Symposium : A
Advanced Silicon Materials Research for Electronic and Photovoltaic Applications III
| start at | Subject | Num. |
|---|---|---|
| Solar Grade Silicon I : B. Ceccaroli and T. Abe | ||
| 08:30 | New Advances in Metallurgical Solar Grade Silicon Based on Recent Norwegian Research Activities Authors : Eivind Øvrelid, Merete Tangstad, Bruno Ceccaroli, Otto Lohne, Gabriella Tranell Affiliations : Eivind Øvrelid (Department of Metallurgy, SINTEF Materials and Chemistry, Trondheim, Norway), Merete Tangstad (Department of Materials Science and Engineering Norwegian University of Science and Technology) Bruno Ceccaroli (Marche AS, Kristiansand, Norway), Otto Lohne (Department of Materials Science and Engineering Norwegian University of Science and Technology, Trondheim, Norway), Gabriella Tranell (NTNU) Resume : The speech will give a short introduction to the well described processes for production of metallurgical grade silicon. Among the new routes for the production of solar grade silicon, the metallurgical refining techniques leading to upgraded metallurgical silicon emerge as serious alternatives to replace the feedstock produced by the Siemens process (commonly called Polysilicon). Most of these new alternative processes consist generally of several successive steps. The most achieved and frequently used among these refining processes will be described. Some recent results based on research studies carried out in Norway will be presented with more details. The intention is to give an overview and an understanding of these emerging processes and provide some tools to optimize and further develop new cost efficient processes for production of both solar grade silicon and solar cells. | 11 1 |
| 09:00 | New concept for the reduction in the number of grains in crystalline silicon grown by directional solidification: optical and electrical characterization Authors : B. Moralejo1, V. Hortelano1, O. Martínez1, J. Jiménez1, A. Piñeiro2, E. Diéguez2, V. Parra3 Affiliations : (1) GdS-Optronlab, Dpto. Física Materia Condensada, Edificio I+D, Paseo de Belén, 1, 47011 Valladolid, Spain. (2) Crystal Growth Laboratory. Universidad Autónoma de Madrid. Ctra de Colmenar km 15. 28049 Madrid, Spain. (3) DC Wafers Investments, S.L. Ctra. de Madrid, Km. 320, 24227, Valdelafuente (León), Spain. Resume : Nowadays, the photovoltaic (PV) market is mostly governed by silicon-based devices. Among them, multi-crystalline Silicon (mc-Si) solar cells, with lower cost of the devices, take the advantage over the mono-crystalline Silicon (c-Si) cells, in spite of the detriment in the energy production efficiency due to structural defects inherent to mc-Si, such as grain and sub-grain boundaries, dislocations, etc. The improvement of the structural properties of the material is a key issue in further development of mc-Si based PV. Recently, the idea of growing silicon ingots with mono-crystalline features but using industrial casting furnaces (directional solidification – DSS – or heat exchange method – HEM –) constitutes an exciting milestone for the future of the PV technology. In this paper, the special features of silicon wafers, manufactured according to an updated casting method, are characterized by optical – electroluminescence (EL) and photoluminescence (PL) – and electrical – light beam induced current (LBIC) and electron beam induced current (EBIC) mappings – methods. The use of these techniques allows combining the characterization of the full wafer in only a few seconds with a very detailed characterization of the structural defects due to the high spatial resolution and sensitivity to charge capture of the LBIC and EBIC techniques. The results show that in spite of the reduction of the number of grains, the key factor continues to be the presence of dislocations and sub-grain boundaries, which can be present in substantial concentration in the quasi mono wafers. The understanding of the role of these defects, and how are they generated can permit improvements in the growth process. | 11 2 |
| 09:15 | Influence of Back-Diffusion of Iron Impurity on Lifetime Distribution near the Seed-Crystal Interface in Seed Cast-Grown Monocrystalline Silicon by Numerical Modeling Authors : Bing Gao, Satoshi Nakano, Koichi Kakimoto Affiliations : Research Institute for Applied Mechanics, Kyushu University Resume : The casting technique for producing monocrystalline silicon has recently become popular due to its cost-efficiency, high throughput and large-scale processes. The quality of monocrystalline silicon produced by this technique depends on the characteristics of the seed-crystal interface, such as seed-crystal interface shape and seed-crystal interface structure. Study of relationship between crystal quality and seed-crystal interface is helpful to optimize and improve the casting process during monocrystalline silicon production.
