Dissertations / Theses on the topic 'Impurity distribution'

Fusion power is the utilization of the energy released in nuclear fusion reactions. It has the potential to become an energy source which is more sustainable, safer and cleaner than the primary energy sources today. The fundamental problem for fusion power is energy confinement. Impurity ions are a major source of energy confinement loss in magnetic confinement fusion devices. Hence, impurity transport in fusion plasma is an important field to study. Ion cyclotron resonance heating (ICRH) has been shown both experimentally and theoretically to influence impurity transport in tokamaks. A poloidal asymmetry in the minority ions produces an electric potential, which causes impurity accumulation on the inboard side of the tokamak. Poloidal asymmetry in the impurity density induces a net radial impurity flux over a flux surface. This project has compared the ICRH-induced impurity transport for four approximate minority ion distribution function models with numerical results from the SELFO code. This has been done computationally for JET-like, concentric tokamak geometry with deuterium plasma, hydrogen minority ions, and tungsten impurities. Two models, the so-called bi-Maxwellian and LFS bi-Maxwellian model, are used in existing literature. Two further models are introduced, called the tri-Maxwellian and LFS tri-Maxwellian model. Unlike the bi-Maxwellian models, these models take into account the existence of thermal and fast ions in the minority population. The results show that there are noticeable differences between the different models, in particular when the resonant surface is on the inboard side. The tri-Maxwellian models offer a clear improvement over the bi-Maxwellian models compared with SELFO. However, there are some features in the SELFO results that none of the approximate models predict, this is because the models neglects wide orbits. A possible barrier in the radial transport has also been identified at flux surfaces where the impurity density asymmetry closely resembles the magnetic field strength asymmetry. The LFS bi-Maxwellian model predicts the radial position of the barrier most accurately and reliably compared with SELFO.
Fusionsenergi är utnyttjandet av energi som frigörs i kärnfusionsreaktioner, och har potential för att bli en energikälla som är mer hållbar, säkrare och renare än de primära energikällorna idag. Det grundläggande problemet för fusionskraft är energiinneslutning. Förorenande joner är en viktig källa för förlust av energiinneslutning i fusionsanläggningar med magnetisk inneslutning. Därför är föroreningstransport i fusionsplasma ett viktigt ämnesområde. Joncyklotronresonansupphettning (ICRH) har visats både experimentellt och teoretiskt att påverka föroreningstransport i tokamaker. En poloidal asymmetri i minoritetsjonerna ger en elektrisk potential, som orsakar förorening samlas på den inre sidan av fusionsanläggningen. Poloidal asymmetri i föroreningsdensiteten på ett fluxyta inducerar en netto radialflux över fluxytan. Detta projekt har jämfört ICRH-inducerad föroreningstransport för fyra approximativa distributionsfunktionsmodeller för minoritetsjon med numeriska resultat från SELFO koden. Detta har gjorts med beräkningar för JET-liknande, kon-centrisk tokamak-geometri med deuterium-plasma, väte som minoritetsjoner, och volfram som föroreningsjoner. Två modeller, s.k. bi-Maxwellian- och LFS bi-Maxwellianmodellen, används i existerande litteratur. Ytterligare två modeller introduceras, kallad tri-Maxwellian- och LFS tri-Maxwellianmodellen. Dessa modeller tar hänsyn till förekomsten av termiska och snabba joner i minoritetsbefolkningen. Resultaten visar att det finns tydliga skillnader mellan de olika modellerna, särskilt när resonansytan är på den inre sidan. Tri-Maxwellianmodellerna visar en klar förbättring över bi-Maxwellianmodellerna jämfört med SELFO. Det finns dock vissa särdrag i resultaten från SELFO som ingen av de approximative modellerna förutsäger, eftersom modellerna försummar breda banor. En möjlig barriär i den radiella transporten har också blivit identifierat på fluxytor där asymmetrin i föroreningsdensiteten liknar asymmetrin i den magnetiska fältstyrkan. LFS bi-Maxwellianmodellen förutsäger den radiella positionen av barriären mest noggrant och tillförlitligt jämfört med SELFO.

