In this article the balance of polysilicon «supply−demand» has been investigated and an attempt of its forecast through 2018 is made. An assessment of the condition of the solar power industry and world production of polysilicon for 2014 is given. Now it is recognized that the capacities of polysilicon producers worldwide exceed the demand but what will be consumption the next years and when overproduction «will be settled» — there is no concerned point of view. Large vendors who make polysilicon with low manufacturing cost are dominating. The first top 10 vendors — Hemlock, REC, OCI, Wacker, GCL, TBEA XinJiang Silicon, LDK, Daqo New Energy, Tokuyama and SunEdison (ex−MEMC) — have a total production capacity of about 250 KMT. There is also an area of uncertainty in which there are about 80—90 producers with various degree of readiness of production, various reached efficiency and productivity. The description of today’s global polysilicon market is given, including technology assessments, supply capabilities, manufacturing costs, and silicon utilization trends, as well as pricing and supply−demand outlook through 2018. We considered 3 potential scenarios so that to outline area of the most probable development options. It is established that, in the absence of macroeconomic delay of world economy, the probability is high that overproduction of polysilicon will be offset by the growing consumption and demand for new polysilicon production capacities will return in the next 3—4 years. 

Elements III and V of groups of the Periodic System and carbon are the most important impurities in silicon. The estimation technique of carbon, boron and phosphorus impurity content in high−purity monosilane has been proposed. The technique involves the preparation of polycrystalline silicon by silane decomposition, growing a test single crystal by the floating zone melting method and analysis of single crystal samples by IR spectroscopy. Calculation of impurity concentration in polycrystalline silicon were performed using results on their content in the test single crystal samples, data on impurity distribution of in the liquid−solid system and sample coordinates along the ingot length. Effective impurity distribution coefficient in the «solid−liquid» system for specific growing conditions have been calculated using the Burton−Prim−Slichter equation. 

Results for the test silicon samples with natural isotopic composition and the enriched 28Si isotope obtained from monosilane samples with different impurity contents have been reported. Results of IR spectroscopic research of impurity composition for the test silicon single crystal are in agreement with the concentration data obtained by chromatography. The concentration of III and V group impurities in monosilane were in the range 4 ⋅ 10−9—2 ⋅ 10−6 at. %, and for carbon 2 ⋅ 10−6—6 ⋅ 10−4at.%. The measurement uncertainty by IR spectroscopy method for carbon impurity does not exceed 15 %, for boron and phosphorus — 20%. We show that the upper limit of carbon content in monosilane detected using this method is determined by its solubility in silicon, while the bottom limit depends on the detection accuracy of the IR spectroscopy technique and possible background contamination.

We studied experimentally and theoretically the possibility to obtain a uniform single crystal of SiGe alloy enriched at the Si side. The content of the second component in a crystal 15 mm in diameter and 40 mm in length grown by the modified floating zone technique from the charge of 79.8 at.% Si and 20 at.% Ge composition with 0.2% B admixture has been investigated using selected area X−ray analysis in different points and in line scanning mode along and across the crystal axis. The longitudinal changes in the germanium concentration of proved to be well described by the analytical equation previously derived for conditions of Sb (Ga) doped Ge growth from a thin melt layer in the presence of a heater submerged into the melt. For a more accurate description of the experimental data we made allowance for the change in the melt layer thickness between the growing crystal and the bottom of the submerged heater. The lateral distribution of the second component, not exceeding 5% over a diameter of the crystal, can be significantly improved by reducing the curvature of the phase interface during the growth. 

Using step−by−step removal of silicon layers, in which dislocation−related photoluminescence is observed after Si+ (100 keV, 1 ·1015 cm−2) ion implantation followed by high−temperature annealing in a chlorine−containing atmosphere, it has been found that a majority of dislocation−related centers of luminescence at ~ 1,5 μm (D1 line) is localized at the depths of Si+ ion ranges. Cross−sectional electron microscopy shows that the dislocations introduced by the implantation treatment (implantation plus annealing) penetrate to depths of ~ 1μm. A phenomenological model of the D1−line dislocation−related luminescence is developed based on the assumption that the K−centers and modified A−centers located in the atmospheres of dislocations are responsible for this luminescence line. The temperature dependence of luminescence intensity calculated on the basis of the model fits well the experimental data for the D1 line.

