Search

Solid-state lithium batteries have become an attractive alternative to traditional lithium-ion batteries due to their enhanced safety and energy density potential. However, their current level is still immature from a cell engineering perspective. This work presents ultimate design principles for achieving high energy density in practical solid-state lithium batteries. Our proposed design logic is based on evaluating the microstructural and architectural parameters that significantly impact energy density. We demonstrate the feasibility of this principle through the fabrication of a 10-layer stacked solid-state pouch cell with a solid polymer electrolyte, resulting in an energy density of 280 Wh kg−1 and 600 Wh L−1. This research provides a promising path forward for the scientific community to pursue practical, high-energy-density solid-state batteries that can meet the growing demand for energy storage solutions.

Crystalline Si core–amorphous SiOx shell nanoparticles were synthesized from SiCl4 using an atmospheric microwave plasma process and their microstructures were investigated. It was found that the amorphous SiOx shell thickness was determined by varying the applied microwave power. The increase of applied power directly increased the length of the flame at the end of the plasma. Through observation of the microstructures with varying lengths of the flame, it was found that a highly crystalline Si phase was formed in the plasma and amorphous SiOx phase was formed in the flame. The longer flame resulted in the thicker SiOx shell around the Si core. The electrochemical properties of the varying SiOx shell thicknesses were also investigated in that the synthesized Si-SiOx nanoparticles were used as anode materials in a lithium-ion battery (LIB). Thin shells such as native oxide did not influence the electrochemical behaviour of the Si core, whilst some shells were too thick to allow Li-ions to react with the Si core. Using Si-SiOx nanoparticles with the optimum SiOx:Si ratio of 18:82, the following electrochemical properties were obtained: a first reversible capacity of 932 mAh/g, an initial columbic efficiency (ICE) of 52.5%, and a capacity retention of 83.7% at the 100th cycle.

Nanoparticles having four different ratios of crystalline silicon (Si) cores and amorphous SiOx shells were synthesized by using a microwave-generated plasma. Their microstructures were analyzed to find quantitative relationships with electrochemical properties. By measuring XRD, SEM, HR-TEM, and Raman spectra of four samples with different core-shell microstructures, the quantitative fractions of crystalline Si core and amorphous SiOx shell in the nanoparticles were calculated. Also, electrochemical properties of the nanoparticles were measured and compared with the calculated fractions. The electrochemical properties such as the initial reversible capacity, the initial coulombic efficiency, and the capacity retention exhibited tendencies remarkably similar to those implied by the calculated fractions. Analysis of each sample's dQ/dV profile also gave us good evidence of understandable relationships between the fractions and electrochemical properties. Details of calculating the fractions from microstructural data are given, with suitable consideration having been given to the fact that several assumptions could lead to errors during the analysis. The approach introduced in this research offers a good means of analysis with which to further the optimal design of nanoparticle microstructure for Li-ion battery anodes.

We synthesized silicon (Si) nanoparticles using an atmospheric microwave plasma process, and investigated the effects of hydrogen (H2) injection on their microstructure during the synthesis. Two nozzles were applied to inject H2 (swirling and rectilinear H2). Our microstructural analysis indicated that the amount and method of H2 injection were critical for completion of the reaction from silicon tetrachloride (SiCl4) to Si, as well as to obtain highly crystalline Si nanoparticles. The swirling H2 was especially critical due to its formation of vortex flow, which allowed relatively long residence time of the H-ions in plasma. The Si nanoparticles synthesized by the atmospheric plasma process had core-shell structures that consisted of crystalline Si cores with amorphous SiOx shells of 5–15 nm thickness. We also investigated the feasibility of the synthesized Si nanoparticles as anode materials in a lithium-ion battery (LIB). For the core-shell structured Si nanoparticles, we obtained the first reversible capacity of 1204 mAhg−1, and a capacity retention of 82.2% at the 50th cycle.

Effects of SiOx or C shells on electrochemical properties of Si nanoparticles were investigated. SiOx shells with thickness of 10~15 nm were formed on homogeneously crystalline Si nanoparticles. Incase of Si-C nanoparticles, there were 30~40 layers of C with a number of defects. Li-ion batteries were fabricated with the above-mentioned nanoparticles, and their electrochemical properties were measured. Pristine Si shows a high IRC (initial reversible capacity) of 2,517 mAh/g and ICE (initial columbic efficiency) of 87%, but low capacity retention of 22%, respectively. SiOx shells decreased IRC (1,534 mAh/g) and ICE (54%), while the retention increased up to 65%, which can be explained by irreversible phases such as LiO2 and Li2SiO3. C shells exhibited no differences in IRC and ICE compared to the pristine Si but an enhanced retention of 54%, which might be from proper defect structures.