As the integrated circuit technology evolves to a greater complexity, the purity of process gases becomes a critical issue requiring extreme diligence in gas handling. There are three major challenges in gas contamination control: Gas purification, gas transport, and metrology of contaminants. In the area of gas purification, a novel integrated purification/filtration technique is developed--the reactive filter. The fundamental concept of a reactive filter involves impregnating the surface of an efficient particle filter with chemically reactive sites. The technique is capable of purifying inert as well as reactive gases to sub parts-per-billion levels of atmospheric contaminants at ambient temperature. Development and evaluation results for the reactive filter are presented. In the area of gas transport, this work has experimentally and theoretically studied the fundamental mechanisms of various sources of impurity intrusion and the fate of impurities in an ultra-pure gas transport system. The contamination sources investigated include impurity adsorption and desorption from transport lines and gas filters, impurity back-diffusion, and impurity permeation through polymeric materials. Fundamental kinetic parameters for impurity adsorption/desorption in stainless steel tubes and ceramic and metallic filters are reported. The desorption kinetics expression developed assumes two energy levels for the desorbing species. A systematic filter outgassing data analysis and normalization technique is presented along with a computer code to simulate a simple gas delivery system. The back-diffusion study developed a mathematical model to predict the extent of back-diffusion for various contaminants in high purity gas lines. Back-diffusion is very sensitive to system pressure. Impurity permeation study was conducted on Kel-F and PFA polymeric tubes. The effect of both temperature and pressure on the permeation coefficients was investigated. In the area of metrology, state-of-the-art analytical tools and techniques like Atmospheric Pressure Ionization Mass Spectroscopy (APIMS) were employed to detect and measure trace impurity levels. APIMS has the advantage of short response times and low parts-per-trillion level detection limit for most impurities of interest.

The production of solar grade silicon is based on the use of expensive high purity carbon and quartz feedstock as well as various silicon refining techniques. Impurities in the feedstock materials enter the silicon during the carbothermic reduction of quartz. The knowledge of the impurity distribution/removal in the feedstock and in the carbothermic reduction process is necessary for targeting less pure and cheaper raw materials. The aim of the present study is to investigate the impurity distribution and behavior in quartz feedstock throughout the metallurgical production of silicon; in particular four areas were investigated: 1) removal of mineral inclusions from quartz prior to carbothermic reduction by means of selective fragmentation; 2) effect of pellets, lumps on the silicon production reactions; 3) high temperature properties of quartz under reducing conditions; 4) distribution of contaminants from quartz to the gas phase during quartz reduction. Pellets and lumpy charge, hydrothermal quartz and quartzite were studied. Because of the complex objective of this study, several experimental techniques as well as characterization methods were used: 1) a novel refining route for hydrothermal quartz based on electrical fragmentation, was tested in a laboratory scale to investigate the advantages of using electric fragmentation in terms of cracks distribution, mineral liberation and fragment morphology; 2) mixtures of quartz and silicon carbide in form of pellets or lumps were heated to 2000 °C for 1 h in graphite crucibles in an inductive furnace to compare the properties of pellets and lumps during the silicon production reaction; 3) quartz pieces were heated in CO(g) atmosphere in a sessile drop experiment to study the ability of different types of quartz to produce SiO(g); 4) mixtures of quartz and silicon and quartz and silicon carbide, in form of lumps or pellets, were heated in graphite crucibles to 1650 °C and 1850 °C to investigate which gaseous compounds form other than SiO(g) during quartz reduction, and to which extent. EPMA was used to investigate the morphology of the reacted charge, XRD to identify different phases in a mixture, XRF to quantify elements in a mixture of phase. ICP-MS was used to investigate the impurity content in the charge and reaction products. It was observed that electrical fragmentation efficiently liberates mineral inclusions in the quartz matrix, and that the use of an inexpensive separation technique such as sieving is well suited to separate these minerals. Pellets and lumps behave differently when quartz and silicon carbide are heated to 2000°C in a reducing atmosphere. More silicon is produced with lumps, i.e. quartz reacts faster to SiO(g,) and more SiO(g) is lost when pellets are used. The rate of formation of SiO(g) is enhanced when the quartz contains cracks and fluid inclusions. It was found that during quartz reduction gaseous compounds leave the quartz but to various extents: the average measured volatility of B, P, K, Fe, Al, Mn, Zn, Pb is respectively 11 %, 25 %, 26 %, 45 %, 1 %, 38 %, 10 % and 36%. Liquid inclusions, common in lumps of hydrothermal quartz, enhance the distribution of the contaminants to the gas phase. The gaseous compounds which form may be entrapped by the condensation process, circulate in the furnace and eventually end up in the silicon.

