Si Ingots Research Articles

Friction and wear properties of αFeSi2–Si alloy, ReSi1.8 and MoSi2 in ethyl alcohol

In this study, Fe–Xat% Si alloy (X=70.5, 80.0 and 96.0), Re–64.3at% Si and Mo–66.7at% Si disk specimens were prepared by spark plasma sintering, and their friction and wear properties were investigated when they were slid against Si3N4 ball specimens in ethyl alcohol. The friction and wear properties of Si ingots were also examined. Fe–70.5at% Si, Fe–80.0at% Si, Fe–96.0at% Si and Re–64.3at% Si disk specimens exhibited friction coefficients as low as 0.15. It is considered that the low friction of the Fe–70.5at% Si, Fe–80.5at% Si and Fe–96.0at% Si disk specimens was due to the formation of low friction silicon alkoxide and polyoxysilane on the worn surfaces of the disk specimens and the paired ball specimens. Re–64.3at% Si disk specimens exhibited the highest microvickers hardness of all the disk specimens prepared in this study. In addition, the microvickers hardness of the Fe–Xat% Si (X=70.5, 80.0, 96.0 and 100) disk specimen increased with increasing the Si content. Moreover, it was difficult to obtain dense Fe–90.0at% Si disk specimens by sintering the annealed and crushed Fe–90.0at% Si powder. However, dense Fe–96.0at% Si disk specimens could be obtained by sintering the Fe–90.0at% Si powder at 1403K.

On the detachment of Si ingots from SiO2 crucibles

In a directional solidification furnace, ingots of silicon are contained in a silica crucible covered with a silicon nitride (Si3N4) powder layer. During the crystallization of the ingot, the layer may suffer various behaviors. When properly applied, it can efficiently act as a mechanical shunt between the Si and the silica wall. However, filling of the crucible with silicon pieces can also scratch locally the layer, resulting in deleterious adhesion between ingot and crucible. The layer can also be too thin, or insufficiently pyrolyzed, so that the Si interacts with the silica crucible.An extended simulation is carried out; taking into account the thermo-elastic stresses induced by the difference in thermal expansion coefficients of the crucible and ingot materials that are attached on small areas. The model compares the remaining elastic energy to the adhesion energy between the silicon and the crucible, in order to predict the temperature at which the Si detaches from the crucible in case of sticking. Two configurations are compared in terms of amount of stresses and extent of the ingot volume likely to be impacted by dislocation generation.

Spectrometric pairs on a unique single-crystal basis in experiments on small-angle neutron scattering

Spectrometric pairs have been fabricated on a unique basis in various combinations of scattering geometry—according to Bragg, Laue, and their combination—to develop and improve the technique for applying a double-crystal spectrometer from a highly perfect single-crystal Si ingot. Possible schemes of their using in combination with a crystal-analyzer in experiments on small-angle neutron scattering have been proposed.

Effect of Crystallinity on Residual Strain Distribution in Cast-Grown Si

We report the correlation between the crystallinity of ingots grown by the directional solidification technique and the residual strain and dislocation distribution. It was found that mono-like ingots have a 20–25% higher averaged residual strain than multicrystalline Si ingots grown under the same conditions. However the existence of local high-strained areas, which were frequently found in multicrystalline Si ingots, is reduced in mono-like Si ingots. In addition, a reduction in dislocation density was observed. This effect and the decrease in local high strain could be attributed to the decrease in grain boundaries in the mono-like ingots.

Effect of Insulators on the Power Consumption and the Oxygen Content of Si Ingots in Czochralski Growth

The effect of insulators in a Czochralski furnace on the power consumption and the oxygen content in the melts as well as in the Si ingots was studied without considering a radiation shield. The top insulator showed better results than the other two insulators for the power consumption and Si melt temperature. The interface at the center of the melt descended in the all conditions of installing the insulators, and all interfaces were concave down. The oxygen content in the Si crystal decreased as the thickness of insulators was decreased for the top and top-side insulators, while increased for the bottom insulator.