Since monocrystalline silicon is mainly used to produce solar cells, minority carrier lifetime is an essential parameter for measuring crystal quality. Thus, quality of the grown crystal at the interface can effectively be indicated by the distribution of minority carrier lifetime across the seed-crystal interface. Witting et al. measured the distribution of minority carrier lifetime across the seed-crystal interface in cast-grown monocrystalline silicon [1]. They found that the lifetime has a minimal value near the seed-crystal interface and gradually increases towards the seed and towards the interior of the grown crystal. From the seed to the grown crystal, the lifetime exhibits a high-low-high distribution within a narrow area. Salo et al. also found a similar phenomenon by measuring structural characteristics of the seed-crystal interface of rapidly grown KDP crystals using the X-ray diffraction method [2]. They found a transitional zone of more than 12 mm in width near the seed-crystal interface, with an increased concentration of defects and nonmonotonic variation of the crystal lattice parameter.
Experiments have shown that a large scale defect type was observed near the seed-crystal interface [1]. These defects are small angle grain boundaries with a high density of dislocations along them [1]. They have high recombination activity and disrupt the crystallinity of the ingot [1]. It is essential to determine their cause to grow high-quality monocrystalline silicon by casting process.
This work wants to clarify the particular distribution of minority carrier lifetime near the seed-crystal interface: first decreases and then increases from a seed to a crystal. Modeling of the temperature- and time-dependent iron diffusion and segregation during crystal growth showed a concentration distribution of an increase followed by a decrease from a seed to a crystal. The consistency between lifetime and iron distribution near the seed-crystal interface indicates that the back-diffusion of iron impurity from silicon melt into the seed at the duration stage before crystal growth is one of the main reasons for lifetime variation near the seed-crystal interface. Therefore, it is essential to reduce the duration time before crystal growth to obtain good-quality monocrystalline silicon.
Acknowledgments:
This work was partly supported by the New Energy and Industrial Technology Development
Organization (NEDO) under the Ministry of Economy, Trade and Industry (METI).
[1] I .Witting, N. Stoddard, G. Rozgonyi, Proc. 18th Workshop Cryst. Silicon Sol. Cells Modules 2008, 155−158.
[2] Vitaly I.Salo, V. F. Tkachenko, Marina I. Kolybayeva, Proc. SPIE 1999, 3578, 519. | 11 3 |
| 09:30 | From silicon kerf to solar grade Silicon: a route to produce low-cost PV ingots Authors : Ilaria Lombardi*, Federico Tappa, Guido Fragiacomo, Sergio Pizzini Affiliations : Ilaria Lombardi*, Federico Tappa, Guido Fragiacomo Garbo Srl, Via Prati Nuovi 9, 28065 Cerano, Italy Sergio Pizzini University of Milano Bicocca, via Cozzi 53, 20125 Milano, Italy Resume : The new diamond wire sawing technology of silicon wafers offers improved performances and costs benefits over the traditional loose slurry method and it is increasingly gaining consensus in the industry. The possibility of recycling the silicon sawdust produced during the slicing operation is appealing both from an economic as well as from an environmental point of view, paving the way for the production of a low cost solar silicon feedstock and increasing the competiveness of the solar power. Garbo has developed a technology for the recycling of fine silicon sawdust which involves the chemical cleaning of the surface impurities and the following compaction to yield granules or bricks with optimized mechanical and packing density properties. In this work, chemical and physical characterization of the recycled feedstock is addressed. Crystallization results via an ad-hoc directional solidification method of the recycled solar grade silicon kerf are presented and discussed. Specifically, an effective DSS apparatus and process is implemented in order to reduce carbon and oxygen melt contamination deriving from gaseous contaminants in the furnace, particularly severe for large surface to volume ratio of the compacted powders. it is shown that modifications of the conventional thermal process allows to obtain ingots with indistinguishable features compared to that obtained with high purity polysilicon. | 11 4 |
| 09:45 | Coffee break | |
| Solar Grade Silicon II : E. J. Ovrelid and I. Yonenaga | ||
