Current development status of polysilicon ingotting technology

Thread Content

Keynote Speech: Current Development Status of Polysilicon Ingotting Technology Zhao Bo: Good afternoon, distinguished leaders and guests! I am very grateful to the organizers for giving me this opportunity to give a presentation here. This report mainly provides a brief overview of the development of polysilicon ingot casting furnace equipment. First, I’ll take a few minutes to give you a brief overview of Seiko Group. China Seiko Group is a high-tech enterprise that operates across different regions and industries; it is a private company that has for many years been ranked among the top 500 private enterprises in China. The group owns three listed companies through direct control or shareholdings. After more than 30 years of development, its assets have exceeded 10 billion, and it employs 10,000 people. There are mainly six core industries, including three listed companies. What you see here are the six industrial platforms under the Seiko Group. Many well-known iconic projects in China, such as the new terminal of the Bird’s Nest, have been constructed or involved in their construction by the Seiko Group. In the field of solar photovoltaics, the main company involved is a holding company under Seiko, named Seiko Technology Co., Ltd., which is engaged in the manufacturing of photovoltaic equipment such as ingots. Seiko Technology’s headquarters is located in Shaoxing, Zhejiang, and it has three manufacturing facilities in Shaoxing, Hangzhou, and Wuhan, all in Zhejiang. This is my rough introduction to our company at the beginning. The technological development of polycrystalline silicon ingot furnaces has evolved entirely in line with the growth of the photovoltaic industry. Initially, solar photovoltaic systems used single-crystalline silicon. As the industry developed, efforts were made to reduce costs and improve production efficiency; abroad, the technology of polycrystalline casting began to be developed in the 1970s and 1980s. This was also a result of technological advancements in this industry. The weight of the crystals used initially was 40 kilograms, but it increased to 80 kilograms and then 150 kilograms. Actual large-scale industrial use of such crystals began after their weight reached 240 kilograms. Abroad, the use of 240-kilogram crystals started around 2003 or 2004. Starting from the second half of 2007, casting technologies capable of producing materials weighing 400 kilograms emerged. Currently, in the field of solar photovoltaics, polycrystalline ingot casting technology has become a mainstream technique. Domestic polycrystalline casting technology has begun to rapidly catch up with international standards in this field. Currently, several manufacturers in China are involved in the development of polycrystalline casting equipment. Seiko Technology Company started developing the first generation of 240-kilogram casting equipment in 2006, and by 2008 it had developed the second generation of 500-kilogram polycrystalline casting equipment; the technical performance of these products has now reached international levels. In fact, it has already surpassed foreign technical levels. This is the design of Seiko Technology’s first-generation furnace, the 240. I will briefly introduce this 240 model: it has been widely used in the field of solar photovoltaics. Abroad, there is also an SM heat exchange method that relies on the movement of crucibles for production; the basic working principle of both methods is similar, but due to differences in the overall design philosophy, their operating conditions vary. I will explain the differences between the two of them later. As you can see, this is the heat field design method for JJL.240, while the design based on heat exchange is carried out through the descent of the crucible. However, due to the different states of the heat field inside, the crystals formed vary. Everyone can see the detailed distribution map of the isotherms within silicon. The actual initial nucleation begins at the corners at the bottom. Considering the advantages and disadvantages of these two methods, in equipment with a capacity of 240 kilograms, one method results in a relatively smaller temperature difference; however, it has its drawbacks, as the corners and the area below them become excessively cold compared to the other method. As you can see, there is a difference in the orientation of the crystals produced by furnaces using HM devices compared to those that use other types of devices. In the HEM method, the crystal growth direction at the bottom is from the center toward the sides, whereas in this case it is from the corners toward the center; this is due to different control directions of the heat field. As you can see, comparing the condition of their silicon wafers in these two methods, JJL has relatively fewer defects. Based on our many years of experience in this industry, we believe there are some shortcomings in the technology related to the 240-kilogram class; in our view, we have designed equipment in the 500-kilogram class as a solution. After I introduced the two design concepts for the 240-kilogram class of this device just now, we designed the 500-kilogram class in this manner. As you can see, it shows the heat field pattern we designed for the 500-kilogram class; next to it is the state of nucleation at the bottom, which represents the state when it has basically reached full development. As you can see, in the nucleation state, the method of moving the base plate, for which we have filed a patent, is employed; this allows for good control over nucleation at the bottom, with the overall curvature being kept very low. The equilateral lines from the bottom to the top are essentially in a balanced state. Such a design allows for more uniform speeds and enables good control over the formation of microcrystalline structures. Let me show you another image: it shows the heat field distribution of a 400-kilogram product manufactured by a company abroad. Even with the bottom parts aligned in this manner, it is not possible to effectively control the problem of undercooling at the edges; this leads to the formation of microcrystals in those areas, as well as a higher concentration of impurities in the central part of the bottom. This is entirely due to design issues. As you can see, this shows the detailed isothermal distribution of JJL500 during nucleation at the bottom; the distribution of the isotherms is quite steady. The advantage of this design is, first of all, that it facilitates the distribution of silicon ingots; since the ingots can be made taller, the overall product efficiency for users increases significantly. This graph shows the precision of the JJL500; as you can see, its precision is significantly higher than that of the previous devices. This precision can achieve a height of over 250 millimeters; at present, the height achievable with foreign 400-kilogram-class systems is roughly around 260 millimeters. These are the advantages of the JJL500 model of equipment: its overall production efficiency is very high, and its height can exceed 300 millimeters; as a result, the overall efficiency also **improves**. It’s not right to compare ourselves to others; we can only compare ourselves to who we were before. With the introduction of this generation of machines, the overall production efficiency should increase by more than three times compared to that of the previous generation of equipment. Thank you all!