Measuring Stress in Si Ingots Using Linear Birefringence

We report near infrared (NIR) linear birefringence measurements on shaped and surface finished Czochralski (CZ) - grown silicon ingots. The linear birefringence was measured using a commercially available NIR system. All measurements were done with the NIR beam configured parallel to one of the <100> directions of silicon. Stress- optic coefficient values of silicon were used to calculate the residual stress from phase retardation values. Two- dimensional scans were performed along flat surface, i.e. the crystallographic [001] and [010] directions. Examples of both, higher and lower stressed Si ingots are presented. Due to the large diameter of silicon ingots, stress resolution and stress detection limits are much enhanced for ingots than for silicon wafer. Phase retardation measurements for most ingots showed only low stress distribution of <30kPa. However, some ingot sections show stress as high as nearly 1.0MPa.

Recycling of p-type mc-si Top Cuts into p-type mono c-Si Solar Cells

Solar cell results and material analysis are presented of 2 p-type Czochralski (Cz) ingots pulled from a charge consisting of 100% and 50% recycled multicrystalline silicon top cuts. The top cuts were pre-cleaned with a dedicated low energy consuming technology. No structure loss was observed in the bodies of the ingots. The performance of solar cells made from the 100% recycled Si ingot decreases towards the seed end of the ingot, which could be related to a non-optimal pulling process. Solar cells from the tail end of this ingot and from the 50% recycled Si ingot demonstrated an average solar cell efficiency of 18.6%. This is only 0.1% absolute lower than the efficiency of higher resistivity reference solar cells from commercially available wafers that were co-processed.

Cell Performance Prediction based on the Wafer Quality

Among many other parameters, wafer quality plays an important role in determining the efficiency of a solar cell and the power output of a module. Researchers have spent a lot of efforts in developing new casting methods to produce low-cost high-quality multicrystalline Si wafers. However, multicrystalline Si ingot technology is, still at this stage, a crystallization process that is difficult to control and produces silicon wafers with a wide range of quality. There is a rising need to develop in-line techniques to accurately assess wafer quality. In this paper, a new wafer characterization method is presented and the results of research to find the correlation between wafer quality and cell performance are reported. Based on the analysis of photoluminescence (PL) of as-cut wafers, and taking the cell process into account, a new method to predict the cell efficiency is demonstrated. The objective of the proposed method is to be able to sort, with a simple in-line technique, multicrystalline as-cut wafers for the reason that high- quality wafers are more suitable for high efficiency processing.

S133012 Siのエッチング援用スライスにおける切断特性の向上([S133]先進砥粒加工技術)

Si wafers for solar cells are fabricated by slicing ingot using mechanical slicing method such as multi-wire sawing. Recently, the slicing method for achieving low kerf loss and thin wafer without generating subsurface damage is strongly required because of the increasing demand of low-cost Si wafers. The etching-assisted slicing method is abrasive free machining process employing chemical etching of Si. The method can achieve a lower kerf loss than the mechanical slicing and damage-free Si surface. In this paper, the removal characteristics of Si ingot are discussed under the different slicing conditions.

Conceptual design of a small nuclear reactor for large-diameter NTD-Si using short PWR fuel assemblies

A design of a small nuclear reactor for a large-diameter NTD-Si using a conventional Pressurized Water Reactors (PWR) full-length assembly was proposed in previous works. The height of the full-length assembly was 400 cm, and the overall size of the reactor and reflector around the core became large. In addition, the irradiation channel became very long, making handling of the Si ingots in the channel more difficult. The use of a short PWR fuel assembly, with a height of 100 cm, was considered in the current work. With the shorter assembly, the design of the reactor became compact and more practical. Gd2O3 and control rods were used to suppress excess reactivity. Criticality, neutron transport, and core burn-up calculations were performed using the MVP/GMVP II code and MVP-BURN code. Steady-state single-channel thermal hydraulic analyses were also performed. The calculation results showed that the reactor could be critical over 1200 days, and that heat removal from core was possible under 1 atm operating pressure. Large-diameter ingot up to 20 cm in height could be doped with sufficient uniformity. The reactor semiconductor production rate was estimated, and varied between 48 tons/year and 70 tons/year for the 50 Ω cm target resistivity depending on the position of the control rod.