P-type single crystalline silicon (Si) wafers with a thickness of 100 μm were obtained in a free abrasive (slurry on) multi-wire sawing process. When wafer thickness decreased to 100 μm, Cu-contamination on sliced wafer surface occurred. We investigated the contamination mechanism and its solution, and finally eliminated this contamination by applying the solution found herein. The thinner the wafer thickness became, the more Cu debris were detached from brass-coated wire, those ultrafine debris (0.08–0.2 μm) became aggregated on SiC abrasive particles. The aggregated Cu debris was strongly attached on wafer surface by mechanical load from wire and thermal energy during the sawing process when wafer was as thin as 100 μm. By adding a coolant to prevent aggregation during the process, Cu-contamination was successively removed. The influences of contamination on physical properties of sliced wafers such as thickness variation, roughness were also investigated. The contaminated wafers exhibited relatively higher reflectivity after texturing process, which resulted in degradation of photoelectric conversion efficiency from 17.6% to 17.0%.

The thermal behavior of silicon nanoparticles (Si NPs) was investigated for the preparation of silicon thin film using a solution process. TEM analysis of Si NPs, synthesized by inductively coupled plasma, revealed that the micro-structure of the Si NPs was amorphous and that the Si NPs had melted and merged at a comparatively low temperature (~750 °C) considering bulk melting temperature of silicon (1414 °C). A silicon ink solution was prepared by dispersing amorphous Si NPs in propylene glycol (PG). It was then coated onto a silicon wafer and a quartz plate to form a thin film. These films were annealed in a vacuum or in an N₂ environment to increase their film density. N2 annealing at 800 °C and 1000 °C induced the crystallization of the amorphous thin film. An elemental analysis by the SIMS depth profile showed that N₂annealing at 1000 °C for 180 min drastically reduced the concentrations of carbon and oxygen inside the silicon thin film. These results indicate that silicon ink prepared using amorphous Si NPs in PG can serve as a proper means of preparing silicon thin film via solution process.

In this study, 1.0 mol% CuO-added (NaxK1−x)NbO3 (CNxK1−xN) ceramics with 0.0 ≤ x ≤ 0.5 were sintered at a relatively low temperature of 960°C. A large Qm value (≥1750) was obtained for the specimens with x ≤ 0.2 owing to the hardening effect of Cu2+ ions, indicating that a large number of Cu2+ ions had replaced Nb5+ ions in these ceramics. The double polarization versus electric field (P–E) hysteresis loop was observed for the specimens with x ≤ 0.2, and was explained by the defect polarization (Pd) formed between Cu2+ ions and O2− vacancies. These specimens also exhibited a sprout-shaped strain versus electric field (S–E) curves with a large strain of 0.14%–0.15% at 8.0 kV/mm because of the internal bias field produced by Pd. For the CN0.2K0.8N ceramic, a double P–E hysteresis loop was maintained at 125°C without remnant polarization. Moreover, this ceramic showed a large strain of 0.175% after 104 cycles when a high electric field of 8.0 kV/mm was continuously applied to the specimens during each cycles. Therefore, CN0.2K0.8N is a useful lead-free piezoelectric ceramic for multilayer actuators.

Thin silicon (Si) wafers with thickness of 100–140 μm were obtained by using a multi-wire slicing process with a SiC slurry. We investigated the process yields of wafer slicing as well as physical/electrical properties of the sliced wafers with various thicknesses. As the wafer thickness decreases, the process yield abruptly decreased due to wafer breakage during the slicing process, and conventional polishing the brick surface was not enough to gain a considerable process yield. However, elimination of defects on the brick surface by mirror polishing resulted in an 83.1% enhancement of yield even for wafers with a thickness of 100 μm. The number of wafers obtained was even higher than that of conventional 180 μm-thick wafers (479 vs. 415 wafers in this research). Investigation of the microstructure of brick surfaces revealed that surface defects on bricks were main parameter to determine the yield of slicing process. Surface defects containing the micro-cracks introduced residual stress, which decreased the slicing process yield especially for thinner wafers. From measurements of physical and electrical properties, it was revealed that the relative total thickness variations (TTVs) and bowings increased and the characteristic fracture strength of sliced wafer and conversion efficiencies decreased as the wafer thickness decreased. For the wafer with thickness of 100 μm, the relative TTV and bowing were 14.1% and 22.5 mm, respectively. The conversion efficiency of a solar cell using this wafer was 17.6%, while that of a conventional Si solar cell using a 180 μm-thick wafer was 18.4%.

