At present, crystalline silicon solar cells are the most mature and widely used solar cells, with a proportion of more than 90% in the photovoltaic market, and will dominate in the future for a long time [1-2]. Among them, the crystal structure of single crystal silicon is perfect, the forbidden band width is only 1.12 eV, and the raw materials in nature are rich, especially the N type single crystal silicon has less impurities, high purity, low lifetime of few children, no grain boundary dislocation defects and The advantages of easy control of resistivity are ideal materials for achieving high efficiency solar cells [1-2].

 

How to improve conversion efficiency is the core issue of solar cell research. In 1954, Bell Labs of the United States first prepared a single-crystal silicon solar cell with an efficiency of 6% [3]. Since then, research institutions around the world have begun to explore new materials, technologies and device structures. In 1999, the University of New South Wales in Australia announced that the conversion efficiency of monocrystalline silicon solar cells reached 24.7% [4], and after the solar spectrum correction in 2009 reached 25% [5], becoming a milestone in the research of monocrystalline silicon solar cells. The University of New South Wales' 25% conversion efficiency record has been maintained for fifteen years. Until 2014, Panasonic Japan and SunPower USA reported the efficiency of 25.6% [6] and 25.2% [7]. . Since then, Japan's Kaneka company [9,14-15], the German Fraunhofer Research Center [10-11], the German Hamelin Solar Energy Research Institute [12-13] have successively reported single-crystal silicon solar cells with an efficiency of more than 25%. The parameters are shown in Table 1.

 

1 theoretical efficiency of single crystal silicon solar cells

 

For homojunction single crystal silicon solar cells, in 2004, Shockley and Queisser theoretically calculated single crystal silicon solar cells with a limit efficiency of 33%, also known as Shockley-Queisser (SQ) efficiency [16], but the efficiency is only Radiation recombination is considered, and non-radiative recombination and intrinsic absorption losses (such as Auger recombination and parasitic absorption) are ignored [17]. In 2013, Richter et al. proposed a novel and accurate method for calculating the ultimate efficiency of single crystal silicon solar cells, considering the new standard solar spectrum, silicon optical properties, free carrier absorption parameters, and carrier recombination and banding. When the thickness of the silicon wafer is 110 μm, the theoretical efficiency of the single crystal silicon solar cell is 29.43% [17]. The simulation of silicon heterojunction (SHJ) solar cells indicates that the optimal back-field structure can simultaneously improve its Voc and Jsc, and the significance of the thickness of the silicon wafer for battery performance. The theoretical limit efficiency of the symmetric structure SHJ battery is 27.02%. [18]. In 2013, Wen et al. analyzed that the interface state defect, band gap compensation and the work function of transparent conductive oxide (TCO) all affect the interfacial transmission performance of a-Si:H(p)/n-CzSi, and thus simulate The theoretical limit efficiency of 27.37% is [19]. In 2015, Liu Jian et al. further proposed that the appropriate a-Si:H thickness, doping concentration and back-field structure can improve the carrier transfer performance of a-Si:H/c-Si heterojunction solar cells, simulation The theoretical limit efficiency is 27.07% [20]. The above studies all believe that the optimal back field can improve the carrier transport, reduce the loss of carriers in the PN junction, and point out that carrier mobility is an important condition to improve the conversion efficiency of SHJ cells [18- 20].

 

For the new undoped silicon heterojunction cell, in 2014, Islam et al. used metal oxide as a novel carrier selective passivation contact layer, which reduced the loss of carriers in the “PN junction” and improved The voltage drop loss in contact with the metal, the ultimate efficiency of the simulation calculation reached 27.98% [21]. Table 2 summarizes the theoretical ultimate efficiencies of monocrystalline silicon solar cells under ideal conditions.

 

2 Analysis of structure and characteristics of high efficiency single crystal silicon solar cells

 

Martin Green analyzed the causes of battery efficiency losses, including the five possible pathways shown in Figure 1 [1, 22]: (1) photons with energy less than the band gap of the cell's absorber layer cannot excite electron-hole pairs. Penetrate directly.

 

(2) Photons with energy greater than the forbidden band width of the absorption layer of the battery are absorbed, and the generated electron-hole pairs are excited to the high energy state of the conduction band and the valence band, respectively, and the excess energy is released in the form of phonons, and the electrons of the high energy state are - The holes fall back to the bottom of the conduction band and the top of the valence band, resulting in loss of energy. (3) Charge separation and transport of photogenerated carriers, loss in the PN junction. (4) A voltage drop loss is caused at the contact of the semiconductor material with the metal electrode. (5) Composite loss due to material defects during photocarrier transport.

