Review: Physical Method Polysilicon in 2008

Thread Content

Reposted from: Dr. Shi’s BLOG Review: UMG in 2008. As a basic energy material that will find its way into thousands of households, and one that utilizes sunlight available everywhere on Earth, polysilicon surely has the potential for a cleaner and cheaper method of production. ----Regarding the phrase \"Martin Green?, or Bradley Shi\", some say it was spoken in 2004 by Martin Green, known as the \"father of solar energy in the world\", but I spent a long time without finding any concrete evidence to support this claim. If he himself didn’t say it, then consider it as if I said it; anyway, I wholeheartedly agree with this statement. Obviously, this sentence refers to the production of polysilicon using the Siemens process. And this “cleaner and lower-cost” method is the physical method, also known as the metallurgical method; it is referred to internationally as UMG. Hereafter, it will be referred to uniformly as physical method polysilicon. Regarding the difference between physical and chemical methods, the author provided a definition at the Silicon Materials forum of the 10th China Solar Photovoltaic Conference held in Changzhou in September 2008: any method in which a chemical reaction occurs within the silicon material itself is referred to as a chemical method ； Any method in which silicon does not undergo any chemical reactions throughout the entire treatment process is called a physical method. Given that silicon is the matrix material, it is reasonable and sufficient to use whether silicon undergoes a chemical reaction as the criterion for distinguishing between chemical and physical methods. To my relief, at the \"2008 International Conference on Solar-grade Polysilicon\" held in Shanghai on December 9, 2008, Professor Chen Chao expressed the same view, which can be regarded as academic recognition of this assertion. UMG, not UFO. Figure 1 was shown by a representative of Siemens at the \"2008 International Solar Grade Polysilicon Conference\" held in Shanghai in December 2008. As can be seen from this image, whether for semiconductor or photovoltaic applications, 75% of the polysilicon used in the world today is produced using this process. If one takes into account that the circulating fluidized bed method is essentially just an improvement on the Siemens method, and that both the zinc reduction method and the silane method are variations of the Siemens method, it can be seen that in 2005 UMG’s production was practically zero. By 2010, however, the share of the Siemens method dropped from 91% to 75%, while the share of physical methods (MG-SOG) increased from zero to 11%. Figure 1: The proportion of various processes in 2005 and 2010. This “11%” is Siemens’ estimate; I believe it underestimates the share of physical methods. However, given Siemens’ possible preference for the Siemens method, it is understandable to have an overestimation of one’s own capabilities and an underestimation of those of competing methods. But in any case, the fact that the physical method can stand out independently from the \"Other\" methods and be identified by its own name already demonstrates Siemens’ emphasis on this physical method. There are certainly sufficient reasons for this emphasis. The main reason is, of course, that physical methods inherently possess many advantages. For polysilicon production on a similar scale, the investment required for physical methods is only 1/5 of that for the Siemens method, and the production costs are only 1/2; moreover, there is no environmental pollution throughout the process. Contrary to what many people think, UMG is not a new technology, nor is it a technology of unknown origin. As early as 1935, physical methods were begun to be used for the purification of silicon. From then on, until 1975, research on physical methods reached a small **. At that time, it was found that the theoretical limit for purity achievable through physical purification methods was only 7N, whereas standard semiconductor devices require silicon purity of at least 9N. At that time, the Siemens method was invented, so the physical method was set aside and ignored for many years thereafter. However, after this method had remained dormant for nearly three decades, photovoltaic applications began to develop rapidly in 2004. Since 6N grade polycrystalline silicon is sufficient for solar energy applications, the metallurgical method was brought back from obscurity and new rounds of research began once again. In particular, the imbalance between supply and demand for polysilicon that began in 2006 led to more people turning to research on polysilicon produced by physical methods. Global 2008: The Dawn of a New Era. Before 2008, physical method polycrystalline silicon remained mostly in the realm of laboratories and academic papers. Although Japan’s JFE announced in 2004 that it had started mass production using electron beam and ion beam methods, from then on, at least until 2008, if JFE’s UMG production capacity still existed, it was not expanded or improved. JFE’s main customer is SHARP. Another source of SHARP’s UMG is Nippon Steel, and it is said that the technicians at Nippon Steel also come from JFE. Norwegian company Elkem also announced