Convective heat transfer and solidification have been studied using a simplified but unified simulation/experimental model of the Czochralsky method for two materials with melting points close to room temperature: heptadecane (low heat conductivity) and gallium (high heat conductivity). Due to the transparency of the heptadecane melt we have been able to visualize the melt flow patterns and the solidified structures in a laboratory experiment to provide the simulation model with source data. Based on calculations we have studied the parameters of melt flow patterns, heat flows on the cooled disc and the dependence of solidification front shape for both materials on convective heat transfer modes: thermogravity and mixed (i.e. with additional crystal rotation) convection.

The use of por−Si electrodes promotes the separation of water molecules inside por−Si nanopores and efficient evolution of hydrogen during water electrolysis. The por−Si/c−Si heterostructure allows solving one of the problems of water photoelectrolysis on silicon electrodes, i.e. their energetic insufficiency. Combined electrochemical and physical deposition of Ni on the surface of por−Si, formation of NiSi−silicide coatings on the surface of the pores and subsequent production of por−Si photoelectrodes based on the NiSi/por−Si/c−Si/Al heterostructure results in an improvement of their corrosion resistance to oxidation and anodic dissolution, an increase in efficiency of hydrogen generation and enhancement of the photoelectrodes’ lifetime..

In this paper a new technique for synthesis of porous silicon layers with silver nanoparticles based on the method of low−energy and high−dose metal ion implantation into Si is proposed. For demonstration of this technique, room temperature Ag+ ion implantation of polished Si wafer with ion energy of 30 keV, ion dose of 1.5 ⋅ 1017 ion/cm2 and ion current density of 8 μA/cm2 was carried out. By high resolution scanning electron and atomic−force microscopy, electron probe microanalysis and Raman scattering we have shown that as a result of ion implantation a thin amorphous layer of porous Si is formed on the surface of irradiated Si with average pore sizes of 150—180 nm, pore depth of about 100 nm and wall thickness between pores of about 30—60 nm. Moreover, porous Si contains Ag nanoparticles with sizes of 5—15 nm. We established that during ion implantation the sputtering of Si surface by Ag+ ions occurs which was not observed before. On the basis of these data we concluded that the proposed physical technique for porous Si formation compared to chemical techniques could be integrated into an advanced process of fabrication and improvement of electronic circuits based on industrial ion implantation. 

In this work the investigations of technology, morphology, electric and photoelectric properties of the silicon photosensitive structures have been represented. The structures included layers of the silicon carbide and the porous silicon. The porous layer was formed on the surface of the single crystal silicon substrates by the method of electrolytic etching in fluoride containing solutions. Plates with different microrelief surface (polished, honed, textured) were used. Carbidization of the samples leading to the formation of heterostructures on SiC/Si was conducted by the method of gas endotaxin in a hydrogen stream in a vertical reactor with cold walls and a graphite container. The structure and composition of the SiC/Si heterostructures on the different surface structures of poly− and single crystal silicon, including surface of porous silicon layer have been investigated. We show that in the process of endotaxy of all types of surfaces forms a single crystal silicon carbide phase of cubic modification. The morphology of the resultant structures has been investigated by scanning and transmission electron microscopes. Different filiform formations were found on the pore−free surface, which are identified as silicon carbide, and the cylindrical or conical structures of the unclear nature were observed on the porous surface. The current−voltage and current−power curves have been plotted for all types of the structures, the general appearance of which indicates the presence of several potential barriers. The photoelectric properties of the structures have been analyzed along with the prospect of their use in solar cells.

New solutions for tensors of effective pyroelectro−magnetic properties of piezoactive composites on the basis of the solution of the boundary value problem of electro−magnetic elasticity have been obtained. For the solution of the boundary value problem, new solutions for singular components of the second derivative functions of Green for displacements, electric and magnetic potentials in homogeneous transversal isotropic piezo electro−magnetic medium with ellipsoidal grain of heterogeneity have been used. Calculation results of the concentration dependences for effective coefficients of pyromagnetic and electromagnetic coherence of ferrite/barium titanate composite with ellipsoidal, spherical, fibrous inclusions for various polydisperse structures and of composite with layered structure have been presented. Considerable influence of the shape of inclusions, features of relative positioning and inversion of properties of phases on the effective coefficients of pyromagnetic and electromagnetic coherence of the composite material have been revealed. The conclusion is drawn on the preferable use of the pyroelectric phase as spherical inclusions, and ferrite as the composite matrix. This allows for more than a five−fold increase in the effective constant of pyromagnetic coherence of the composite material in comparison with its value for the same structure but with inversion of properties of phases for constant volume fractions of the ferrite and pyroelectric phases.