 

The various ways of energy loss above can be summarized as optical losses (including (1), (2), and (3)) and electrical losses (including (3), (4), and (5)). In order to improve solar cell efficiency, it is necessary to simultaneously reduce optical loss and electrical loss. Effective measures to reduce optical loss include low-refractive index anti-reflection film on the front surface, front surface suede structure, back reflection and other trapping structures and techniques, while full back contact technology without metal electrode shielding on the front surface can maximize Improve the utilization of incident light. To reduce electrical losses, we need to improve the quality of silicon wafers, improve PN junction formation technology (such as ion implantation, etc.), new passivation materials and technologies (such as TOPCon, POLO, etc.), metal contact technology, etc. [1]. A single crystal silicon solar cell with various structures has been proposed for how to reduce optical loss and electrical loss. Currently, single crystal silicon solar cells with conversion efficiency exceeding 25% mainly include the following six types.

 

2.1 passivated emitter back field point contact (PERC) battery family

 

The team led by Martin Green of the University of New South Wales (UNSW) proposed a single crystal silicon solar cell with PERC structure, achieving a high conversion efficiency of 22.8% on P-type FZ silicon wafers [23]. The basic structure is shown in Figure 2a. . In 1999, the UNSW team again announced that its PERL solar cell (shown in Figure 2b) had a conversion efficiency of 24.7% [4-5]. Compared with traditional monocrystalline silicon solar cells, the main features and advantages of PERL solar cells include: (1) Silicon oxide as a passivation layer on the back surface of PERL solar cells, the recombination rate of the interface is significantly reduced. (2) The back metal electrode contacts the heavily doped emitter through the small hole, and this structure can form a good ohmic contact, thereby reducing the resistance loss [4]. (3) The inverted pyramid trapping structure provides better trapping effect, and the MgF2/ZnS as the double anti-reflection layer reduces the reflection of light, which together significantly improves the short-circuit current of the solar cell [23]. In order to solve the problem of equivalent series resistance increase caused by insufficient back contact, they used the light boron doping on the back of the whole silicon wafer, and then used the localized heavy boron doping to prepare the metal contact region, thereby forming a PERT battery. The structure is shown in Figure 2c. It achieves high conductance and low back surface recombination rates, improves open circuit voltage and fill factor, and achieves a high efficiency of 24.5% on a 4cm2 P-type MCZ silicon wafer [25]. The structure of the PERC solar cell is shown in Figure 2a. It has the advantages of excellent back surface passivation and its preparation technology. In recent years, it has received extensive attention from the industry and has become the next generation of high-efficiency high-end battery products in the industry.

 

FraunhoferISE uses a technology that is non-lithographic, fast processing and suitable for various silicon substrates. The PERC battery has a efficiency of more than 21% and has a good industrialization prospect [27]. In 2017, Longji Leye and Jingke reported respectively a single crystal silicon PERC battery with an efficiency of 23.26% [28] and 23.45% [29]. In 2018, they reported on the efficiency of 23.6% and 23.95% of the battery [30], becoming a milestone in the photovoltaic industry. In the preparation process of PERC cells, the design of the back electrode and the formation of good ohmic contact between the metal electrode and the silicon substrate are two key steps [1-2]. At present, the ohmic contact technology between the metal electrode and the silicon substrate is more and more mature, and has been widely used in the production line.

 

2.2 Interdigitated back contact (IBC) solar cells

 

In 1975, Schwartz first proposed a back contact solar cell [31]. After years of development, people have developed interdigitated back contact (IBC) solar cells, the structure of which is shown in Figure 3. The most striking feature of the IBC solar cell is that the PN junction and the metal contact are on the back of the solar cell. The front surface completely avoids the obstruction of the metal grid electrode, and combines the pyramidal surface structure of the front surface and the light trapping structure composed of the anti-reflection layer. It can maximize the use of incident light, reduce optical loss, and have a higher short-circuit current. At the same time, the optimized metal gate electrode on the back reduces the series resistance [32]. Usually, the SiNx/SiOx double-layer film is used on the front surface, which not only has an anti-reflection effect, but also has a good passivation effect on the surface of the suede silicon. The front unshielded solar cell not only has high conversion efficiency, but also has the advantages of beautiful appearance, and is suitable for application in photovoltaic building integration, and has great commercial prospects. At present, IBC battery is the most complicated and structurally difficult battery in the commercial crystalline silicon battery, marking the highest level of R&D and manufacturing of crystalline silicon.

 

SunPower, Inc., a leading industrial leader in IBC batteries, has developed three generations of IBC solar cells. Among them, the highest efficiency of the third-generation IBC solar cells prepared on N-type CZ silicon wafers in 2014 reached 25.2% [33]. Trina Solar has been committed to the research and development of IBC monocrystalline silicon cells. The large-area 6-inch (243.2cm2) N-type monocrystalline silicon IBC battery independently developed in May 2017 has an efficiency of 24.13% [30]; 2018 In February of this year, the efficiency of the battery was further increased to 25.04%, the open circuit voltage reached 715.6mV, and it was independently tested and certified by the Japan Electric Safety and Environmental Technology Laboratory (JET). This is the single-junction single-crystal silicon solar cell with a local efficiency of more than 25% for the first time in China. It is also the highest conversion efficiency of monocrystalline silicon solar cells prepared on a large 6-inch crystalline silicon substrate in the world. It marks an important step in the research of high-end photovoltaic cell technology [30].