Silicon-carbon (Si-C) composites with various weight ratios were prepared through heat treatment of water-soluble carboxymethyl cellulose (CMC) with Si nanoparticles synthesized by using an inductively-coupled plasma. Microstructures of the Si-C composites were thoroughly investigated by using X-ray diffraction (XRD), scanning electron microscopy (SEM) and Raman spectroscopy. The results indicate that we obtained a micro-sized Si-C composite with homogeneously-distributed crystalline Si nanoparticles in an amorphous C-matrix. Pores, which were due to the volatilization of CO2 from CMC during heat treatment, was detected when the concentration of carbon was increased. The electrochemical properties of those Si-C composites for use as anode materials in lithium-ion batteries (LIBs) were also investigated. The C-matrix enhanced the capacity retention, as well as the rate capability of Si nanoparticles, due to the dense and homogeneous microstructures of the composite. The Si-C composites (7:3 weight ratio) retained a reversible capacity of > 1,000 mAh/g with a capacity retention of 88.9% even after 100 cycles. The reversible capacity ratio at a 1.5 C-rate was about 80% as compared with that at a 0.1 C-rate.

The removal of boron is one of the main challenges in the purification of metallurgical grade silicon destined for low-cost photovoltaic applications. However, boron is very difficult to remove in its elemental form due to its large segregation coefficient in silicon and its low vapor pressure. The removal of boron by slag treatment is today regarded as a highly promising method, but its refining efficiency is relatively low. Also, the reduction of boron by plasma treatment exhibits a high refining efficiency, but the processing cost is high due to the large amount of electricity consumed by the process. In this regard, the use of an oxidizing reactive gas in the refinement process offers some advantages both in terms of low energy consumption and promising refinement rates. Boron can be extracted in various gaseous forms as B(x)O(y) and/or B(x)H(z)O(y) phases, but the vapor pressure of B(x)H(z)O(y) is much greater than that of the other specie at a temperature of 1700 K. The present study reports a modified oxidative refining method designed to enhance the vaporization of boron as B(x)H(z)O(y) by blowing gaseous water onto the silicon melt in a segmented crucible to enhance the electromagnetic force, whereby the processing cost can be dramatically reduced due to the use of a reusable quartz crucible in a graphite crucible. An initial boron content of 13 ppm in the metallurgical grade silicon was significantly decreased to 0.3 ppm by the employment of 1.7SLM Ar + 100 ml/h H2O. Also, a mechanism capable of reducing boron based on thermodynamic considerations is proposed.

A room temperature refining process using a new acidic mixture was investigated for the purification of metallurgical grade-silicon (MG-Si) with an average particle size of 617 μm. MG-Si typically contains precipitated metal impurities, such as iron (Fe), aluminum (Al) and titanium (Ti), in the triple points of grain boundaries. These metal impurities tend to concentrate within the laminar structure at grain boundaries during solidification due to their different segregation coefficients and are leached with acid at different rates because of their different reactivities with acid. The addition of acetic acid (CH3COOH) to a conventional acid mixture composed of nitric acid (HNO3) and hydrofluoric acid (HF) resulted in improved extraction of the main metal impurities. This new acid mixture efficiently extracted Al, which is known to be particularly difficult to remove with conventional acid leaching agents, such as hydrochloric acid (HCl), nitric acid (HNO3), and hydrofluoric acid (HF). The recorded decreases of the Fe and Al concentrations in the material followed two distinct leaching events. The first decrease was due to the extraction of impurities located on the particle surface; the second occurred when the acid leaching time exceeded 15 h. This second leaching event may be due to the removal of metal impurities located deep in the material's pores and deep within the bulk structure of the MG-Si particles. After 25 h of leaching MG-Si with an acid mixture composed of HNO3, HF, and CH3COOH, the purity of the material increased from 99.74 to 99.99% with extraction efficiencies of 99.92 for Fe and 99.98% for Al and Ti.