Thesis (M.S. in Applied Physics)--Naval Postgraduate School, Dec. 2004.
Thesis Advisor(s): Nancy M. Haegel, Gamani Karunasiri. Includes bibliographical references (p. 43-45). Also available online.

Electron scattering by a single or multiple impurities affects the quantizaton of conductance of a semiconductor nanochannel. The theoretical model of electron transport in a hardwall nanostructure with an impurity requires an analysis of the electronic transverse energy levels, eigenfunctions and hopping integrals resulting from cross channel or transverse confinement. Theoretical equations for the electronic transverse energy levels, wavefunctions and hopping integrals in the case of a repulsive, finite strength rectangular barrier arbitrarily positioned in the nanochannel are presented. The effects of size, strength and location of the impurity are discussed.In order to find the electronic transverse energy levels, wavefunctions and hopping integrals, two FORTRAN computer programs were developed. The first, called Program Data Input, writes the computational parameters to a data file. The second, Program Single Impurity, uses this data file in performing the calculations of the electronic transverse energy levels, eigenfunctions and hopping integrals.
Department of Physics and Astronomy

L'incorporation de b et c est d'autant plus faible que la teneur en eau de l'encapsulant (b::(2)o::(3) fondu) est plus grande. B et o segregent toujours en queue de lingot, alors que c segrege en tete dans les lingots riches en gallium et en queue dans les lingots riches en arsenic. Par ailleurs, le coefficient de segregation de b, tout en restant toujours inferieur a 1, prend des valeurs differentes suivant la stoechiometrie de la charge initiale. Ces resultats ouvrent la voie, en relation avec des mesures complementaires (electriques, optiques), d'etudes permettant de preciser l'influence de b, c et o sur les proprietes electroniques de gaas

La miniaturisation des transistors mos n'est plus envisageable sans une etape de metallisation auto-alignee des contacts source/drain et grille (procede salicide: self aligned silicide). Dans les technologies mos actuelles, le siliciure le plus couramment utilise est le disiliciure de titane: tisi#2. La siliciuration ayant une influence a la fois sur les dopants et sur la topographie, elle a un impact evident sur les dispositifs. Afin de garantir des simulations electriques precises, la modelisation des procedes technologiques submicroniques doit donc etre capable de prendre en compte cette etape. Dans ce travail, nous avons developpe un modele general de croissance des siliciures et nous l'avons implemente dans un simulateur de procede bidimensionnel. Nous avons demontre qu'il etait possible de simuler la croissance d'un siliciure aussi bien par diffusion du silicium que par diffusion du metal, ainsi que la siliciuration par depot selectif en phase vapeur. Les parametres du modele ont ete ajustes afin d'obtenir des cinetiques de croissance du tisi#2 en bon accord avec les resultats experimentaux. Nous avons montre que notre modele permettait de rendre compte d'effets topographiques bidimensionnels (absence de croissance sous l'espaceur, courbure de la couche de tisi#2 sur les grilles de polysilicium,), a condition de faire dependre certains parametres des contraintes mecaniques generees pendant la croissance de la couche. L'influence de la siliciuration sur la redistribution des dopants a ete analysee ; la simulation a confirme que la segregation du bore a l'interface tisi#2/si etait responsable de la degradation des resistances d'acces des contacts siliciures sur silicium dope au bore. L'injection de lacunes pendant la siliciuration a egalement ete calibree grace aux resultats experimentaux et, par des simulations de procedes et de dispositifs, nous avons montre que ce phenomene pouvait induire une redistribution des dopants a longue distance. L'outil que nous avons developpe apporte donc une aide a l'optimisation du procede de siliciuration dans les technologies silicium

Dans la region interfaciale, on retrouve toutes les configurations possibles de liaisons autour du tetraedre de si. Les profils de concentration de ces especes different selon que l'on fait l'oxydation ou l'oxynitruration. La distribution de ces especes a l'interface suit le modele theorique r. B. M. (random bonding model). Ce modele permet de comprendre la morphologie de l'interface et donc d'etablir la nature des liaisons chimiques entre les differents atomes (si-o-si et si-n-si)

