Strategy for
 large‐scale
 monolithic
 Perovskite
 /Silicon tandem solar cell: A review of recent progress

Funding information KOREA East-West Power Co., LTD. (EWP), Grant/Award Number: 2.190433.01; Korea Institute of Energy Technology Evaluation and Planning, Grant/Award Numbers: 20163010012450, 20193091010460; National Research Foundation of Korea, Grant/Award Number: 2019M1A2A2072416 Abstract For any solar cell technology to reach the final mass-production/commercialization stage, it must meet all technological, economic, and social criteria such as high efficiency, large-area scalability, long-term stability, price competitiveness, and environmental friendliness of constituent materials. Until now, various solar cell technologies have been proposed and investigated, but only crystalline silicon, CdTe, and CIGS technologies have overcome the threshold of mass-production/commercialization. Recently, a perovskite/silicon (PVK/Si) tandem solar cell technology with high efficiency of 29.1% has been reported, which exceeds the theoretical limit of single-junction solar cells as well as the efficiency of stand-alone silicon or perovskite solar cells. The International Technology Roadmap for Photovoltaics (ITRPV) predicts that silicon-based tandem solar cells will account for about 5% market share in 2029 and among various candidates, the combination of silicon and perovskite is the most likely scenario. Here, we classify and review the PVK/Si tandem solar cell technology in terms of homoand hetero-junction silicon solar cells, the doping type of the bottom silicon cell, and the corresponding so-called normal and inverted structure of the top perovskite cell, along with mechanical and monolithic tandemization schemes. In particular, we review and discuss the recent advances in manufacturing top perovskite cells using solution and vacuum deposition technology for large-area scalability and specific issues of recombination layers and top transparent electrodes for large-area PVK/Si tandem solar cells, which are indispensable for the final commercialization of tandem solar cells.

possibility of a large area, 4 but also the socio-economic criteria such as price competitiveness, 5 human body, and environmental friendliness 6 of constituent materials. A wide variety of solar cell concepts have been proposed, and intensive research has been conducted to satisfy the above-mentioned criteria till now, but most of these technologies have not exceeded the threshold of final mass production commercialization.
Currently, the cost of crystalline silicon solar cells represented by Al-BSF or PERC has already fallen below 10 cents per watt and is predicted to decrease continuously to 7 cents by reducing the amount of polysilicon used, as the wafer thickness decreases or the increase in power generation per cell as the wafer size increases from M2 to M12. 7 Therefore, there is no doubt that crystalline silicon solar cells will maintain the best price competitiveness among all solar cell technologies in the future, but the efficiency is already approaching the practical limiting efficiency of single-junction solar cells, requiring an innovative strategy. 8 A multi-junction or tandem solar cell manufactured by stacking a semiconductor PN junction with a large bandgap on top of another semiconductor PN junction with a small bandgap is one of the alternatives that can overcome the fundamental efficiency limitations of a silicon solar cell. 9 The III to V compound semiconductor-based triple tandem solar cell was recognized as one of the ultra-high-efficiency solar cell candidates with high efficiency of 37.9%. 10 However, III to V compound semiconductor technology has failed to commercialize mass production because it requires expensive vacuum equipment and precursors for thin-film synthesis and is known to be harmful to the human body and environment. 11,12 As an alternative, studies on implementing tandem solar cells of various structures by stacking various low-cost solar cells have been intensively studied in recent years. Top cell candidates include organic solar cells, 13 CIGS solar cells, dye-sensitized solar cells, 14 and perovskite (PVK) solar cells 15 with relatively high bandgap energy, while bottom cells include narrow bandgap materials such as silicon solar cells and CIGS. 16 Representative heterojunction tandem solar cells are as follows: Dye-sensitized solar cells/silicon tandem solar cells using PEDOT:FTS with higher transparency and lower charge transfer resistance than Pt used as an interfacial catalyst layer, 14 solution-processed organic/organic tandem solar cells guided by a semi-empirical analysis, 17 PVK/CIGS tandem solar cells with conformal monolayer contacts with lossless interfaces, 18 PVK/organic tandem solar cells with lowloss interconnecting layers, 19 a monolithic all-PVK tandem solar cell with a strategy to reduce Sn vacancies in mixed Pb-Sn narrow-bandgap PVKs 20 have been reported.