 

2.3 silicon heterojunction (SHJ) solar cell

 

Although PERL batteries and IBC batteries can achieve extremely high efficiency, they are all based on homogenous PN junctions [34]. The theoretical calculation of AFORS-HET shows that the heterojunction is beneficial to the solar cell to obtain a higher open circuit voltage, thus obtaining higher battery efficiency [17]. Due to the difference of the forbidden band width, conductivity type, dielectric constant, refractive index and absorption coefficient of the two semiconductor materials in the heterojunction, it is more widely used than the homojunction [1]. Since the 1980s, Sanyo Corporation of Japan and the subsequent Panasonic Corporation have been in the leading position in the field of monocrystalline silicon heterojunction solar cells (HIT, also known as SHJ), after the intrinsic a-Si:H passivation layer, back Continuous optimization and adjustment of key technologies such as field structure, high conductivity and high transmission ITO, trap structure, metallization gate line and wafer thickness [35-37], 2013 will be 101.8cm2 SHJ solar cell efficiency Increased to 24.7% [38], the open circuit voltage (Voc) reached 750mV, much higher than the open circuit voltage of the homojunction battery, the basic structure is shown in Figure 4.

 

The rapid development of silicon heterojunction (SHJ) solar cell research is closely related to its own advantages. Its advantages are as follows [40-43]: low temperature preparation process, heterogeneous high Voc, double-sided fleece structure Surface lighting, full passivation layer contact structure, no photolithographic opening, one-dimensional transport of carriers, and low cost and high efficiency. Kaneka Corporation of Japan is committed to the study of monocrystalline silicon heterojunction solar cells. They use double-sided woven silicon wafers with intrinsic a-Si:H as a passivation layer to achieve high open circuit voltage, which is also high. An important reason for efficiency. The silicon wafer adopts double-sided texturing technology to reduce optical loss. It has TCO on both sides and has dual functions of optical transparency and conductivity. In addition, they also electroplated Cu on the Ag electrode, reducing cost and improving conductivity [44], thereby further optimizing the performance of the SHJ solar cell with an efficiency of 25.1% [9].

 

In recent years, China has made great progress on SHJ batteries. The SHJ battery conversion efficiency reported by Hangzhou Cylon reached 23.1% (effective area 229.9cm2) [45]. The Shanghai Institute of Microsystems and Information Technology of the Chinese Academy of Sciences has achieved a 22.5% efficiency in the 125mm × 125mm N-CZ silicon wafers since 2015 [46], by improving the quality of the wafer and the suede trap structure, 2017 In February, the efficiency of SHJ cells prepared on large-area (156 mm × 156 mm) N-CZ silicon wafers reached 23.5% [47]. In terms of industrialization, many companies at home and abroad have gradually promoted the development of their industrialization chain.

 

2.4 Interdigitated back contact heterojunction (HBC) solar cell

 

In order to further improve the conversion efficiency of single crystal silicon solar cells, the advantages of high short circuit current of IBC battery and high open circuit voltage of SHJ battery can be combined into interdigitated back contact heterojunction (HBC) solar cells. Show. Compared with IBC structured solar cells, HBC solar cells use a-Si:H as a double-sided passivation layer, which has excellent passivation effect and can achieve higher open circuit voltage [6]. In the process of growing PN junctions, they doped with a regional mask, which reduces the carrier loss of carriers. Compared with the solar cell of SHJ structure, the front surface has no electrode shielding, and the SiN anti-reflection layer replaces TCO, and the advantage of reducing optical loss is more significant (in the short wavelength range). Combining the two advantages of the front surface, the HBC battery can Get a higher short circuit current.

 

In 2017, the battery developed by Kaneka Corporation of Japan achieved conversion efficiency of 26.3% [14] and 26.63% [15]. The front surface of the company's HBC battery (SHJ+IBC) has no metal electrodes, and the P and N layers on the back are arranged in a staggered arrangement, which greatly reduces the series resistance Rs, and the metal electrodes in contact with the P and N layers can form very Good ohmic contact increases the short circuit current. In addition, the excellent intrinsic passivation layer is capable of obtaining a high open circuit voltage. These two advantages also determine that Kaneka can achieve the highest efficiency of the world's crystalline silicon cells.

 

2.5 tunneling oxide passivation contact (TOPCon) solar cell

 

The Fraunhofer Research Center in Germany uses a chemical method to prepare a layer of ultra-thin silicon oxide (~1.5nm) on the back side of the cell, and then deposits a layer of doped polysilicon, which together form a passivated contact structure. This technique is called tunneling. Wear oxide layer passivation contact (TOPCon) [10] technology. Due to the difference in work function between n+ polysilicon and the absorber layer, the former will