in 2007 that it had used a \"new, metallurgical-like\" method to purify polysilicon, but it faced some difficulties because it did not specify the potential degradation associated with this silicon material. The American company Corning also announced successful purification of polysilicon using metallurgical methods, although it can only mix this UMG with polysilicon produced by chemical methods in a 10% ratio for use in crystal pulling. Entering 2008, UMG seemed to slow down its activities; from the beginning of the year until May, there was little activity at all, to the point that many industry experts thought the physical method had disappeared completely and been rendered obsolete. It was TIMMINCO from Canada that broke this silence. This Toronto-based publicly traded company, which is involved in the processing of non-ferrous metals, has, since the end of April, cautiously introduced at some small-scale financial industry seminars the fact that its BSI plant uses metallurgical methods to purify polysilicon for use in solar cells. It was not until June that this news was made public in the mainstream media. This news caused TIMMINCO to instantly overtake Elkem and DC Chemical and become the star in the field of physical purification of polysilicon. The company has never officially released its own metrics and data, but PPT files from its various \"presentations\" are available on the Internet. The data in these files indicate that the company’s purity level is 5N, with boron levels below 0.8 ppm and phosphorus levels between 1 and 3 ppm. The cost is only 15 dollars, while the selling price is as high as 66 dollars. Of course, these are all “unofficial” reports. The company also announced that among its clients is the world’s largest solar cell panel manufacturer, namely Germany’s Q-Cell. These news items were certainly enough to drive the company’s stock price from 1 Canadian dollar to several dozen Canadian dollars. At the INTERSOLAR conference held in Germany in early June, another company registered in Canada with production facilities in China – CSI – announced that it had adopted UMG 100% in the production of E-series solar cells. These cells have an efficiency of 13.3%, and their price is about 15% lower than that of \"regular\" cells. The shock value of this news lies in the fact that what they announced was not research results, but rather that sales had already taken place, with the buyers being two companies from Germany and Italy. This means that UMG has entered the commercial phase, albeit in an “economical” manner. In November, two American companies announced one after another that they had produced solar cells using UMG. One is CaliSolar, located in California; the company has successfully raised $100 million to expand the production capacity of its UMG batteries. Another company is Blue Square Energy, which is trying to raise $25 million in funding. The UMG conversion efficiency of both companies exceeded 14%. At that time, the world entered a global recession due to the U.S. subprime mortgage crisis. The ability to secure financing on such a scale even in the context of an economic crisis also demonstrates UMG’s appeal. China: A challenging first half. Physical polysilicon in China came into focus starting in 2005. As early as 2006, many companies announced that they had successfully purified 6N polysilicon using metallurgical methods; however, these polysilicons produced by physical methods and claimed to be of high purity have never made it onto the market. By 2007, many more companies announced that they had successfully produced polysilicon using physical methods (please visit http://blog.sina.com.cn/bradley to read the author’s series of blog posts titled “Heroes of Physical Method Polysilicon”). Numerous polysilicon projects using metallurgical methods were announced to be in operation or under construction, with many of them claiming to have signed contracts for thousands of tons of polysilicon. Based solely on these reports, China has clearly become a major producer of physical-method polysilicon ； But just like the news from 2006, if these reports are not fake news, they are at best hype with ulterior motives. Entering 2008, physical method polysilicon in China seemed to return to a more rational approach. In sync with UMG around the world, everyone suddenly ceased all activity; there was no longer as much noise as before. At the \"Polysilicon Materials and Equipment Seminar\" held at Jiuhua Villa in Beijing at the end of April, various experts and representatives present stated that the purity of polysilicon achieved through physical purification methods at that time was only above 5N. When asked when large-scale production and application could be achieved, Professor Tan Yi from Dalian University of Technology said it should be in the \"not-too-distant future\"” ； Zheng Zhixiong, the owner of Nan’an Sanjing, said, “I have already been in large-scale production.”” ； Tong Xingxue, the president of Sevay, said from the user’s perspective that \"we won’t consider using the metallurgical method for polysilicon in the next two years\"” ； The most optimistic view is that by the end of 2008, physical vapor deposition polysilicon will achieve two goals: “it will be possible to manufacture