Among these various heterogeneous tandem solar cell technologies, perovskite/silicon (PVK/Si) tandem solar cells are considered to have the highest potential in terms of high efficiency and price competitiveness. The German research team reported a PVK/Si tandem solar cell with an efficiency of 29.1%, which exceeds the theoretical limit of a single junction solar cell close to a standalone silicon or PVK solar cell. 21 Comprehensive optical simulations conducted by various research institutes show that when the current matching of the monolithic PVK/Si tandem solar cell is optimized, the ideal current density value is 21 mA/cm 2 , and the corresponding efficiency is reported to be 31%. [22][23][24] In order to achieve such an ideal current density, novel device structures, and new materials that can minimize light reflection at the surface and parasitic light absorption by electron and hole transport layers need to be designed. Because silicon solar cells are known to be very stable devices, the long-term stability issue of PVK/Si tandem solar cells stems entirely from top PVK cells. 25 Recently, through intensive research on the improvement of long-term stability of PVK solar cells, it was found that forming A and X sites in a mixed composition rather than a single composition greatly improves the stability of the solar cell. 26,27 More stable PVK thin films with minimal lattice mismatch can be formed through mixed-cation and mixed-halide approaches designed in consideration of a tolerance factor. 28 In this review article, we have categorized and discussed the differences, advantages, and disadvantages of tandem solar cells depending on the type of doping or junction of bottom silicon solar cells. In particular, we review the recent advances in manufacturing top PVK cells using solution and vacuum deposition technology for large-area scalability along with specific issues of recombination layers and top transparent electrodes for large-area PVK/Si tandem solar cells, which are indispensable for the final commercialization of tandem solar cells.   Figure 1B,C). In order to maximize the efficiency of a 4T tandem solar cell, it is necessary to maximize the efficiency of each top and bottom cell at the matching point of current and voltage, respectively. The highest theoretical 4T tandem solar cell has a PCE of 46% ( Figure 1E). 15 Taking advantage of these 4T tandem solar cells, various 4T tandem solar cells are being studied. Si solar cells, which occupy the mainstream of the solar cell market, are the most suitable candidates for commercialization in tandem solar cell research. 29 In addition, PVK with a tunable bandgap is a suitable candidate as the top cell of a Si-based tandem solar cell. 30 The structure of a typical 4T tandem PVK/Si solar cell can be classified as follows. There are PIN structured PVK top cell/n-si based homojunction Si bottom cell (Figure 2A), 31 NIP structured PVK top cell/p-si based homojunction Si bottom cell ( Figure 2B), 32 PIN structured PVK top cell/n-si based heterojunction Si bottom cell ( Figure 2C) 31 and NIP structured PVK top cell/n-si based heterojunction Si bottom cell ( Figure 2D). 33 Representative 4T PVK/Si tandem solar cells based on a homojunction Si bottom cell as a function of the type of top cell structures such as NIP or PIN. Bailie et al developed the first 4T PVK/Si tandem solar cells with 17% efficiency by using a silver nanowire transparent electrode on an NIP structure PVK top cell based on CH 3 NH 3 PbI 3 and applied it to a low-quality multi-crystalline Si bottom cell. 16 Duong et al analyzed the electrical and optical power loss in detail and produced an NIP structured PVK top cell of >80% near-infrared transmittance based on the design for the optimal tandem cell with ITO transparent electrode. 34 This was applied to the PERL cell, and an efficiency of 20.1% was reported. Ren et al increased the efficiency of the NIP structured PVK top cell by using a PVK film with a low density of bandgap states applied with postdeposition oxygen treatment and developed a highly transparent MoO 3 /Au/MoO 3 electrode. 35 This was applied on a p-Si-based homojunction Si bottom cell to fabricate a 4T tandem solar cell and reported efficiency of 23.6%. Ren et al obtained high transmittance in the longwavelength range using a MoO 3 /Au/MoO 3 transparent electrode. 36 40 In addition, the 2D PVK precursor as MABr was introduced on the 3D PVK film, and this result showed that the device performance was improved by reducing the surface defects of the 3D PVK film. This top cell was placed on the PERL cell to make a 4T PVK/Si tandem solar cell, and an efficiency of 26.2% (1 cm 2 ) was reported.
As representative 4T PVK/Si tandem solar cells of NIP structured PVK top cell or PIN structured PVK top cell based on heterojunction Si bottom cell, for the first time, Löper et al obtained a transmittance of >55% in the near-infrared spectral region by using an ITO transparent electrode design without a metallic component on a PVK top cell with an NIP structure based on CH 3 NH 3 PbI 3 . 41 This technology was applied to a heterojunction Si bottom cell to make a 4T PVK/Si tandem solar cell, and due to high transmittance, the current density value of the bottom cell was increased, and an efficiency of 13.4% was reported. Werner et al used a MoO x buffer layer to protect sputtering damage and used an indium zinc oxide (IZO) layer having an absorption of less than 3% at 400 to 1200 nm and sheet resistance of 35 Ω/sq as a transparent electrode. 