(Keynote) Development History of High Efficiency Silicon Heterojunction Solar Cell- from Discover to Practical Use

Abstract

Silicon heterojunction (SHJ) solar cells are attracting attention as high-efficiency Si solar cells. The basicSHJ structure is p-type amorphous Si (a-Si)/i-type (non-doped) a-Si/n-type mono-crystalline Si (c-Si). Wehave developed and actively evolved this SHJ solar cells from early 1990s, and introduced the moduleequipped with SHJ solar cells named as well-known “HIT”. The features of HIT are: (1) high efficiency,(2) good temperature characteristics, that is, a small output decrease even in the temperature environmentactually used, (3) easy application to double-sided power generation (bifacial module) using symmetricstructure. Although bifacial module has been receiving much attention in recent years, we launched it asearly as 2000.How was this excellent solar cell born? Originally, it was born in the process of developing thin-filmpolycrystalline Si (poly-Si) solar cells as the bottom cells of tandem-type solar cells with a-Si top cells.As an initial stage of the development, we used p-type a-Si as a window layer and n-type c-Si wafer as aphotovoltaic layer because the characteristics of c-Si was clearer than thin film poly-Si. The insertion of ilayerbetween p-type a-Si and n-type c-Si came from two ideas to improve the interface properties. Onewas a structural approach (control the Si bonding network between c-Si and a-Si with microcrystal bufferlayer), and the other was an attempt to suppress the mutual diffusion of impurities. For the former, it wasconfirmed that it was better to use a-Si instead of microcrystals to passivate the c-Si surface effectively,and for the latter, it was found that better characteristics could be obtained by surface passivation of ilayerwith few defects, instead of the effect of impurity separation. The surface passivation technologyusing amorphous semiconductor film was a new invention different from conventional passivation usinginsulating films such as SiOx and SiNx. With this new structure, a large improvement of about 30 mV inopen circuit voltage (Voc) was confirmed. Next we applied this approach to the back side, and again wefound that i-layer could greatly reduce interface defects. Thus, the basic structure of the symmetrical SHJsolar cell was determined. With this structure (Ag electrode/TCO/(p/i)-a-Si/n-c-Si/(i/n)-a-Si/TCO/Agelectrode), we achieved a conversion efficiency of 20% (cell size of 1 cm x 1 cm) in 1994, and after thatthe efforts for mass production started. To expand to a practical size, we newly developed the electrodeforming technology using low-temperature curing type silver paste, and successfully started the massproduction in 1997.In parallel, our R&D team continued to make various conversion efficiency improvements whichincluded optimizing a-Si deposition conditions, developing new TCO materials and deposition methods,and lowering the resistivity of silver paste. In addition, we succeeded in demonstrating in experiments forthe first time that the open-circuit voltage increased associated with the thinning of wafers that had beenpredicted in simulations. The excellent surface passivation capability of i-layer minimized losses whenthinning wafers and maintained the conversion efficiency, which was a very advantageous feature fromthe viewpoint of cost reduction. At the end of 2012 (announced in 2013), we achieved a cell conversionefficiency of 24.7% with a wafer thickness of 98 μm, demonstrating that both low cost and highefficiency can be obtained. Some of these technologies were introduced into mass production.In 2014, we applied SHJ technology the back-contact type solar cell, and achieved a conversionefficiency of 25.6% with both the high Isc due to elimination of front electrode and the high Voc due toexcellent surface passivation. This was a new world record at that time for non-concentrating siliconbasedsolar cells at the research level, the previous one having stood for 15 years.