Crystallization is a key unit operation used for obtaining purified products by many process industries. The key properties of the crystalline products, such as size and shape distribution, purity and polymorphic form are controlled by the crystallization process. All these properties impact significantly the downstream operations such as drying or filtration. Therefore, monitoring and controlling this process is fundamental to ensure the quality of the final product. Process analytical technology (PAT) brings numerous new methods and opportunities in the process analytics and real time process monitoring systems, which can be integrated into the control algorithm and provide high level optimal control strategies as well as deeper understanding of the process. Process monitoring helps develop mathematical models which can, in one hand, help in better understanding the processes and consecvently the development and application of advanced control methods in order to achieve better product quality. In this work, image processing and image analysis based direct nucleation control (IA-DNC) is developed in order to investigate the evolution of the crystal properties, such as crystal size, and crystal shape distribution. The IA-DNC approach is also compared to alternative DNC techniques, in which particle number were measured by Focused Beam Reflectance Measurement (FBRM) in order to control crystal size. A control approach is introduced that control the nucleation and disappearance of crystals during cooling and heating segments related to the changes of the number of counts (measured by Particle Vision Measurment, so called PVM or combination of FBRM and PVM). The approach was applied to investigate crystallization of compounds with different behavior: potassium dihydrogen phosphate (KDP) water, contaminated KDP -water and Ascorbic acid water systems. The results demonstrate the application of imaging technique for model-free feedback control for tailoring crystal product properties. The second main aim of the thesis is to investigate and control crystallization processes in impure media in the presence of multiple impurities, with an impact on the crystal shape via growth kinetics. The broad impact of the crystal growth modifiers (impurities) on the growth kinetics is observed in real time by using in situ video imaging probe and real-time image analysis. A morphological population balance model is developed, which incorporates a multi-site, competitive adsorption mechanism of the impurities on the crystal faces. The kinetic parameters of primary nucleation, growth and impurity adsorption for a model system of potassium dihydrogen phosphate crystallization in water in the presence of two impurities, were estimated and validated with experimental results. It was demonstrated that the model can be used to describe the dynamic evolution of crystal properties, such as size and aspect ratio during crystallization for different impurity profiles in the system. Manual, feedback and hybrid feedback-feedforward control techniques are developed and investigated numerically for continuous processes, while model-based and model-free control approach for crystal shape are developed for batch processes. The developed morphological population balance model is implemented and applied in the model-based control approaches, which are suitable to describe multicomponent adsorption processes and their influence on the crystal shape. Case studies show the effectiveness of crystal growth modifiers based shape control techniques. Comparison of different control approaches shows the effectiveness of the techniques. The third part of the thesis deals with purification of crystals when adsorption of impurities on crystal surfaces and its incorporation into crystals are considered. A purification method, called competitive purity control (CPC) is proposed and investigated. A morphological population balance model, including nucleation, growth and competitive impurity adsorption kinetics is developed to describe the case when multiple impurities can adsorb competitively on the crystal surface. The model is also combined with liquid phase chemical reaction model, in order to investigate the purity control case when an additive is introduced in the system that reacts with the impurity forming a non-adsorbing reaction product. Both competitive purity control approaches proposed: the adsorption based competitive purity control (A-CPC) and the reaction based competitive purity control (R-CPC); are investigated using detailed numerical simulations then compared with the alternative widely used purification method, called recrystallization. In the last contribution chapter, an integrated process optimization of a continuous chemical reactor and crystallizer is performed and studied numerically. The purpose of this study is to show the way in which the byproduct produced in the chemical reactor may affect the crystallization process and how its negative effect can be reduced by applying integrated process optimization. Sensitivity analysis of the system was performed by considering the flow rate and the concentration of substances in the input stream of the chemical reactor as manipulated process variables. Model based integrated process optimization and the sensitivity analysis in order to obtain improved quality product in terms of crystal size, shape and purity.

Ce travail de thèse porte sur la modélisation de la diffusion du Be dans les couches en InGaAs obtenues par épitaxie par jets chimiques (EJC). Cette diffusion peut se produire au cours des traitements thermiques de dispositifs tels que le transistor bipolaire à hétérojonction (TBH) InGaAs/INP ce qui limiterait les performances fréquencielles de ce dispositif. Dans un premier temps, les mécanismes de diffusion à l'état solide ont été abordés. Une attention particulière a été accordée au mécanisme substitutionnel-interstitiel de diffusion (SID). Dans le cadre de la technologie bipolaire III-V, les procédés d'épitaxie par jets chimiques et de gravure ionique réactive (GIR) sont également explicités. D'autre part, la caractérisation de la diffusion par spectrométrie d'émission d'ions secondaires (SIMS) s'est avérée essentielle pour la validation de nos modèles théoriques. Pour expliquer les résultats expérimentaux, un nouveau mécanisme de diffusion interstitiel-substitutionnel généralisé est proposé. Ce mécanisme suppose la diffusion simultanée selon les modèles dissociatif et kick-out de la diffusion interstitielle-substitutionnelle. Le changement de la dépendance fonctionnelle du coefficient de diffusion effectif du Be en fonction de sa concentration dans le volume de l'échantillon est expliqué par l'effet du niveau de Fermi