42 The CH 3 NH 3 PbI 3 -based PVK top cell of the NIP structure using this transparent electrode obtained >60% transmittance in the 800 to 1200 nm wavelength range. This top cell was applied on a heterojunction Si bottom cell to increase the current density to make a 4T PVK/Si tandem solar cell with an efficiency of 18.18%. Werner et al fabricated a CH 3 NH 3 PbI 3 -based PVK top cell with an NIP structure that increased the area from 0.2 to 1 cm 2 . 43 This was applied to a heterojunction Si solar cell on top to make a 4T PVK/Si tandem solar cell and reported efficiencies of 23.0% (1 cm 2 ) and 25.2%  46 Based on this design, a 4T PVK/Si tandem solar cell was fabricated by applying a PVK top cell based on the NIP structure of Cs 0.1 FA 0.9 PbI 2.865 Br 0.135 on the IBC bottom cell to produce efficiencies of 25.5% (0.13 cm 2 ) and 23.9% (4 cm 2 ). Wang et al introduced an ultrathin gold nanomesh layer using the Frank-van der Merwe growth method to fabricate a MoO x /Au/MoO x transparent electrode. 33 An NIP structured CH 3 NH 3 PbI 3 based PVK top cell was introduced for this electrode, and a 4T PVK/Si tandem solar cell was fabricated by applying the top cell on the heterojunction Si bottom cell, and an efficiency of 27% was reported. Jaysankar et al introduced an Al 2 O 3 passivation layer using an ALD process between the PVK layer based on Cs 0.15 (CH 5 NH 2 ) 0.85 Pb(I 0.71 Br 0.29 ) 3 and the spiro layer used as hole transport layer (HTL). 47 48 In addition, by introducing 2D/3D PVK hetero-structure passivation to the top cell, V oc of about 45 mV was increased. This top cell was applied to the IBC bottom cell to make a 4T PVK/Si tandem solar cell and reported an efficiency of 25.7%. Rohatgi et al pointed out the existing high-cost heterojunction Si sub-cell and fabricated a tunnel oxide passivated contact (TOPCon) sub-cell, which has higher commercialization potential than the heterojunction Si cell. 49 A Cs 0.05 FA 0.8 MA 0.15 PbI 2.55 Br 0.45 based PVK top cell of PIN structure was placed on top of the TOPCon bottom cell to make a 4T PVK/Si tandem solar cell and reported an efficiency of 26.7%. Dewi et al compared the previously reported filtered-based measurement method with a size matching scheme using a mask and mentioned the importance of optimal measurement schemes. 50 In the case of 4T PVK/Si tandem solar cells measured by this analysis method the efficiencies, 24.7% and 23.5%, respectively, showed a difference of approximately 1%.
The analysis of the flow of 4T PVK/Si tandem solar cell research that has been carried out so far ( Figure 3 and Table S1) shows that it has the advantage of simply applying it to the previously optimized Si solar cell by making a PVK top cell with excellent long-wavelength transmittance. However, there is an air gap in the path where the light source passes through the PVK top cell and is transmitted to the Si bottom cell. This air gap causes a difference in refractive index between the PVK top cell and the Si bottom cell. In other words, optical loss due to the difference in refractive index is a major cause of low efficiency. Therefore, in order to overcome the limitation of the efficiency of the 4T tandem solar cell, it is necessary to minimize the optical loss by reducing the parasitic absorption of each functional layer. In addition, there is a need for an interfacial layer technology capable of minimizing optical loss that can replace the air gap between the PVK top cell and the Si bottom cell.

| Monolithic 2-terminal PVK/Si tandem solar cell
The monolithic 2-terminal (2T) tandem solar cell has the advantage of less parasitic absorption, as it is a simple integrated type without additional glass substrate and thick transparent electrode for PVK top cell, but sophisticated technologies such as process optimization and current matching technology are required ( Figure 1A). 15 Additionally, the cost of the glass substrate and thick transparent electrode of 4T tandem solar cells is also a drawback in commercialization compared to 2T tandem solar cells. 51 In order to maximize the efficiency of a 2T tandem solar cell, it is necessary to maximize the efficiency of each top and bottom cell at the matching point of current and voltage. The highest theoretical 2T tandem solar cell has a PCE of 45.7% ( Figure 1D). 15 Among 2T tandem solar cells, the reason why PVK/Si tandem solar cell research is growing rapidly is that PVK thin films are easier to adjust the bandgap and thickness than other thin films. 52 That is, when a PVK thin film is applied as the top cell of a tandem device by using the specificity of such a PVK thin film, it is advantageous for the current matching of a tandem solar cell. To fabricate a 2T PVK/Si tandem solar cell, the polarity of the PVK top cell and Si bottom cell should be matched, and a semi-transparent PVK solar cell with high transparency in the nearinfrared region should be fabricated directly on the Si bottom cell. 28 A typical 2T PVK/Si tandem solar cell design with a PVK top cell based on a Si solar cell is as follows. There are NIP structured PVK top cell/n-Si based homojunction Si bottom cell ( Figure 2E), 53 PIN structured PVK top cell/p-Si based homojunction Si bottom cell ( Figure 2F), 54 NIP structured PVK top cell/n-Si based heterojunction Si bottom cell ( Figure 2G) 55 and PIN structured PVK top cell/n-Si based heterojunction Si bottom cell ( Figure 2H). 56 As the advantages of Si and PVK solar cells, the following 2T PVK/Si tandem solar cells are rapidly being studied.