| 10:15 | New Advances in Polysilicon Processes Correlating Feedstock Properties and Good Crystal and Wafer Performances. Authors : Bruno Ceccaroli (1), Otto Lohne (2), Eivind Øvrelid (3) Affiliations : (1) Marche AS, Kristiansand, Norway; (2) Department of Materials Science and Engineering Norwegian University of Science and Technology, Trondheim, Norway; (3) Department of Metallurgy, SINTEF Materials and Chemistry, Trondheim, Norway Resume : Metallurgical purification routes to solar grade silicon have gained a respectable success for the manufacture of standard multicrystalline solar cells. However, gas phase chemically refined silicon (commonly designated as Polysilicon) remains unchallenged for semiconductor applications and for the manufacture of monocrystalline solar cells, particularly the most popular high efficiency cell architectures. Therefore, in the medium term future virgin Polysilicon should keep a predominant share of the silicon feedstock to electronic and photovoltaic wafers. For photovoltaic to meet the grid parity cost target, significant efforts have been deployed during the past years aiming at drastically reducing the cost of Polysilicon. This has been achieved not only through massive capacity expansions but also through simplifying the process as well as developing new gas/solid purification and deposition processes. The replacement of the historical Siemens by the Fluidized Bed Reactor is the most accomplished example of such attempts. Worth noting is also the replacement of chlorine by other halogens in the volatile silicon bearing compound or hydrogen by metal in the deposition/reduction process. However, such cost efficient progress might have been on the expense of the ultimately desirable purity. This contribution will review the most accomplished and promising attempts to develop the Polysilicon process. The potential and the limitations of the emerging Polysilicon products will be illustrated with respect to their ability to produce high quality wafers for semiconductor or high efficiency solar cells. | 12 1 |
| 10:45 | STRUCTURE OF RIBBON SILICON: ELECTRON MICROSCOPY ANALYSIS Authors : Edward Tsidilkovski*, Sergei Rouvimov** Affiliations : *) Evergreen Solar Inc., Marlborough MA, USA; **) Portland State University, Portland OR, USA Resume : Continuing increase in commercial solar cell efficiency raises the importance of the substrate material’s quality, enhancing its effect on photovoltaic performance. The properties of solar cells made from multi-crystalline silicon are strongly affected by structural defects including dislocations, grain boundaries, impurities and precipitates. Specifics of the defect formation and distribution in ribbon silicon wafers stems out of peculiar growth conditions, including high temperature gradients and low ratio of melt volume to ribbon surface. Carbon impurity introduced into wafers from the graphite parts of the furnace hot zone plays especially important role in defect formation in ribbon silicon. Both the impurity and defect concentrations may vary across a ribbon wafer as a function of growth and processing conditions. Here we present an analysis of the defects in ribbon wafers and review the mechanisms of the defect formation. Using electron backscattered diffraction combined with high resolution TEM we have identified the dominant defect types, their structural features, and their characteristic spatial distribution with respect to the crystallographic orientations. It was shown that in spite of high non-equilibrium concentration of the light element impurities, these elements do not precipitate in the ribbon material. Correlation to the lifetime maps allowed determination of the electrically active defects, thus providing an opportunity for effective defect engineering. | 12 2 |
| 11:00 | Exploring polysilicon deposition conditions through a laboratory Siemens prototype. Authors : A.Ramos, J.Valdehita, J.C.Zamorano, C.Cañizo, A.Luque. Affiliations : Instituto de Energía Solar - Universidad Politécnica de Madrid Resume : Since 2006, polysilicon for photovoltaics -PV- has a higher share of the demand than that for microelectronics. This is changing not only the market but also the requirements: polysilicon cost impacts significantly on the PV cost and on the energy payback time. On the other hand, the level of silicon purification for photovoltaic applications is not as critical as for microelectronics, allowing process simplifications that can lead to cost reductions. Although a “metallurgical route” has been proposed as an alternative to produce the so called “solar silicon”, the “chemical route” based on the synthesis, distillation and deposition of chlorosilanes is the dominant one.
Work and research performed at the Instituto de Energía Solar on this topic is centered on the chemical route, specifically on the polysilicon deposition step by chemical vapor deposition (CVD) from trichlorosilane.
This CVD deposition accounts for the largest contribution to the energy consumption of the whole process.
To achieve the objectives of lower production cost and low energy payback time, in-depth knowledge of the production process is needed. Polysilicon deposition process is a quite complex process if looking at the big number of parameters involved.
The work that will be presented comprises research on solar silicon production, exploring CVD deposition conditions through a laboratory CVD prototype and characterizing the chemical reactions of the deposition process.