M.Sc.
Impurity resonance scattering effects are investigated in the Cr-Ga alloy system. This system has a triple point on its magnetic phase diagram where the paramagnetic (P), incommensurate (I) and commensurate (C) spin-density-wave (SDW) states co-exist. Alloying Cr with the nonmagnetic nontransitional element Ga affects the magnetic properties of Cr in a very unique way. In order to investigate the presence of resonant impurity scattering effects in binary Cr-Ga alloys, electrical resistivity measurements were carried out in the temperature range between 6 K and 85 K. The results of the investigation show: • A nonmonotonic increase in the residual resistivity of the Cr-Ga system with an increase in the Ga content, due to the presence of resonant impurity scattering of conduction electrons. • A low-temperature resistivity minimum observed in some of the Cr-Ga alloys, taken as further evidence for the presence of resonant impurity scattering effects on the conduction electrons. The impurity resonance scattering effects on the electrical resistivity of a Cr + 1.2 at.% Ga alloy, doped with V and Mn to tune the Fermi level through the impurity level, are also investigated. The investigation was complemented by thermal expansion and velocity of sound measurements in the temperature range 77 K to 450 K for the Cr + 1.2 at.% Ga alloy only. This specific Ga concentration was chosen to allow for studying resonant scattering effects in both the ISDW and CSDW phases of the system. This is possible because concentration of 1.2 at.% Ga is just above the triple point concentration. Doping with Mn to increase the electron concentration (eA) drives the alloy deeper into the CSDW phase region of the phase diagram, while doping with V, on the other hand, will drive the alloy towards the ISDW phase region. The results of the study are summarized as follows: • Two relatively sharp peaks, attributed to resonant impurity scattering effects, are observed in the curve of the residual resisitivity as a function of dopant concentration in the ISDW phase of the ternary (Cr0.988Ga0.012)1-xVx and (Cr0.988Ga0.012)1-yMny alloy systems. v • At 0 K the (Cr0.988Ga0.012)1-yMny alloy system transforms from the ISDW to the CSDW phase at y ≅ 0.0032, giving a CSDW phase for y > 0.0032. A peak is observed in the residual resistivity at about this Mn content. This peak can then either be ascribed to a jump occurring in the residual resistivity when the CSDW phase is entered from the ISDW phase or to resonant scattering effects. The conclusion is that the peak is rather related to the latter effect. • The resistivity as a function of temperature of the above two ternary alloy series show well-developed or weak minima at low temperatures for some of the samples. This is taken as further evidence of the influence of impurity resonant scattering effects on the resistivity of these alloys. • The resistivity and thermal expansion coefficient of the polycrystalline Cr0.988Ga0.012 alloy of the present study behaves anomalously close to the ISDW-CSDW phase transition temperature and warrant further investigation. The concentration-temperature magnetic phase diagram of the (Cr0.988Ga0.012)(Mn,V) alloy system was constructed from the magnetic transition temperatures obtained from electrical resistivity measurements. Theoretical analysis of the phase diagram was done using the two-band imperfect nesting model of Machida and Fujita. The results show: • A triple point at (0.21 at.% V, 225 K) where the ISDW, CSDW and P phases coexist on the magnetic phase diagram. • The curvature of all three theoretically calculated phase transition lines in the region of the triple point is of the same sign as that observed experimentally. • The theoretical fit is very good for the ISDW-P and ISDW-CSDW phase transition boundaries, while there is some discrepancy for the CSDW-P phase transition line. This may be attributed to the fact that the theory is one dimensional and that it does not include electron-hole pair breaking effects due to impurity scattering and also not effects of changes in the density of states due to alloying.
Dr. A.R.E Prinsloo Prof. H.L. Alberts