Representative 2T PVK/Si tandem solar cells based on an NIP structured PVK top cell and n-type homojunction Si bottom cell are described in this section. Mailoa et al 57 developed the first monolithic 2T PVK/Si tandem solar cell using an NIP structured CH 3 NH 3 PbI 3 -based PVK top cell and an n-Si homojunction Si lower cell, resulting in an efficiency of 13.7%. Werner et al introduced homojunction Si capable of high-temperature heat treatment of TiO 2 , which was used in the high-efficiency NIP PVK solar cell process. 58 In addition, by introducing a zinc tin oxide layer as a recombination layer, it was shown that the device was driven even after a 500 C heat treatment process. A 2T PVK/Si tandem solar cell was fabricated with this top cell and a bottom cell, and an efficiency of 16% was reported. Wu et al mentioned the merits of Si homojunction solar cells made by hightemperature processes, which are the mainstream in the existing solar cell market, and developed an n-Si-based homojunction PERL cells that maximize the longwavelength absorption by controlling the refractive index of SiN x . 59  , which is used as an ETL in an NIP PVK top cell capable of low-temperature processing, can be introduced into a cell of n-Si homojunction PERL structure without a recombination layer. 60 Zheng et al applied SnO 2 as an ETL in an NIP CH 3 NH 3 PbI 3 -based PVK top cell capable of lowtemperature processing. This top cell was introduced on a cell with an n-Si homojunction PERL structure without an ITO layer, and the current matching point was determined by adjusting the thickness of CH 3 NH 3 PbI 3 . In addition, the V oc was also increased by heavy doping of the emitter of the Si bottom cell. The efficiencies of 2T PVK/Si tandem solar cells using these technologies were reported to be 20.5% at 4 cm 2 and 17.1% at 16 cm 2 . Zheng The efficiency chart of reported 4-terminal PVK/Si tandem solar cell et al introduced (FAPbI 3 ) 0.83 (MAPbBr 3 ) 0.17 and applied it to an n-Si-based homojunction Si solar cell with back side-texturing and reported efficiency of 21.8% at 16 cm 2 by introducing a new metal grid design. 61 62 An efficiency of 24.1% was reported by developing a 2T PVK/Si tandem solar cell that minimized optical loss by introducing a TiO 2 layer using ALD in an n-Si-based homojunction Si lower cell. Zhu et al optimized high-quality PVK films by adjusting the ratio of N,N-dimetylformamide to dimethyl sulfoxide. A 2T PVK/Si tandem solar cell was fabricated with an NIP FAMACs PVK top cell with a high-quality PVK film and an n-Si heterojunction bottom cell. 63 As a result, the efficiency of the tandem solar cell was 22.80%. Zheng et al developed an ARC film that reduced optical loss and improved UV stability by applying a down-shifting material. 53 The 2T PVK/Si tandem solar cell was fabricated with an NIP structured PVK top cell and an n-Si homojunction bottom cell and reported an efficiency of 23.1%.
Representative 2T PVK/Si tandem solar cells based on an PIN structured PVK top cell and p-type homojunction Si bottom cell can be described as follows. Kanda et al deposited a transparent electrode on the opposite ends of the CH 3 NH 3 PbI 3 -based PVK top cell and the p-Si homojunction Si bottom cell, and made a 2T PVK/Si tandem solar cell by mechanically contacting it to face each other and reported efficiency of 13.7%. 64 Hoye et al pointed out the possibility of commercialization of n-type Si solar cells used in tandem solar cells based on Si solar cells accounts for only 5% of the global solar cell market. 65 Using an PIN structured CH 3 NH 3 PbI 3 -based PVK top cell, a 2T PVK/Si tandem solar cell was fabricated using a p-type Si-based bottom cell, and an efficiency of 16.2% was reported. Kanda et al reported efficiency of 15.5% by making a 2T PVK/Si tandem solar cell by mechanically contacting a CH 3 NH 3 PbI 3 -based PVK top cell with a p-Si-based homojunction Si bottom cell that has increased efficiency through process optimization. 66 Based on the p-Si-based Al-BSF homojunction bottom cell, a mainstream of the solar cell market, a (FAPbI 3 ) 0.8 (MAPbBr 3 ) 0.2 based PVK solar cell with an PIN structure, which has an HTL with an optimal band alignment, was tuned. As a result, the efficiency of the 2T PVK/Si tandem solar cell was 21.19%. Kanda et al introduced the texture of the bottom Si solar cell by adjusting the thickness of the PVK thin film of an NIP structured CH 3 NH 3 PbI 3 -based PVK solar cell and reported efficiency of 15.9% by mechanically contacting it. 32 Choi et al developed a transparent conductive adhesive to mechanically bond the existing high-efficiency NIP structured PVK top cell and a p-Si-based Al-BSF homojunction bottom cell, which occupies the mainstream of the Si solar cell market. 67 In addition, a 2T PVK/Si tandem solar cell was fabricated based on an optical design that optimized the current density of a tandem solar cell in consideration of the refractive index of each layer, and an efficiency of 19.40% was reported.