Our laboratory prototype can reproduce all working conditions of the industrial process except the pressure inside the reactor chamber. As the relation between the pressure and the main parameters involved in the process is straightforward and can be theoretically stated, the laboratory prototype can give valuable information about the phenomena involved in the deposition and the operation conditions.
We report in this paper on some of the experiments that we have developed, and the learnings involved.
In particular:
1. The influence of the temperature on the growth rate.
The experiments conducted in this laboratory prototype reactor can be performed at temperatures of the deposition surface up to 1250 ºC. Several experiments have been successfully carried out, which evidence that the growth rate increases slower in the low temperature regime of the deposition surface (900–1100ºC) than in the high temperature regime (above 1100ºC), as will be shown.
2. Characterization of chemical reactions monitored by a mass spectrometer.
Working with trichlorosilane - HSiCl3- and Hydrogen -H2- (reactive gases of the polysilicon deposition reaction) is not straightforward as it can result on damage of the measurement devices in a mass spectrometer. Introducing some cares to avoid this issue, a quantitative analysis of the spectra is successfully accomplished.
Cautions related to the compatibility with the operating conditions are reported, as well as quantitative results and a comparison with the deposition efficiency.
3. Deposition process conditions that cause the appearance of ‘pop-corn’ phenomenon.
There are different conditions that can favor dendritic growth (known as “pop-corn effect”): heat, gas flow velocity, radiation flux and mixture composition. Experiments developed with this prototype reactor prove that ‘pop-corn’ is strongly dependent with temperature. It has not been possible to avoid it in long deposition processes at temperatures above 1100 ºC.
However, although the pop-corn is believed to be an undesired phenomenon, our experiments show that the polysilicon growth rate is higher if there is dendritic growth than if the deposition is more homogeneous. Advantages and disadvantages of this phenomenon will be set out.
In summary, quantitative results of CVD polysilicon deposition from trichlorosilane will be presented from a set of experiments that cover a range of deposition conditions. Both the physical and chemical basics and the applications to optimize the deposition performance will be described. | 12 3 |
| 11:15 | Silicon processing for photovoltaics based on incoherent radiation power Authors : J.M.Serra Affiliations : Faculdade de Ciências da Universidade de Lisboa/SESUL, Campo Grande Ed-C8,1749-016 Lisboa, Portugal Resume : It is well known that impurity contamination is a very important factor in silicon processing. Contamination arises from crucibles or graphite susceptors during crystal growth, from the use of fibers or dies as in the case of string ribbon or efg technologies respectively.
Every hot part in a crystal growth system will contribute to increase impurity contamination.
Optical processing using halogen lamps provides a clean method, with available fast rise and fall ramps, with great flexibility in many situations related to materials processing and in particular in photovoltaics. Optical power provided by halogen lamps is very cheap, which is an important aspect for an industry that needs to reduce process costs. Techniques based in lamp power radiation have been in use in our laboratory at the University of Lisbon for many years. We want to draw the scientific community to some advantages of this type of optical processing. Several examples of the use of such optical techniques are presented, illustrating its application to silicon ribbon growth, partial melting control in thin wafers and as a doping technique for junction formation. An example is also shown on its successful application for photovoltaics in a gas-to-wafer path, bypassing both the Siemens and the traditional wafering steps. | 12 4 |
| 11:30 | Short Presentations Graduate Student Award Authors : Applicants Graduate Student Award Affiliations : - Resume : - | 12 5 |
| 12:15 | Lunch break | |
| Czochralski Silicon Crystal Growth : W. von Ammon and J. Vanhellemont | ||
| 14:00 | A new model for intrinsic point defects in silicon crystals grown from the melt: Questioning v/G model Authors : Takao Abe, Toru Takahashi and *Koun Shirai Affiliations : Shinetsu Handotai, Isobe R&D Center、2-13-1 Isobe, Annaka, Gunma 379-0196, Japan *ISIR, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0047 Resume : Almost all on intrinsic point defects have postulated the coexistence of vacancies and silicon interstitials near the growth interface. It is generally considered that in the vicinity of interface vacancies are abundant species and move fast, while interstitials are rare species and move slowly. On this basis, people tried to construct the formation mechanisms of secondary defects such as A- and D- defects. Recently, we have established by experiment and theory that the temperature gradient at the interface decreases as the pulling velocity increases, and vice verse. The impact of this observation on the formation mechanism of secondary defects is not small, because the most important part of formation of secondary defects occurs at that region. Thermal stresses induced by the temperature gradient could influence on the formation of defects. This effect has not, so far, been considered seriously. In particular, this is true by considering that thermal stress is favorable for formation of interstitial. On the other hand, there is evidence that an enough amount of vacancies are always present just at the interface. Only when the temperature gradient is large enough (this is the condition for slow growth rate), I-rich regions can be formed by overcompensating the pre-existing vacancies by an enough amount of interstitials. These are derived by our observation on the kinetics of reaction between defects by changing the growth rates. | 13 1 |