solar cells at a scale suitable for large-scale production.” At that time, no one really believed that physical laws could soon be put into practical use. By early June, at the International Conference on Photovoltaic and Silicon Materials held in Shanghai, physical methods became a hot topic of discussion. At the \"Photovoltaic Leaders Summit\" hosted by the conference, Peng Xiaofeng, chairman of SunPower, spoke at length about his interest in metallurgical polysilicon and the research he has conducted on it. Zheng Zhixiong from Nan’an Sanjing ‘revealed his family’s secrets’ by talking at the meeting about the shortcomings of polysilicon produced by metallurgical methods. However, his famous statement with a Fujian accent shocked everyone: \"Is it possible to buy a Mercedes at the price of a Santana?\" ”This highlights the low cost of polysilicon from a contrasting perspective. But up to this point, physical method polycrystalline silicon has remained in the background, and it seems that no one can say for sure whether it can be used or not. China 2008: A tumultuous second half, but the situation changed rapidly. Two days later, Suzhou Trina Solar announced that it had produced solar cells using 100% physical-method polycrystalline silicon, which was a great encouragement for manufacturers of physical-method polycrystalline silicon in China, but it also brought considerable pressure. In September, Ningxia Yinxing Polysilicon announced that it would use polysilicon produced by metallurgical methods to manufacture batteries, and has already installed a solar power plant with a capacity of 100KVA, using polysilicon produced by physical methods. Jiangxi Sunwoda announced that it has signed a contract with the German company Q-CELL for the processing of 20,000 tons of silicon wafers. What is striking about this news is that the material used for these wafers is all UMG. Against this backdrop, at the International Solar Conference held in Changzhou in late September, physical methods became a hot topic. Professor Zhou Lang from Nanchang University presented at the conference on how several major solar companies at the Spanish photovoltaic exhibition supported the metallurgical method for producing polysilicon ； Professor Yang Deren from Zhejiang University presented information on the technical aspects of physical method polycrystalline silicon ； Shanghai ProNew Energy Co., Ltd. also presented its progress in physical method polycrystalline silicon at the conference, providing detailed data on composition as well as information on silicon wafers and solar cells, including phosphorus and boron contents, the resistivity of silicon wafers and minority carrier lifetimes, as well as the photovoltaic conversion efficiency of solar cells. At this conference, although a few experienced professionals spoke out urging everyone to avoid using physical-method polysilicon, the majority of industry insiders’ doubts regarding the metallurgical method were no longer about whether it could be used or not, but rather when it could be utilized on a large scale. More than a month after the Changzhou conference, in early November, Trina Solar in Changzhou also announced that it had successfully produced solar cells using the metallurgical method for polycrystalline silicon, achieving an efficiency of 14%. It became the second company in China, after JA Solar, to use the metallurgical method for producing polysilicon cells. Soon after, it was reported that Ates also began using domestically produced physical-method polysilicon. The price plunge at the end of the year: The price of polysilicon dropped within the month of November. It was in November that the financial tsunami was identified as transforming into a global economic crisis. The international situation is bleak. Since Christmas and New Year are approaching, orders also decrease at this time compared to previous years. But the economic crisis has caused concerns that the market will decline significantly next year. Therefore, with declining orders, everyone wants to keep cash in hand; as a result, many people are reluctant to hold inventory and rush to sell what they have, regardless of the costs. Firstly, the price of solar cells has dropped by over 40%, falling from 28 yuan per watt to 15 yuan per watt. This has also led to a drop in the price of silicon wafers; the price of single-crystal silicon wafers with a size of 125 dropped from over 50 yuan to 25 yuan per wafer. Given that the polysilicon held by many people has a complex origin and varying costs, the spot price of virgin polysilicon has dropped from 3.2 million RMB to 1.2 million RMB per ton. Recycled materials present an even greater complexity and challenge; what used to be purchased at over 2 million yuan per ton is now unsellable even at 300,000 yuan per ton. In this case, the cost advantage of physical polysilicon becomes somewhat subtle. First, the reduced demand for polysilicon has eased the supply-demand imbalance, resulting in a significant decrease in the demand for physical methods ； Second, the decline in the price of polysilicon produced by chemical methods has made the lower purity of polysilicon produced by physical methods more apparent, reducing the urgency for the use of dopants. All of this