Representative 2T PVK/Si tandem solar cells based on an NIP structured PVK top cell and an n-type heterojunction Si bottom cell are as follows. Werner et al fabricated an NIP structured CH 3 NH 3 PbI 3 -based PVK top cell that minimized the light absorption by controlling the IZO thickness used as the recombination layer and the spiro-OMeTAD thickness of the HTL. 68 The 2T PVK/Si tandem solar cell made of this top cell on an n-Si heterojunction Si bottom cell reported efficiencies of 21.2% (0.17 cm 2 ) and 19.2% (1.22 cm 2 ). Albercht et al introduced SnO 2 ETL for low-temperature processing because the TiO 2 ETL layer used in an NIP structured PVK solar cell requires hightemperature heat treatment. 69 In this case, the passivation quality of the high-efficiency n-Si heterojunction Si bottom cell is degraded. This NIP structured CH 3 NH 3 PbI 3 -based PVK top cell was applied to an n-Si heterojunction Si bottom cell, and an efficiency of 18% was reported. Werner et al introduced an n-Si heterojunction bottom cell with a rear texture introduced to increase long-wavelength absorption and applied this to an NIP structured CH 3 NH 3 PbI 3 -based PVK top cell to fabricate a 2T PVK/Si tandem solar cell and reported an efficiency of 20.5%. 43 Bush et al introduced an PIN structured PVK solar cell with a wide bandgap using cesium formamidinium lead halide PVK and introduced a SnO 2 based buffer layer through ALD with minimized parasitic absorption, and showed excellent long-term stability. 70 78 Hou et al fabricated an NIP structured PVK top cell using Cs to reduce the roughness of the FAMA-based PVK surface. 79 In addition, the V oc of the n-Si heterojunction Si bottom cell was increased by optimizing the minority carrier lifetime by controlling the silane dilution ratio vs hydrogen. These cells were tandemized, and an efficiency of 20.43% was reported. Kim et al designed an optimized monolithic 2T PVK/Si tandem solar cell by considering each functional layer based on optical simulation. 54 Hou et al optimized the pyramid size of the pyramidally textured PDMS ARC film. 80 An efficiency of 21.93% was reported by applying an optimized ARC film by making a 2T PVK/Si tandem solar cell with an NIP structured PVK top cell and an n-Si heterojunction bottom cell. Kamino et al developed a low-temperature silver grid screen printing technology to replace the high-temperature screen printing technology used in the industry. 81 This technique was applied to a 2T PVK/Si tandem solar cell in which the n-Si heterojunction bottom cell and the Cs 0.17 FA 0.83 PbI 0.83 Br 0.17 based PVK top cell with an PIN structure were tandemized, and 22.6% efficiency was reported at 57.4 cm 2 . Park et al reported efficiency of 23.5% by optimizing the current matching of tandem solar cells using a 3T EQE analysis method and applying a CH 3 NH 3 PbI 3 based PVK top cell with an PIN structure to an n-Si heterojunction bottom cell. 82 Köhnen et al found that fill factor changes due to current mismatch in the 2T PVK/Si tandem solar cells. 22 In consideration of this, the  85 In addition, 1-butanethiol was introduced to overcome the limit of charge collection of the micro-thick PVK film formed on the pyramid texture surface, which increased the diffusion length, and prevented phase segregation. An efficiency of 25.7% was reported by tandemizing the Cs 0.05 MA 0.15 FA 0.8 PbI 2.25 Br 0.75 PVK top cell with an PIN structure, which has these technologies and an n-Si heterojunction bottom cell. Xu et al introduced triple-halide alloys using chlorine, bromine, and iodine to reduce the V oc deficit caused by photo-induced phase segregation of wide band-gap PVK solar cells. 86 An efficiency of 27% was reported by tandemizing an PIN structured PVK top cell that applied this technology and an n-Si heterojunction bottom cell. Chen et al reduced the pyramid texture size and introduced a nitrogen-assisted blading process to successfully coat PVK with a thickness of 0.5 to 1 μm. 87 This technique was applied to an PIN structured PVK top cell and tandemized with an n-Si heterojunction bottom cell to report an efficiency of 26.2%. Kim et al fabricated a PVK top cell with 20.7% efficiency in 1.7 eV bandgap through anion engineering of phenethylammonium-based twodimensional additives. 88 This was applied to an PIN structured PVK top cell and applied to the n-Si heterojunction bottom cell, and an efficiency of 26.7% was reported.
The 2T tandem solar cells based on an PIN structured PVK top cell and p-type heterojunction Si bottom cell are described as follows. A double-sided textured p-Si heterojunction bottom cell was fabricated and tandemized with a PVK top cell with an PIN structure using a hybrid vacuum deposition process to report an efficiency of 25.1%. 83 Recently, research on 2T PVK/Si tandem solar cells has developed rapidly, and many results have been reported in a short period of time. When the general tandem solar cell is divided into large and small areas based on 1 cm 2 ( Figure 4A,B and Table S2), the largest area of reported PVK/Si tandem solar cell is 57.4 cm 2 . 81 This showcases the importance of a large area technology for PVK top cells because it is considerably smaller than the area of a commercial silicon solar cell. The next chapter will introduce the fabrication process and electrode design of PVKs for large areas considering optical and electrical losses.