| 14:30 | Optimal Design of Czochralski Process for Single Crystal Silicon Ingot with Oxygen Concentration Profile Authors : YuJin Jung and Jae Hak Jung Affiliations : School of Chemical Engineering, Yeungnam University, Gyeongsan 712-749 Resume : Growth of single crystals silicon ingot by Czochralski process is always related to transport of impurities. Oxygen is one of the most important impurities in Czochralski grown silicon crystals for solar cell and electronic devices. And oxygen impurity usually provide from furnace elements which have SiO and SiO2 deposit layers. In this study, a modified numerical model of oxygen transport as a way of prediction of oxygen concentration in growing Si crystals is described. The proposed numerical model accounts for oxygen transport in the melt and SiO deposition on the walls of the furnace. Test calculations for the updated model were performed for a Czochralski process setup with a small crucible diameter and weak turbulent melt flow. The calculated oxygen profile along the crystallization front is compared with available experimental data. 2-D calculation results have evaluated the number of finite volumes in 2-D grid, required for precise description of oxygen transport in the small growth system. In this study, the numerical predictions performed with the CGSim software package agree well with available experimental data obtained in optimized crystal growth process reported for the first time. Finally we can give the direction of optimal design of Czochralski process which can reduce the oxygen impurity in the ingot products. | 13 2 |
| 14:45 | The effect of mass transfer on the temperature gradient of a crystal growing from melt Authors : Koun Shirai, Takao Abe Affiliations : ISIR, Osaka University: Shin-Etsu Handoutai Resume : The change in the temperature gradient on the crystal side while the rate of crystal growth from melt is varied has long been debated. Abe and Takahashi have recently reported an unambiguous experimental demonstration that the temperature gradient is a decreasing function of the growth rate, which contradicts the theories, experimental results, and widely held notion of other researchers.
The present paper provides a theoretical basis for this seemingly peculiar effect of the growth rate on the temperature gradient. The essential matter is the effect of mass transfer, the role of which had been commonly disregarded in old studies. Although the rate of mass transfer is not large compared to that of heat conduction, it is proven that the temperature gradient is subjected to the mass transfer in a definite manner. Our analysis shows that the effect becomes significant when the crystal diameter is large, which is consistent with the experimental observation.
Another effect of the mass transfer is the change in the shape of melt/crystal interface. In old studies, the temperature gradient was determined by Stefan's equation: however this treatment confuses the cause and effect. The temperature gradient should be determined by the fundamental equation of heat conduction. When the gradient is determined in this way, the shape of the melt/crystal interface spontaneously adjusts to satisfy Stefan's equation. | 13 3 |
| 15:00 | Growth and Characterization of Silicon-Germanium crystals Authors : Ichiro YONENAGA Affiliations : Institute for Materials Research, Tohoku University Resume : Silicon-germanium (SixGe1-x) alloy is a complete solid-solution with the diamond structure and is the focus of keen interest as material for both microelectronic and opto-electronic devices and various functional materials because of the potential for the band-gap and lattice parameter/strain engineering. SiGe thin films are being used for high-speed microelectronic devices, and also bulk SiGe alloys for use as X-ray and neutron monchromators, solar cells, thermoelectric power generators and lattice-matched substrates. For such functional applications as well as for further utilization of the potential of the materials, it is essential to clarify their intrinsic physical properties. In addition, SiGe is quite interesting from various basic aspects: solidification of highly miscible alloy, distorted atomistic structure, carrier transport, defect physics, etc. Here, this paper reviews the Czochralski growth of bulk SiGe, discussing relevant problems, and covers the fundamental properties of the alloys as the atomistic bonding structure of quasi-Pauling type, electron and hole mobilities, thermal conductivity, O local-vibration, muonium, mechanical properties, etc. | 13 4 |
| 15:30 | Coffee break | |
| Back | ||
Show my program |
|
|
| Une réalisation advisa.fr |  |