Si nanoparticles were synthesized by inductively coupled plasma and a specially designed double tube reactor. By injection of large amount of PH3 during the synthesis, the effects of phosphorous incorporation on their microstructures and chemical binding environments were investigated. Injection of PH3 gas during the synthesis resulted in a change from crystalline to amorphous phase, a reduction of particle size as well as a process yield. All of the above results were attributed to a lower plasma density when higher amount of PH3 was injected. From energy-dispersive X-ray spectroscopy, secondary ion mass spectrometry, and X-ray photoelectron spectroscopy analysis, it was revealed that P was doped in Si nanoparticles. However, secondary phases such as P4 and P2O5 were formed as amorphous ones in nano-scale when a relatively large amount of PH3 was injected. In addition, those nanoparticles were applied as an active material in the lithium-ion battery's anode. Unexpectedly, amorphous Si nanoparticles with secondary phases showed improved electrochemical properties. P-doping in Si nanoparticles could not directly advance cycling performance by improvement of electrical conductivity of Si nanoparticles. It was rather assumed that a secondary phase influenced and enhanced electrochemical properties by additional capacity due to a formation of Li3P and forming an effective buffer against large volumetric change of Si nanoparticles during the charge/discharge. The initial reversible capacity of amorphous Si nanoparticles synthesized with 100 sccm of PH3 flow rate was 2113 mAh g− 1, and that at the 100th cycle was still about 1000 mAh g− 1, which was twice as high as that of Si nanoparticles synthesized without PH3 injection.

The oxidation behavior of photovoltaic ribbon was measured for 48 hours over the temperature range of 300°C to 800°C using thermogravimetric apparatus (TGA). It was found that weigh-gained of both samples was found to be increased with increasing temperature and it was dramatically increased at 800°C for 30 minutes of initial oxidation time. Pure copper with 99.55% purity was successfully recovered from the photovoltaic ribbon of spent solar module using cyclic oxidation method.

This paper presents a metallurgical route that consumes much less energy than the gasification route such as Siemens process-based photovoltaic (PV) value chain including metallurgical-grade silicon (MG-Si), solar-grade silicon (SoG-Si) purified by the metallurgical route, Czochralski (Cz) ingot growth, 6″ wafer fabrication, solar cell fabrication, and PV performance evaluation. Mono-like crystalline silicon ingot for achieving high purity was continuously cast with MG-Si for the fabrication of feedstock for Cz ingot growth through the advanced electron beam melting (EBM) system. The system was developed with the combination of vacuum refining part and directional solidification (DS) part for the purification of large amounts of MG-Si, where the two parts are connected by an overflow of Si melts to undergo continuous refining and casting processes in a chamber. To grow the mono-like crystalline Si ingot in our EBM system, a single crystalline seed was placed on the graphite dummy cylinder with DS zone. The mono-like crystalline Si ingot with high purity of >99.999% (above 5N) was blended 50–50 with undoped SoG-Si chunk for the compensation of boron dopant concentration in a crucible of Cz grower. The 156 × 156 mm 2 wafers were cut from Cz ingot grown with purity of >99.99999% (above 7N) by slurry-based multi-wire sawing and were fabricated by the conventional solar cell process. The measured photovoltaic performance recorded cell efficiency of 17.1%; hence its higher potential for application to the PV industry.

Fractional melting is a good method of refining metallurgical-grade silicon. It can be used to replace the traditional Siemens process for that purpose. It is important to observe the changes in the microstructure of silicon during heating to increase the efficiency of fractional melting. The microstructures and the chemical compositions of impurity phases in heated samples were observed by using scanning electron microscopy and wavelength dispersive spectroscopy. The affinities among the various impurities were investigated using the Miedema model. Ti and Al had strong affinities with V and Ca, respectively. It is possible to use elements that have strong affinities with B and P to effectively remove B and P from the silicon matrix, which are otherwise difficult to remove because of their high segregation coefficients.

Phosphorous (P) incorporated silicon nanoparticles (Si NPs) were synthesized by using inductive coupled plasma (ICP) and a specially designed double tube reactor.Their microstructures were investigated by injecting various amounts of PH3 gas during the synthesis. Injection of PH3 gas during the synthesis resulted in a change from crystalline to amorphous phase, a reduction of particle size as well as process yield. These results were attributed to a lower plasma density when higher amount of PH3 was injected.From EDS, SIMS and XPS analysis, it was revealed that P was successively incorporated in Si NPs.However, secondary phases such as P4 (red P) and P2O5were formed as amorphous ones in nanoscale when a relatively large amount of PH3 was injected.