| SOLUTION AND EVAPORATION PROCESS FOR LARGE-SCALE PVK LAYER DEPOSITION
Among the solution processes such as spin coating, spray coating, slot-die coating, and the spin coating has been widely used for the deposition of PVK layers because of the easy control of the chemical composition and thickness to optimize PVK solar cells. In particular, the efficiency of PVK solar cells fabricated through spin coating began to increase rapidly after the advent of the antisolvent dropping method, resulting in high-quality, dense, and pinhole-free PVK films. 89 However, in largescale film coatings, spin coating is expensive because the material utilization rate is too low, and the spin coater is expensive to spin heavy substrates at high rotation speeds. In addition, nonuniform and insufficient antisolvent coverage in PVK samples has made it difficult on a large scale to obtain high-quality, high-density, and pinhole-free PVK films with the anti-solvent dropping method. Unlike a spin coating, other solution processes such as spray coating, slot-die coating, and blade coating are suitable for large-scale PVK film deposition because they are easy to scale-up with continuous coating and have high material utilization ( Figure 5). [90][91][92] However, many processing factors such as precursor, solvent, coating speed, and heating temperature must be optimized to achieve the same PVK film quality by spin coating. Zuo et al introduced a blow-assisted drop-casting method for MAPbI 3 films in air and achieved a champion PEC of 15.57% through a slot-die coating method with NH 4 Cl additive to optimize the morphology of PVK films. 92 Cotella et al achieved a PCE of 9.2% by controlling the crystallization of PVK films using a slot die coating method with a preheated substrate and cold air knife. 93 Ulicna et al investigated PVK film morphology through a spraycoating process and anti-solvent dipping methods, resulting in a PCE of 17.29%. 94 Park et al developed a reproducible megasonic spray-coating method for MAPbI 3 film formation. With megasonic spray-coating, PVK films were prepared at low temperature with an anti-solvent-free process, resulting in a PCE of 14.2% in a 1 cm 2 active area. 95 Recently, a high-efficiency PVK solar module using a large-scale solution process rather than spin coating has been reported through process factor optimization and the development of a new method. Currently, 11.1% PCE of 168.75 cm 2 PVK module and 16.9% PCE of 63.7 cm 2 size by slot-die coating and blade coating have been reported, respectively. 96,97 The previous results are quite promising, but the PVK compositions are both simple MAPbI 3 , and control of PVK composition in a While the solution process has been mainly used for PVK layer coating until now, the evaporation process for PVK layer coating is also continuously being studied because the evaporation process provides highly uniform and pinhole-free films even in a large area ( Figure 6). [98][99][100] Basically, organometal halide PVKs are chemical compounds that are usually synthesized by the reaction of metal halides with organic ammonium halides. As organometal halide PVK easily decomposes at high temperature and vacuum conditions, instead of evaporating organometal halide PVK itself, the two components of organometal halide PVK evaporate to form a PVK layer. There are two common ways to evaporate two components of PVK: one is a one-step co-evaporation method, and the other is a two-step sequential evaporation method. However, the vapor pressure of organic ammonium halides under vacuum conditions is so high that unstable vacuum deposition and the absence of solvent that helps the reaction between the two components have hindered the achievement of a high-quality PVK layer during the evaporation process. At first time, the hybrid process that using evaporation process for inorganic components and solution process for organic components on inorganic components were introduced for large-scale PVK/Si tandem solar cells. 68,74,77 This hybrid process has advantages of both evaporation process and solution process, which shows uniform and pinhole-free PVK films with easy control of chemical composition. However, in-line process for mass production is not possible in hybrid process and optimization of both evaporation process and solution process is too complex. In order to overcome the limitations of hybrid process, all evaporation process for large-scale PVK layer deposition has been researched continuously. Zheng et al studied the effect of solvent annealing on grain growth in 2D PVKs prepared by single-source thermal evaporation, and through solvent annealing, the crystallinity of 2D PVKs was improved, and a PCE of 4.67% was achieved. 101 Lei et al first formed a PbI 2 layer by normal evaporation and then fabricated MAI by flash evaporation to obtain a homogeneous large-scale PVK film. All evaporated PVK devices achieved a champion PCE of 15.06% at 16 cm 2 active area. 102 By optimizing the evaporation process parameters such as deposition rate, substrate temperature, chamber pressure, etc., PCEs of more than 20% have recently been reported with co-evaporated PVK layers. 103 Moreover, 18.13% PCE of a 21 cm 2 PVK module by coevaporation method was reported. 104 However, to optimize the bandgap of PVK and increase the efficiency of PVK/Si tandem solar cells, uniformity and control of PVK composition in large-scale evaporation processes are also required. Notably, newly developed evaporation processes, such as flash evaporation and closed-space vapor transport, can be an innovative breakthrough to overcome the decomposition problem of PVK materials. 105,106 Applying a newly developed evaporation process and optimizing the parameters in the evaporation process will

| CHALLENGES AND PROGRESS IN LARGE-SCALE PVK/Si TANDEM SOLAR CELLS
Undoubtedly, a high-quality and pinhole-free PVK layer by the scalable deposition process mentioned above is necessary for efficient large-scale PVK/Si tandem solar cells. However, it is not the only obstacle to scale up PVK/Si tandem solar cells. The first step to consider in large-scale PVK/Si tandem solar cells is the recombination layer. In the 2T monolithic PVK/Si tandem solar cell, the vertical resistance of the recombination layer should be low enough to connect the bottom and top cells without resistance loss, and the TCOs were utilized as recombination layers owing to their intrinsic high transparency and low resistance. However, defects in the PVK layer or charge transport layers (CTL) are inevitable in large-scale PVK/Si tandem solar cell fabrication, and the shunting effect due to pinholes and defects is important for the failure of large-scale PVK/Si tandem solar cells. To prevent the failure of large-scale PVK/Si tandem solar cells, it is necessary to increase the lateral resistance of the recombination layer to localize the shunting effect and minimize leakage current. An important step for largescale PVK/Si tandem solar cells is the development of a high lateral resistance recombination layer (Figure 7). In PVK/Si tandem solar cells 19.1% PCE of 12.96 cm 2 was achieved using p+/n+ hydrogenated nanocrystalline silicon recombination junction deposited by plasmaenhanced chemical vapor deposition instead of TCO. 74 In addition, 17.6% PCE of 16 cm 2 PVK/Si tandem solar cell was achieved through the optimized conductivity of recombination layer formed by the combination of a doping-level control emitter and a solution-processed charge-selective metal oxide nanoparticle layer. 60 However, the lateral resistance of the transparent electrode should be low to collect the generated charge carrier efficiently to the external circuit without loss, while the lateral resistance of the recombination layer should be low in large-scale PVK/Si tandem solar cells. After the introduction of a metal-oxide-based sputtering damage buffer layer, TCOs with high transparency and conductivity were sputtered as the top transparent electrode of PVK/Si tandem solar cells, but TCO itself has limitations as the top transparent electrode that conflicts with electrical conductivity and transparency.