Improvement of Minority Carrier Lifetime by HCN Treatments

We have developed a method of improving minority carrier lifetime by use of HCN treatments. Metal contaminants, Fe, Ni, and Cu, are introduced during slicing Si ingots to produce Si wafers, and also during alkaline etching processes for removal of a damaged layer and formation of mat-textured surfaces. The concentrations of metal contaminants are greatly decreased by cleaning with HCN solutions having pH of 12, while they are decreased only slightly by pH 9 HCN solutions. Two kinds of metal contaminants, i.e., contaminants present deeply in saw damaged regions and on the surface due to re-adsorption during alkaline etching, can be removed by pH 12 HCN solutions while only the latter can be by pH 9 HCN solutions. The minority carrier lifetime is increased by removal of a damaged layer by alkaline etching but decreased below that of as-sliced wafers after formation of mat-textured surfaces. The lifetime is increased by the HCN treatment and it is attributed to i) removal of metal contaminants and ii) passivation of interface states by the formation of Si-CN bonds from Si dangling bonds.

Record Efficiency of PhosTop Solar Cells from n-type Cz UMG Silicon Wafers

Highly purified n-type Upgraded Metallurgical Grade (UMG) silicon carries a large potential for high efficiency low cost solar cells. In this study, the industrially producible “PhosTop” solar cell concept is employed to manufacture large-area n-type rear junction solar cells from such a Si material with a screen-printed Al-alloyed full-area emitter featuring a selective phosphorous front surface field (FSF) and a SiO2/SiNx:H passivation on the front.Since resistivity at the seed end is about seven times as high as at the tail end of the UMG Si ingot and carrier lifetime decreases from seed to tail end, a clear dependence of the solar cells’ IV characteristics on the original position of the corresponding wafers in the UMG Si ingot is observable. Maximum conversion efficiency is reached (on a wafer which has been taken out at about one fifth of the ingot's length distant from the seed end) by η = 19.0% being, to the authors’ knowledge, the highest efficiency so far reported on industrial type solar cells manufactured from 100% UMG Si.

Toward the realization of an ultra-high-quality multi-crystalline Si ingot for solar cells

Toward the realization of an ultra-high-quality multi-crystalline Si ingot for solar cells

다결정 Si ingot 응고 시 도가니 열전도도 및 Ar 유입량 변화에 대한 열유체 해석

This study presents the results on the changes of crucible thermal conductivity and inflow of Ar, and constructed the mathematical model about heat transfer into furnace. As process variables, simulation model was designated thermal conductivity of crucible to <TEX>$0.5W{\cdot}m^{-1}{\cdot}K^{-1}$</TEX>, <TEX>$1W{\cdot}m^{-1}{\cdot}K^{-1}$</TEX>, <TEX>$2W{\cdot}m^{-1}{\cdot}K^{-1}$</TEX>, <TEX>$4W{\cdot}m^{-1}{\cdot}K^{-1}$</TEX>, and inflow rate of Ar to 15 L/min, 30 L/min, 60 L/min. Initial condition and boundary condition were set respectively in two terms of process. Each initial conditions were set up by the preceding simulation of heat and fluid flow. The primary goal is the application of unidirectional growth of Si ingot using the result. In the result of the change of heat conductivity of crucible, the higher thermal conductivity of crucible shows the shorter solidification time and the bigger temperature difference. And the flow patterns are changed with the inflow rate of Ar. Finally, we found that the lower crucible's thermal conductivity, the better crucible is at polycrystalline Si ingot growth. But in case of Ar inflow, it is hard to say about good condition. This data will be evaluated as useful reference used in allied study or process variable control of production facilities.

Effect of silicon/crucible interfacial energy on orientation of multicrystalline silicon ingot in unidirectional growth

We investigated the effects of the Si crystal/crucible and Si melt/crucible interfacial energies on the orientation of the nucleus of multicrystalline Si during unidirectional growth. We calculated the Gibbs free energy upon nucleation on the crucible by considering the shape of the nucleus of the Si crystal and found that the nucleus with the 〈111〉 or 〈100〉 upper direction is stable. It was experimentally shown that {111} planes are dominant at the bottom of a multicrystalline Si ingot grown at a low rate, which was explained by the difference between the Si crystal/crucible and Si melt/crucible interfacial energies at nucleation.