A silicon substrate with the dimensions of 100 × 140 × 0.3 mm was grown directly from liquid silicon with gas pressure. The silicon melt in the sealed melting part was injected into the growth part at applied pressure of 780–850 Torr. The solidified silicon substrate was then transferred by the pull of the cooled dummy bar. A desirable structure with a liquid-solid interface perpendicular to the pulling direction was formed when the mold temperature in the solidification zone of the growth part was much higher than that of the dummy bar, as this technique should be able to overcome thermal loss through the molds and the limited heat flux derived from the very narrow contact area between the silicon melt and the dummy bar. In addition, because the metallic impurities and expansion of volume during solidification are preferably moved to a liquid phase, a high-quality silicon substrate, without defects such as cracks and impurities in the substrate, could be manufactured in the interface structure. The present study reports the experimental findings on a new and direct growth system for obtaining silicon substrates characterized by high quality and productivity, as a candidate for alternate routes for the fabrication of silicon substrates.

The oxidation rate of pure copper was measured for 48 hours over the temperature range of 500°C to 700°C using thermogravimetric apparatus by means of continuous and discontinuous oxidation methods. It was found that weigh-gained of both samples was found to be increased with increasing temperature and rate laws were determined to be parabolic and linear rate law each other. The comparison of oxidation rate of copper was shown to be higher in discontinuous that in continuous process due to direct oxidation through the oxide scale and copper substrate.

The most desirable approach to the hydrometallurgical process consists in using larger silicon particles by exposing the metallic impurities contained in the silicon to its surface via a thermallyactivated elution prior to chemical treatment. The present study reports experimental findings concerning the effect of a thermal pre-treatment using a mixture of 5-wt% nitric acid and 2.5-wt% hydrofluoric acid for the purification of metallurgical-grade silicon particles of different sizes on the hydrometallurgical process. The extraction rates of metallic impurities from inside the silicon were in inverse proportion to the size of the silicon particle. However, the effect of the thermal pre-treatment on the extraction rate became greater with increasing particle size.

With direct growth technologies for obtaining ribbon-type silicon (Si) wafers from molten Si, the horizontal growth process is much faster than the vertical one in terms of the growth rate of the wafer. However, such a fast growth rate is likely to create grains of a relatively small size in the Si wafer, resulting in limited efficiency caused by grain boundary recombination. In this regard, controlling the grain growth behavior of the Si wafer with processing parameters such as the temperature of the substrate and its moving speed is very important in the horizontal growth process. The present investigation produced the experimental findings concerning the processing parameters for controlling the grain size. The size of the grains on the surface of the Si wafer increased to around 180 μm at a lower substrate temperature and a slower the moving speed of the substrate due to the increasing thickness of the Si wafer. However, the growth rate of the grains was observed to have little relation to the variations in the wafer thickness resulting from the processing parameters unless the nucleation behavior had been controlled.

Silicon (Si) nanoparticles were synthesized using a double tube reactor and inductive coupled plasma (ICP), and their microstructures with various process conditions were investigated in terms of plasma density and residence time of reactive gas in the plasma zone. Homogeneous amorphous or single crystalline Si nanoparticles were synthesized depending on applied power to ICP coil. The SiH4 flow rate mainly determined the particle size. Average particle sizes of the synthesized Si nanoparticles were 14–34 nm, and the maximum production rate was as high as 1000 mg h−1. The purity of synthesized Si nanoparticles was higher than 99.99%. When Si nanoparticles were applied as an active material in the anode of a Li-ion battery (LIB), the high first reversible capacities of 2540 (initial coulombic efficiency (ICE) = 61.4%) and 1358 mAh g−1 (ICE = 62.5%) for crystalline and amorphous Si nanoparticles were attained, respectively. However, capacity fadings were detected in both nanoparticles, and capacity retentions at 100th cycle were only 17.1% and 34.8% for crystalline and amorphous Si nanoparticles.

A new polycrystalline silicon (Si) wafering process directly from Si melt is introduced. As Si is known to have high latent heat, maintaining the steady state solidification condition in target area could be a main issue to achieve proposing wafering process. In addition, as the proposed process is based on horizontal growth, another critical issue is to keep grains growth parallel to the wafer growth direction, which results in large grain sizes. At first, simple numerical modeling was used to evaluate the possibility of realizing this new conceptual process, and to determine the main process parameters. From the simulation results, growth velocity and heat transfer rate at the solidification zone was identified as main process parameters for the steadily grown Si wafer within target area. Based on the simulation results, a growth system was set up experimentally, and the feasibility of the process was examined. A Si wafer with dimensions of 156×156×0.3 mm3 was successively obtained and grains growth parallel to wafer growth direction as calculated in the simulation were observed.