To overcome the limitations of TCO itself, the sheet resistance of the transparent electrode was effectively reduced by using a metal grid on top of TCO. To obtain a low resistance of a transparent electrode with a metal grid, a metal with low resistivity, a short distance between the metal fingers, and a large cross-sectional area of the metal finger are desirable. However, the lightshading loss caused by the top surface of the metal grid prevents light from entering the solar cell and limits the increase in efficiency by introducing a metal grid. Therefore, for efficient large-scale PVK/Si tandem solar cells, optimal metal grid design, and metal finger height-to-F I G U R E 6 Schematic illustration of (A) PVK coevaporation method with controlling substrate temperature. Reproduced with permission: Copyright 2020, American Chemical Society. 103  width aspect ratio (geometry) are required (Figure 8). It was reported that the PCE of a PVK/Si tandem solar cell with 16 cm 2 size increased from 17.6 to 21.8% after metal grid design optimization. 60,61 It was also reported that shading loss decreased to less than 0.8% after increasing the metal finger aspect ratio using a graded-opening shadow mask. 86

| OBSTACLES FOR COMMERCIALIZATION AND OUTLOOK
Recently, the record efficiencies of 2T PVK/Si tandem solar cells are expected to exceed 30%, and the area of large-scale PVK layer deposition process has become large enough to cover the size of commercial silicon solar cell wafers. Although the efficiency of 2T PVK/Si tandem solar cells is higher than 29%, which is the theoretical limit of Si single-junction solar cells, and the rapid development of large-scale deposition processes for PVK layers are promising, the cost of PVK/Si tandem solar cells should be considered for commercialization. According to the recently reported cost analysis for PVK solar cells and PVK/Si tandem solar cells, the process equipment and maintenance costs of the evaporation process may be low with high throughput in mass production. PVK solar cells have generally reported high material cost issues rather than processing costs in cost analysis. In particular, the cost barrier of PVK solar cells is identified by expensive organic charge transport materials such as spiro-OMeTAD, PTAA, and PC 60 BM. [107][108][109] The cost of high-performance organic charge transport materials is usually expensive owing to the complex synthesis steps and additional cost for ultra-high purity. As an alternative, the cost of inorganic charge transport materials such as NiO, CuSCN, SnO 2 , and Nb 2 O 5 is much cheaper than organic materials, and several large-scale compatible processed inorganic CTLs have been reported in PVK solar cells ( Figure 9A). [110][111][112][113] From the perspective of eco-friendly renewable energy, it is necessary to solve the toxicity and longevity of PVK/Si tandem solar cells, which are the drawbacks of PVK, to gain an advantage over other energy sources. Most PVK solar cells are based on toxic lead (Pb) due to their excellent optoelectronic properties. Although the total amount of Pb in PVK/Si tandem solar cells is very low, it has been reported that the Pb of halide PVK is 10 times more dangerous than the Pb that already exists on the earth. 114 To lower its toxicity, Pb-free and less-Pb PVK based solar cells have been researched using safe tin (Sn)-based PVK. [115][116][117] However, Sn-based PVK solar cells have lower efficiencies and faster degradation than Pbbased PVK solar cells due to phase instability and easy oxidation from Sn 2+ to Sn 4+ . In addition, new approaches have been suggested to prevent Pb leakage in PVK solar modules by trapping Pb with cation-exchange resins containing abundant and inexpensive Ca 2+ and Mg 2+ . 118 Solar cells are exposed to degradation sources such as moisture, oxygen, heat, and ultraviolet light in actual use, and to ensure the long lifetime of PVK/Si solar cells, the stability of PVK solar cells against the degradation sources should be further increased. [119][120][121] To solve the easy decomposition of methylammonium lead triiodide PVK by moisture and heat, optimized PVK compositions were developed by mixing cations and halogen anions to achieve high stability against moisture and heat. Unlike thermally decomposed metylammonium lead triiodide PVK, formamidinium lead triiodide PVK, and cesium lead triiodide PVK showed high resistance to thermal decomposition. By mixing the cations, the thermal stability of PVK improved as well as crystal phase stability. For halide anions, Bromide-based PVK showed higher stability against moisture and heat than the iodide-based PVK, and the PVK composition was further optimized by mixing both halogens for better stability. Further optimization of the PVK composition by incorporating equivalent small metal ions and forming 2D/3D hetero-structured PVK to increase stability have been reported. [122][123][124] It is well known that organic charge transport materials are easily decomposed by moisture and oxygen and require a high level of encapsulation. Inorganic charge transport materials are generally advantageous for stability owing to their original robust properties. Moreover, it was reported that a densely formed inorganic layer could act as a diffusion barrier to prevent the escape