Evaluation of residual strain in directional solidified mono‐Si ingots

AbstractCast‐grown mono‐like Si ingots for photovoltaic application are getting increasing attention due to the possibility of obtaining highly efficient solar cells at a low‐cost production process. The reduction of crystallographic defects is essential to reach that potential. In this study, the residual strain and dislocation distribution in monolike ingots grown by directional solidification was experimentally determined. It was found that both strain and dislocations are mainly concentrated in the periphery of the ingots (© 2013 WILEY‐VCH Verlag GmbH &amp; Co. KGaA, Weinheim)

Effect of the Silicon Nitride Coating of Quartz Crucible on Impurity Distribution in Ingot‐Cast Multicrystalline Silicon

Multicrystalline silicon ingots were fabricated via the directional solidification method in a vacuum induction furnace using silicon nitride (Si3N4) as quartz crucible coating. The effect of Si3N4 coating on the impurity content, transference, and distribution in Si ingots was investigated. The results can be summarized as follows: The impurities contained in Si ingots, including Fe, Ca, Mn, Cu, P, and B, were found to decrease at varying degrees when the inner surface of the quartz crucible was coated with Si3N4. A mixed layer composted by Si3N4 and Si was observed on the surface of the Si ingot. This mixed layer provided microdefects for the nucleation of metallic impurities. The Fe and N contents in the Si/Si3N4 mixed layer changed with the same tendency, and FeSi2 particles formed on the Si/Si3N4 crystalline boundary. Both Fe and Ca precipitated in the SiC particles near the Si ingot/Si3N4‐coating interface.

Electrical and optical characterization of extended defects in silicon mono‐cast material

AbstractMono‐cast Si growth is currently a very promising approach to optimize the cost per watt in the production of PV devices, simultaneously increasing the installed energy/m2 ratio. However, due to the novelty of this growth approach, the material properties have not yet been studied in detail. In this work, by combining PL imaging with electrical characterization techniques (LBIC and EBIC), both the large scale and the local properties of the wafers grown from mono‐cast Si ingots have been analyzed. PL imaging shows two kinds of defects, which were studied in detail by LBIC. Electrical active defects are found in areas corresponding to a unique large grain, assigned to sub‐GBs, decorated by impurities. These defects have strong charge trapping, with deleterious consequences for the device performance. The presence of those intra‐grain defects is also confirmed by EBIC maps, which allows comparing the electrical activity of both grain and sub‐grain boundaries (© 2012 WILEY‐VCH Verlag GmbH &amp; Co. KGaA, Weinheim)

Effect of Si&lt;sub&gt;3&lt;/sub&gt;N&lt;sub&gt;4&lt;/sub&gt; Coating on Strain and Fracture of Si Ingots

This study is concerned with the effect of reduced Si3N4 coating during Si growth carried out by the directional solidification cast method. Three crystals were grown and compared in terms of cracking behavior and strain distribution. It was found, that the outer areas of the grown Si ingots exhibit much higher strain than the center pieces and this corresponds with the appearance of cracks in higher density in the outer edges of the crystals. It was also found that cracks mainly propagated in vertical direction and less in horizontal direction due to the temperature gradient during growth.

Centimeter-Sized Bulk GaN Single Crystals Grown by the Na-Flux Method with a Necking Technique

Centimeter-sized bulk GaN single crystals with large dislocation-free areas were fabricated by the Na-flux method with a necking technique. This necking, which is the key technique in Czochralski growth of dislocation-free Si ingots, was realized using a newly developed GaN point seed. Structural properties of grown crystals were investigated using panchromatic cathodoluminescence (CL) measurements and X-ray diffraction. Prism-shape and well-faceted bulk GaN crystals with dimensions as large as 0.85 cm (width) and 1 cm (length) were grown by this technique. The GaN single crystal grown for 400 h have full-width at half-maximum values for c// and c⊥ as narrow as 42.8 and 32.5 arcsec, respectively, indicating an extremely high quality. Panchromatic CL images of (0001) GaN wafers sliced from grown crystals revealed that large areas of the wafers were dislocation free. We concluded that the necking technique in Na flux GaN growth may be a major breakthrough for fabricating large dislocation-free GaN ingots.