Lithium ion transport was accelerated within graphite by controlling its d-spacing as well as its functional groups. By oxidizing bare graphite under a mild condition, expanded graphites (EG* where * = functional groups) were obtained with increasing d-spacing from 0.3359 nm to 0.3395 nm as well as with functional groups formed on the plane or at the edges of graphites. The subsequent thermal reduction of EG* led to an insignificant change of d-spacing (0.3390 nm), simultaneously eliminating a portion of the functional groups (EG). The enlargement of d-spacing reduced kinetic hindrance of lithium ion movement within the expanded graphites (EG* and EG) by reserving more space for the ionic transport route. In addition, the activation energy of lithium ion intercalation in EG* was reduced by surface charge polarization of graphites induced by hydrogen bonds between oxygen atoms of carbonates in electrolytes and hydrogen atoms of surface functional groups of the expanded graphites, even if the degree of graphitization decreased. Re-graphitization induced by the subsequent thermal reduction increased delithiation capacities (QdLi) of EG as an anode for lithium ion batteries especially at high currents: QdLi at 50 C = 243 mA h g−1 for EG versus 66 mA h g−1 for bare graphite.

A novel strategy for introducing ion-permselective properties in a conventional polyethylene (PE) separator to inhibit the shuttle effect of polysulfides in high-performance lithium–sulfur batteries is reported. This was accomplished by taking advantage of the pH-responsive multilayers of weak polyelectrolytes such as poly(allylamine hydrochloride) (PAH) and poly(acrylic acid) (PAA) assembled on the PE separator using layer-by-layer (LbL) assembly. It was found that the cationic permselectivity (permeability of cation/anion) of an ultrathin multilayer coated separator is highly tunable with respect to the number of bilayers and external pH, benefiting from fine tuning of the internal charge density of the multilayered films. The movement of polysulfide anions was significantly inhibited by five bilayers of PAH/PAA (ca. 98% with multilayers assembled at pH 3/3), while the movement of Li cations was preserved. As a result, the ion-permselective separator demonstrated a high initial reversible capacity of ca. 1418 mA h g−1 with multilayers assembled at pH 3/3 because of the good permselectivity and the enhanced wetting properties of the LbL treated separator for electrolytes, leading to a significantly improved Coulombic efficiency as compared to a conventional PE separator, i.e., almost 100% over 50 cycles. We anticipate that the permselectivity controllable coating method will be applied for various other membrane technologies.

The recovery of copper from spent photovoltaic ribbon was conducted using thermal treatment method at the range of temperature of to under inert atmosphere. The coating layer consisted of lead of 68.99 wt.% and tin of 31.21 wt.% was melted down at elevated temperatures and was collected on the bottom of crucible. The chemical composition of copper ribbon after thermal treatment was analyzed by ICP-MS (Inductively coupled plasma mass spectrometry) and the purity of copper was found to be obtained up to about 96 wt.% regardless of temperatures. The cross-sectional area of the specimen was also examined by SEM (scanning electron microscopy) and EDX (energy dispersive X-ray microscopy).

A ribbon-type polycrystalline silicon wafer was directly fabricated from liquid silicon via a novel technique for both a fast growth rate and large grain size by exploiting gas pressure. Effects of processing parameters such as moving speed of a dummy bar and the length of the solidification zone on continuous casting of the silicon wafer were investigated. Silicon melt extruded from the growth region in the case of a solidification zone with a length of 1cm due to incomplete solidification. In case of a solidification zone wieh a length of 2 cm, on the other hand, continuous casting of the wafer was impossible due to the volume expansion of silicon derived from the liquid-solid transformation in solidification zone. Consequently, the optimal length of the solidification zone was 1.5 cm for maintaining the position of the solid-liquid interface in the solidification zone. The silicon wafer could be continuously casted when the moving speed of the dummy bar was 6 cm/min, but liquid silicon extruded from the growth region without solidification when the moving speed of the dummy bar was 9 cm/min. This was due to a shift of the position of the solid-liquid interface from the solidification zone to the moving area. The present study reports experimental findings on a new direct growth system for obtaining silicon wafers with both high quality and productivity, as a candidate for an alternate route for the fabrication of ribbon-type silicon wafers.