of volatiles, and the thermal stability of PVK solar cells with an inorganic layer was significantly improved. It was reported that a semi-transparent PVK solar cell, with a densely formed CTL and transparent electrode, passed the thermal cycling test, damp heat test, and UV stress test ( Figure 9B,C). [125][126][127] Although the above obstacles in the commercialization of PVK/Si tandem solar cells have been discussed in detail in the literature and many potential solutions have been suggested, the high processing temperature of top metal grid metallization of conventional Si solar cells has been overlooked. Large-scale solar cells require a low series resistance of the top electrode, which requires metallization of the top metal grid. However, screen printing that requires a high temperature of 800 C in a Si solar cell is not possible in PVK solar cells, where conventional silver paste metallization is susceptible to heat. In heterojunction Si solar cells, low-temperature plating metallization was developed to prevent degradation of a-Si:H based passivated contacts, but for PVK solar cells vulnerable to moisture, the plating process immersed in a bath with metal salt is also impossible. Although a metal grid was formed on a PVK/Si tandem solar cell by evaporation, its material utilization rate is too low, and high usage of Ag significantly increases the total cost. Recently, a cold metallization process suitable for PVK solar cells was developed from low-temperature cell connection technology, the so-called SmartWire Connection Technology developed by Meyer-Burger. 128 It is a simple lamination process between metal wire embedded polymer foil and the top surface of the solar cell. Low melting-point alloy layer coated copper wire can be used to reduce the wire interconnection temperature to 160 C or less. In this SmartWire Connection Technology, both material usage and optical shading loss can be reduced in conventional wide busbars by optimizing the grid design and wire geometry, and further improvement of PVK/Si tandem solar cell for commercialization should be considered in this new approach.

| CONCLUSION
In this review, we summarize large-scale monolithic PVK/Si tandem solar cells with representative structures of Si-based PVK tandem solar cells, optical and electrical development of PVK/Si tandem solar cells, the process of large-scale PVK solar cells, and challenges of large-scale PVK/Si tandem solar cells. In the 4T tandem solar cell, optical loss of tandem solar cell exists due to the air gap between the PVK top cell and the Si bottom cell. However, as the 2T tandem solar cell is a simple integrated type, it has the advantage of less parasitic absorption but requires a technique such as current matching. Optical and electrical developments have been studied to achieve high-efficiency PVK/Si tandem solar cells. To overcome the reduction in efficiency due to optical loss, a textured surface is used to reduce the reflections between interlayers and optimize the transmittance and conductivity of the top electrode. Meanwhile, electrical properties are improved by many processes, such as the introduction of a sputtering buffer layer. For the commercialization of PVK/Si tandem solar cells, the optical and electrical properties must be improved, and the development of a large-scale PVK layer deposition process is critical. Recently, the spin coating process has been widely used in the solution process by easily controlling the chemical composition, forming a high-quality and high-density PVK film by anti-solvent dropping, and optimizing the thickness and bandgap of the PVK layer. However, spin coating is not suitable for large-scale processes. Therefore, other solution processes, such as spray coating, slotdie coating, and blade coating, were mainly used in large-scale PVK solar cells. A dense and pinhole-free PVK film can be easily formed during the evaporation process, but the control of chemical reactions between organic cations and metal halides is very difficult. Various processes have been developed during the evaporation process, including substrate temperature control, flash evaporation, two-step sequential process, and proximity evaporation to form high-quality PVK layers. However, in PVK/Si tandem solar cells, scale-up is not the only obstacle, but the recombination layer resistance control is also important. The vertical resistance should be low to prevent electrical loss of the recombination layer, but the lateral resistance of the recombination layer should be high to eliminate the shunting path. To overcome the limitation of the TCO layer, the properties of metal grids such as low resistivity metal, short distance between the metal fingers, and large crosssectional area of metal fingers are important. However, PVK/Si tandem solar cells have to overcome several obstacles to commercialization. The commercialization of PVK solar cells with the high material cost is being pointed out, and the toxicity and stability of PVK materials are critical issues for commercialization. Additionally, conventional unfeasible metallization temperatures for PVK/Si tandem solar cells are being overlooked, and these hurdles must be overcome to commercialize PVK/Si tandem solar cells. ORCID Myoung Hoon Song https://orcid.org/0000-0002-8106-7332