The present project has made a comprehensive assessment of the effect of Nb doping on various charge-transfer related properties of TiO2. Of particular focus, the electrical properties of Nb-doped TiO2 (0.65 at %) have been investigated using the simultaneous measurement of electrical conductivity and thermoelectric power. This investigation was undertaken at elevated temperatures (1073 K -- 1298 K) in equilibrium with a gas phase of controlled oxygen activity (10-10 Pa < p(O2) < 75 kPa). In addition, the effect of segregation on the surface versus bulk composition of Nb-doped TiO2 was also investigated at a function of temperature and oxygen activity. Specifically, the following determinations were undertaken: The effect of oxygen activity, p(O2) and temperature on both electrical conductivity and thermoelectric power The effect of Nb on the defect disorder and related electrical properties of TiO2 The determination of equilibration kinetics and the associated chemical diffusion data for Nb-doped TiO2 The determination of Nb bulk diffusion in TiO2 The effect of p(O2), temperature and dopant content on Nb segregation and the related surface composition of Nb-doped TiO2 The obtained electrical properties enable the determination of a defect disorder model for Nb-doped TiO2, which may be considered within the following p(O2) regimes: Strongly Reduced Regime. In this regime, the predominant ionic defect was anticipated to be oxygen vacancies compensated electronically by electrons. While the transition to this regime (from higher p(O2)) was clearly observed, the predominant defect disorder existing beyond this transition was not confirmed due to an inability to obtain sufficiently low oxygen activity. Metallic-type conductivity behaviour was observed within this transition region. Reduced Regime I. In this regime, the predominate defect disorder defined by the electronic compensation of incorporated Nb ions by electrons was clearly observed. Reduced Regime II. In this regime, the predominate defect disorder defined by the ionic compensation of incorporated Nb ions by quadruply-charged titanium vacancies, was clearly observed. The present project included the determination of diffusion data which included: Temperature dependence of 93Nb tracer diffusion in single crystal TiO2 over the temperature range 1073 K -- 1573 K Chemical diffusion coefficient over the temperature range 1073 K -- 1298 K and oxygen activity range, 10-10 Pa < p(O2) < 75 kPa These pioneering studies are significant as they enable the prediction of the processing conditions required to reliably 1) incorporate Nb into the TiO2 lattice, and 2) achieve equilibrium with the gas phase. Finally, the present project included investigations on the effect of Nb segregation on the surface composition of Nb-doped TiO2, with the following outcomes: Due to segregation, the surface can be significantly enriched in Nb compared to the bulk The extent of enrichment increases as the bulk Nb content or the oxygen activity is decreased Following enrichment, the surface Nb concentration could be sufficiently high to assume a unique surface phase The outcomes of the present project are significant as they can enable the processing of TiO2 with enhanced charge transport and controlled surface properties.

A primarily software based Fourier Deep Level Transient Spectroscope (FDLTS) is built. The raw capacitance transient is acquired and digitized using capacitance meter HP4280A whereas the signal analysis is done using a customized software module. The software module calculates both the conventional DLTS spectrum and the Fourier DLTS spectrum. This home-made FDLTS set up was compared to a commercial conventional box-car DLTS system (Sula Technology's DLTS) as well as to a commercial Fourier DLTS system (Bio-rad) and it was found to be either equivalent to the commercial systems or even better in some respects. In one case, Fourier analysis using the home-made setup, led to the detection of a trap completely undetected by the commercial conventional DLTS. The FDLTS system together with the commercial conventional DLTS were used to study possible gold contamination in an industrial process. The study was accomplished by comparing conventional and Fourier DLTS spectra and corresponding calculated trap properties using Schottky barrier diodes fabricated on the suspect wafers and an intentionally gold diffused reference sample wafer. During the investigation minority carrier emission in DLTS using Schottky barrier diodes was observed. The study revealed the presence of some possible gold-like contamination which trapped minority carriers (i.e. electrons) in p type silicon.
Graduation date: 2003