Silicon sheets were fabricated via spin-casting, which is a ribbon manufacturing method for photovoltaics. Although the advantages of spin-casting include rapid processing and mass production with no silicon loss, crystalline defects such as random grain boundaries (GBs), sub-GBs, or dislocations are generated from rapid solidification. These defects are an important factor in the decreased conversion efficiency of solar cells. Spin-cast sheets were annealed at 950 °C, 1200 °C, and 1350 °C for 2 h to eliminate defects and to increase the grain size. The effect of thermal processing on the sheets was analyzed by comparing them with as-cast sheets and investigated by electron backscattered diffraction, photoluminescence, and etch-pit density. It was confirmed that sub-GBs were almost completely assimilated with neighboring grains with the 1350 °C heat treatment. Suppression of dislocations was observed through the decrease in the D-band peak with low-temperature photoluminescence (PL, 10 K), and few dislocation clusters were observed from the Secco-etched sheet annealed at 1350 °C.

An electron beam melting (EBM) system was fabricated to investigate its effectiveness in the refining of metallurgical grade silicon. The metallurgical refining of silicon is a promising process for reducing the production cost of silicon feedstock for solar cells. Especially, EBM is an excellent technology for evaporating not only metal impurities such as Mg and Ca, but also non-metal impurities such as P, C and O. In this study, a Cu cold crucible was placed in a high vacuum chamber equipped with a 100-kW electron-beam (e-beam) gun. Silicon was charged without any pretreatment such as crushing or acid-leaching. Among the various process parameters, the e-beam pattern was a key to determine the purification efficiency. Two patterns of e-beam were applied, circular and spiral, and their effects on the microstructure and impurity removal were investigated. A different interface between the solid and liquid silicon phase was formed during the solidification depending on the e-beam patterns. Those results were identical to those of the numerical analysis. When a spiral pattern was applied, the horizontal interface was formed and grains grew vertically and uniaxially, which enhanced the impurity removal efficiency. The charged silicon purity of 99.806% was enhanced up to 99.996% after only 44 min of EBM and refining.

A new metal-strip-casting process called continuous strip-casting (CSC) has been developed for making thin metal strips. A numerical simulation model to help understand solid-liquid interface behavior during CSC has been developed and used to identify the solidification morphologies of the strips and to determine the optimum processing conditions. In this study, we used a modified level contour reconstruction method (LCRM) and the sharp interface method to modify interface tracking, and performed a simulation analysis of the CSC process. The effects of process parameters such as heat-transfer coefficient and extrusion velocity on the behavior of the solid-liquid interface were estimated and used to improve the apparatus. A Sn (Tin) plate of dimensions 200 x 50 x 1 mm3 was successfully produced by CSC for a heat-transfer coefficient of 104 W/m2 K and an extrusion velocity of 0.2 m/s.

SiO(x) nanoparticles were synthesized using a specially designed induction melting system equipped with a segmented graphite crucible. The graphite crucible with the segmented wall was the key to enhancing the evaporation rate due to the increase of the evaporation area and convection of the silicon melt. Injection of the gas mixture of oxygen (O2) and argon (Ar) on silicon (Si) melt caused the formation of SiO(x) nanoparticles. The evaporated SiO(x) nanoparticles were then cooled and condensed in a process chamber. The effects of the O2/Ar ratio in the injection gas on the microstructures of the SiO(x) nanoparticles were then investigated. Synthesized SiO(x) nanoparticles were proven to be of a homogeneous amorphous phase with average diameters of 30-35 nm. The microstructures were independent from the O2/Ar ratio of the injected gas. However, x increased from 1.36 to 1.84 as the O2/Ar ratio increased. The purity of the synthesized nanoparticles was about 99.9%. SiO(x) nanoparticles could be applied as the active anode material in a lithium (Li) ion secondary battery.

Silicon (Si) wafer was grown by using direct growth from Si melt and contaminations of wafer during the process were inves-tigated. In our process, BN was coated inside of all graphite parts including crucible in system to prevent carbon contamination. Inaddition, coated BN layer enhance the wettability, which ensures the favorable shape of grown wafer by proper flow of Si melt incasting mold. As a result, polycrystalline silicon wafer with dimension of 156× 156 mm and thickness of 300 ± 20 um was suc-cessively obtained. There were, however, severe contaminations such as BN and SiC on surface of the as-grown wafer. While BNpowders were easily removed by brushing surface, SiC could not be eliminated. As a result of BN analysis, C source for SiC wasfrom binder contained in BN slurry. Therefore, to eliminate those C sources, additional flushing process was carried out before Siwas melted. By adding 3-times flushing processes, SiC was not detected on the surface of as-grown Si wafer. Polycrystalline Siwafer directly grown from Si melt in this study can be applied for the cost-effective Si solar cells.Key words: Silicon, Wafer, Direct growth, Contamination, Silicon carbide