has led to a period of sluggish sales for physical polysilicon in China. At the same time, as prices fall and the marginal profits of chemical polysilicon decrease, many producers of chemical polysilicon are also wondering when the price of polysilicon will reach its bottom. Since the cost of chemical polysilicon in China is currently nearly twice as high as that abroad, if the additional production capacity generated by foreign manufacturers comes online, the price of polysilicon could drop to levels lower than the costs incurred by domestic producers. At this time, foreign manufacturers still enjoy considerable profits, while domestic manufacturers using chemical methods are in a **very poor situation. Due to its inherent cost advantages, physical polysilicon is not only a strong competitor to Siemens’ method used abroad but also ensures its own invincibility. Therefore, many new investors, including some who invest in chemical polysilicon, have also turned their attention to the physical method. On December 9, 2008, at the \"2008 International Solar-grade Polysilicon Conference\" held in Shanghai, Ding Zhongjiang, deputy general manager of Emei Semiconductor Factory, said in his speech: \"In the long term, the improved Siemens process will not become the dominant polysilicon production method in the photovoltaic industry.\" Among the various new polysilicon production processes, the metallurgical method should rank first and is the most promising. ” These words indicate that the mainstream in the polysilicon industry holds the view that physical methods are not only worth considering but also must be given due attention. One after another, polysilicon companies emerged. At one time, physical method polysilicon seemed to be the key to turning things into gold. Many people rushed in, and everywhere became a battlefield for physical method polysilicon. Entering 2008, after a period of inactivity in the first half of the year, many companies withdrew from this competitive field. The main reason is that although physical method polycrystalline silicon has many advantages, the technical challenges are not as simple as one might think; efforts made at first may yield results, but then decline and eventually lead to failure. After several attempts that yielded no results, many people in the technical field felt discouraged, and those who had invested lost confidence as well. However, there are still people who persist, and they have made good progress. These people can be roughly divided into three categories: The first category consists of those who were originally involved in metal silicon. Using his understanding of silicon and some knowledge acquired from the outside world, he began the purification process. Most of these companies have relatively crude technical approaches and lack systematic planning; since their main source of income comes from metallic silicon, the progress of polysilicon purification technology has little impact on their livelihoods ； But it is precisely for this reason that there is no strong determination or motivation, which leads to a mindset of trying something without being sure if it will work. The second category consists of many companies with immature technologies, but still possessing some funds. These companies have made little progress in purification technology this year, so they are now in a difficult situation and have no choice but to persevere with their efforts. Such companies often find it more difficult to resolve problems, as they are unable to admit to investors that their technology is not yet mature. How long these companies can survive depends on their rate of spending money. Because an investor will not rashly continue to invest before seeing the promised results. The third category consists of those companies that have a clear technical approach and have made significant progress this year. The phased products of these companies have begun to gain recognition from downstream manufacturers. Typically, such companies have a fairly complete technical team as well as sufficient funding. Given the smooth progress of technology, subsequent financing will not be a major issue either. What these companies need to do next is to stabilize the achievements already made on a large scale, and further improve the quality of their products. Among these three types of companies, it is possible for them to continue operating, but there will also be companies that withdraw from the market continuously. This financial crisis will accelerate change: those that can continue will do so even better, while those that cannot withstand it will withdraw more quickly. In short, physical vapor deposition polysilicon in 2008 was not more popular on the surface than in 2007; in fact, it even regressed to some extent. But in reality, there has been essential progress, or even a leap forward. To borrow a phrase from an article written in Jinggangshan, “It is like standing on the shore and looking out at a ship in the sea, whose mast tops can already be seen.” ； It is situated on the summit of a high mountain, from where one can see, in the distance to the east, the rising sun glowing brightly and about to emerge ； It is a baby that is about to mature, agitating within the mother’s womb. ” (End)