulted.

 

 

 

 

 

 

The method of separate determination of two−pole sample volume resistance and contact resistance is suggested. The method described is applicable to high−ohmic semiconductor samples: semiinsulating gallium arsenide, detector cadmium−zinc tel- luride (CZT), etc. The method is based on near−contact region illumination by monochromatic radiation of variable intensity from light emitting diodes with quantum energy exceeding band gap width of investigated material. It is necessary to obtain sample photo−current dependence upon light emitting diode current and to find the straight−line part of this dependence. The extrapola- tion of this linear part to ordinate axis gives cut−off current value. As bias voltage is known, it is easy to calculate sample volume resistance value. Afterwards, using dark current value, it is pos- sible to determine total contact resistance. The method was approbated on n−type semiinsulating GaAs sample. Contact resistance value was shown to be approximately equal to sample volume resistance. It means that influence of contacts must be taken into account when electrophysical data is analyzed.