SiO1.39 nanoparticles synthesized by using an evaporation and condensation process were composited with phenol resin and conducting carbon (C), and their microstructures and electrochemical properties were investigated. A dense, micro-sized SiO1.39-C composite was synthesized to prevent capacity fading of the SiO x nanoparticles. To densify and coarsen this composite, we carried out pelletizing, annealing and crushing of the sample. The electrical conductivity was also improved by the composition without any structural change. Crushed particles were used as an anode material for a Li-ion battery, and its cyclic performance was evaluated. By composition, the coulombic efficiency was observed to increase from 93.6 to 97.5% at the 5 th cycle, and it continued increasing with each additional cycle. For the composite annealed at 1,000 °C in a 4-vol.% H2/Ar atmosphere, the specific capacity and the coulombic efficiency at the 20 th cycle were measured as 426 mAhg−1 and 99.7%, respectively.

Remarkable improvements in the electrochemical performance of Si materials for Li-ion batteries have been recently achieved, but the inherent volume change of Si still induces electrode expansion and external cell deformation. Here, the void structure in Si-encapsulating hollow carbons is optimized in order to minimize the volume expansion of Si-based anodes and improve electrochemical performance. When compared to chemical etching, the hollow structure is achieved via electroless etching is more advanced due to the improved electrical contact between carbon and Si. Despite the very thick electrodes (30 ∼ 40 μm), this results in better cycle and rate performances including little capacity fading over 50 cycles and 1100 mA h g−1 at 2C rate. Also, an in situ dilatometer technique is used to perform a comprehensive study of electrode thickness change, and Si-encapsulating hollow carbon mitigates the volume change of electrodes by adoption of void space, resulting in a small volume increase of 18% after full lithiation corresponding with a reversible capacity of about 2000 mA h g−1.

Concerning horizontally direct growth technology, graphite material with excellent thermal and chemical resistance is frequently used as both the crucible and the substrate in equipment operated at high temperatures. However, this can cause a major problem, namely, contamination due to the carbon and metallic impurities derived from graphite components. This study presents the experimental findings on a method of preventing the incorporation of unintended impurities into the silicon wafer by modifying the surface of the graphite. Coating the crucible and substrate with silicon carbide (SiC) could significantly reduce carbon (1.5 × 1017 atom/cm3) and metallic (undetected under analysis resolution) contamination by suppressing physical contact between the silicon melt and the graphite. It is expected that the purity of the silicon wafer will be improved to meet the demand for next-generation solar cell production with the optimum growth conditions required for obtaining the desired property of the silicon wafer.

The high cost of crystalline silicon solar cells has been considered as one of the major obstacles to their terrestrial applications. Spin on doping (SOD) is presented as a useful process for the manufacturing of low cost solar cells. Phosphorus (P509) was used as an n-type emitters of solar cells. N-type emitters were formed on p-type EFG ribbon Si wafers by using a SOD at different spin speed (1,000~4,000 rpm), diffusion temperatures (), and diffusion time (5~30 min) in atmosphere. With optimum condition, we were able to achieve cell efficiency of 14.1%.

Ribbon silicon solar cells have been investigated because they can be produced with a lower material cost. However, it is very difficult to get good texturing with a conventional acid solution. To achieve high efficiency should be minimized for the reflectance properties. In this paper, acid vapor texturing and anti-reflection coating of SiNx was applied for EFG Ribbon Si Wafer. P-type ribbon silicon wafer had a thickness of 200 μm and a resistivity of 3 Ω-㎝. Ribbon silicon wafers were exposed in an acid vapor. Acid vapor texturing was made by reaction between the silicon and the mixed solution of HF : HNO3.. After acid vapor texturing process, nanostructure of less than size of 1 μm was formed and surface reflectance of 6.44% was achieved. Reflectance was decreased to 2.37% with anti-reflection coating of SiNx.

Induction melting process using segmented graphite crucible was investigated to melt silicon feedstock for solar cells. Induction melting is the key technology used in silicon melting process, such as ingot-growing, metallurgical refining and ribbon production. However, contamination of silicon from the crucible in induction melting is the main source of degrading the silicon. In this paper, new structure of crucible was suggested to minimize the contamination from the crucible wall. It was the segmentation of graphite crucible wall that introduced the non-contact between melt and inner wall of crucible. Numerical and experimental studies of induction melting process of silicon using the suggested crucible were conducted. For numerical analysis, 3D models of crucible, induction coil and silicon were constructed and electromagnetic force and temperature distribution in silicon and crucible were calculated. To evaluate a process with the suggested crucible, induction melting system was built up based on the simulated results and experiments of silicon melting were carried out.