碩士
國立中央大學
機械工程研究所
100
Directional solidification (DS) method is frequently used to produce the mc-Si ingots in the PV wafer industrial. The efficiency of an mc-Si solar cell depends strongly on the impurity content and level of the mc-Si wafers. In this study, numerical simulation of the transient temperature, velocity and concentration of oxygen and carbon and silicon carbide has been carried out in order to clarify the influence of gas shield on impurity transport of oxygen, carbon and silicon carbide in a directional solidification system furnace during the growth process. The installation of such a gas flow guidance device changes the gas and melt flow pattern in the furnace, which would affect the transport of carbon oxide and silicon oxide in the gas and oxygen and carbon in the melt. As the solidification fraction enlarges, the oxygen concentration in the melt decreases, because of the reduction in the amount of crucible surface immersed below the silicon melt. Results show that a decrease of the relative position of gas shield and gas-melt interface reduces the oxygen concentration in the melt but enlarges the carbon impurity. The distance between gas guidance device and free surface are presented using 2.7cm, 5.5cm, 8.3cm, 11cm, 13.8cm. The distance with 2.7cm can reduce oxygen concentration about 30% and show the minimum oxygen concentration in the ingot. Carbon concentration decreases with reducing height in the ingot, but carbon concentration increases when the height of gas shield changes from 5.5cm to 2.7cm. Thus, the optimum position to install the gas flow guidance device for obtaining the best carbon and oxygen levels in the ingot during the solidification process is 5.5cm.

博士
國立中央大學
機械工程研究所
99
The silicon (Si) solar cell is still the highest market share nowadays. To accomplish the goal of grid parity, for the solar cell, the wafer quality has to improve, and the production cost has to reduce. The high concentration of carbon and oxygen impurity cause the dislocation and thermal donor in the mc-Si wafer, respectively. The carbon and oxygen are the main impurity for affecting the efficiency of solar cell. The simulations of thermal flow field, carbon concentration, oxygen concentration and silicon oxide concentration in directional solidification system (DSS) are carried out by the finite volume method (FVM) in this study. The distributions of carbon and oxygen concentration in the grown ingot were measured by the SAS Company and the measurement results are compared with that of the simulation predictions. The simulation results are in good agreement with the experimental ones. The simulation shows that the melt convection is induced by buoyancy force. The flow pattern in the melt changes during the growth process. In order to improve the uniformity of carbon distribution in the melt, a heat insulation of crucible is used to increase the temperature gradient and vortex intensity of melt. Using wafers of the whole ingot obtained from the modified case, the average conversion efficiency of solar cells can be improved up to 1.8% of the one of standard case. As the solidification fraction enlarges, the oxygen concentration in the melt diminishes, because of the reduction in the amount of crucible surface immersed below the silicon melt. When the solidification fraction is small, the oxygen concentration is higher with a higher furnace pressure than with a lower one due to there being less SiO evaporation at the free surface. When the solidification fraction increases, because of the cooling effect of the argon gas, the oxygen concentration is smaller when the furnace pressure is higher rather than lower. Hence, to adjust variably the furnace pressure during the mc-Si ingot growth is good method for reducing the oxygen concentration. To increase the argon flow rate can bring more evaporated SiO gas above the free surface outwards the furnace; hence the oxygen concentration in the melt is decrease. In the present study, the gas guiding plate is used to increase the argon velocity above the free surface. The effect of reducing oxygen concentration for the gas guiding plat is similar to the one of increasing the 25% amount of original argon flow rate. The gas guiding plate can decrease the oxygen concentration without increase the argon flow rate, the furnace enhancement can be used on advancing the wafer’s quality.

碩士
國立臺北科技大學
電機工程系
106
In recent years, the technology of remote sensing has improved significantly. To solve the problem of over-sized data and bands, band selection method has been widely utilized to reduce the dimensions of images. In this way, Hughes phenomenon can be avoided and the computation time can also be reduced. Two-dimension Impurity Function Band Prioritization (2D-IFBP) method has been proposed by other researcher. It utilizes Mutual Information (MI) and bivariate normal distribution to calculate the overlapping rate between two classes from the two-dimensional domain. In addition, 2D-IFBP is used to modify the band order of IFBP to obtain the better result. In this thesis, some other bivariate distributions will also be applied such as uniform distribution, exponential distribution, and T distribution to evaluate its characteristic of Mutual Information. In this experiment, the data sets of Salinas, Washington DC mall and Pavia University are used to verify the influence of different distributions on 2D-IFBP. The experimental results show that the first and the second data sets are more like normal distribution but the data set of Pavia University is more like T distribution. Moreover, T distribution has the same dimension reduction rate as normal distribution but it possesses the better accuracy rate than the other one. Hence, trying to utilize different distributions for the specific data sets is more appropriate than conventional method.