Due to stable and high power conversion efficiency (PCE), it is expected that silicon heterojunction (SHJ) solar cells will dominate the photovoltaic market. So far, the highest PCE of the SHJ-interdigitated back contact (IBC) solar cells has reached 26.7%, approximately approaching the theoretical Shockley–Queisser (SQ) limitation of 29.4%. To break through this limit, multijunction devices consisting of two or three stacked subcells have been developed, which can fully utilize the sunlight by absorbing different parts of the solar spectrum. This article provides a comprehensive overview of current research on SHJ-based tandem solar cells (SHJ-TSCs), including perovskite/SHJ TSCs and III–V/SHJ TSCs. Firstly, we give a brief introduction to the structures of SHJ-TSCs, followed by a discussion of fabrication processes. Afterwards, we focus on various materials and processes that have been explored to optimize the electrical and optical performance. Finally, we highlight the opportunities and challenges of SHJ-TSCs, as well as personal perspectives on the future development directions in this field.
Crystalline silicon solar cells have dominated the photovoltaic market for decades. System components including installation, cabling, and inverters account for the main cost of photovoltaics, because these costs are area-dependent. Improving the power conversion efficiency (PCE) is the most powerful means to reduce the levelized cost of electricity. Silicon heterojunction (SHJ) solar cells include c-Si/a-Si , , , , , , , c-Si/poly-Si , , , , c-Si/SiOx , , , , , , and c-Si/MoOx ,  heterojunctions solar cells. SHJ solar cells represent the new research direction due to their high performance. So far, the world record of 26.7% cell efficiency was set by Kaneka in 2017 . It uses an SHJ structure combined with an interdigitated back-contact (IBC). However, this efficiency is only 2.7% lower than the theoretical efficiency limit of 29.4% for silicon single-junction solar cells . This means it is really close to its performance limit. There is an urgent need of novel methods to be adopted.
A well-known strategy to overcome the Shockley–Queisser (SQ) theoretical efficiency limit is to employ multijunction architecture. In these devices, a wide band gap top cell absorbs high-energy photons, and a narrow band gap bottom cell absorbs low-energy photons. This structure can minimize the thermalization losses and improve the utilization of the solar spectrum. According to simulation results, the maximum limiting PCE of dual-junction silicon based tandem solar cells (TSCs) is 45%, and that of triple-junction silicon based TSCs is 50% , . For silicon based TSCs, various top cell materials have been endeavored, but current research focuses on III–V and perovskite semiconductors. III–V solar cells have been known for their high efficiency and excellent reliability, which reduces the invest risk and improves the stability of tandem modules. A PCE of 35.9% was reported for a mechanically stacked four terminal GaInP/GaAs/Si tandem device . This is the highest PCE of silicon based TSCs. Meanwhile, perovskite solar cell is another ideal candidate for SHJ-based tandem solar cells (SHG-TSCs) due to its tunable bandgap, easy fabrication, and high PCE of 25.5% . Using a low temperature solution method, perovskite solar cells can be well compatible with c-Si/a-Si SHJ solar cell. Record one-sun PCE of 29.5% has been achieved for two terminal perovskite/SHJ TSC by Oxford PV . However, the lager Voc loss of nonradiative recombination and the high sensitivity of perovskite to low levels of humidity, remain as significant challenges that require further research and development.
This article reviews current research activities that center on perovskite/SHJ TSCs and III–V/SHJ TSCs. The text of this article is divided into six parts: Section 2 discusses several structures related to the SHJ-TSCs, as well as fabrication strategies that have been developed for achieving ideal SHJ-TSCs. The following three sections thoroughly illustrate these strategies with specific examples. The final section concludes with personal comments on the directions of future research on such new solar cells.
2 Design consideration for TSCs
2.1 The structure of TSCs
SHJ-TSCs mitigate carrier thermalization losses by using several subcells with different bandgaps. Figure 1 illustrates four different SHJ-TSCs structures. As shown in Figure 1(a), two terminal (2T) devices use a tunnel junction or transparent conductive adhesive to connect the two sub cells. 2T devices have advantages in module integration. This is because 2T devices can be connected in standard series or parallel with little space between the cells . In addition, 2T devices cut down the parasitic absorption by removing the intermediate electrode. So far, all industrially scaled multi-junction solar cells use 2T structure. The fabrication of 2T devices faces three major challenges: current matching, high performance recombination junction and performance losses under varying spectra . Among these factors, tunnel junction plays a key role in obtaining high efficiency by providing a low electrical resistance and high optical transmittance interlayer between the subcells. It is expected that 2T devices will be favorable for future applications with the solution of these problems.
In 3T configurations, the two subcells can be connected by a tunnel junction or recombination layer or wiring , , , . The third electrode can be formed at the interface of the top cell and bottom cell, and the third electrode can extract additional generated power by the variation in solar radiance. In addition, when the IBC bottom cell is used, the third electrode can also be formed at the rear side of the bottom cell (Figure 1(b)). Theoretically, 3T devices enable the same performance as 4T devices. For example, Schnabel et al. used IBC bottom cell and GaInP top cell, yielding a PCE of 27.3% for 3T TSC .
For 4T configurations, there are two typical structures: mechanically stacked (Figure 1(c)) and optical splitting 4T tandem devices (Figure 1(d)). 4T configuration eliminates the need for recombination junction and requires neither current matching nor processing compatibility between subcells, hence it offers a more flexible option. The two subcells are subjected to optical coupling but they are electrically independent. Each sub cell can be fabricated independently, regardless of fabrication methods and processes. Therefore, the efficiency of each sub cell can be maximized. However, due to the parasitic absorption and resistance from intermediate electrode, significant optical and electrical losses are inevitable. Meanwhile, the complexity of fabricating these devices and implementing them into modules impede their application. Therefore, 4T devices have not been found any practical applications, until now.
2.2 Band gap matching and optoelectronic performance optimization
The theoretical SQ efficiency limit for single junction solar cells with a bandgap of 1.34 eV is 33.7% . So far, the maximum reported efficiency of GaAs with a bandgap of 1.42 eV is 29.1% (Figure 2(a)). As shown in Figure 2(b), the highest Voc of perovskite solar cell with a bandgap of 1.75 eV is 1.24 V . Because of band gap energy and nonradiative voltage loss, the Voc deficits of perovkite solar cell are much higher than GaInP and GaAs solar cell. The champion cell based on wide band gap n–i–p perovskite solar cell now has a PCE of 21.6% , a significant gap from the SQ efficiency limit. To break this efficiency limitation, one must rely on the optimized multijunction architecture.
The bandgap combination is an essential requisite for the successful fabrication of SHJ-TSCs. As shown in Figure 2(c), the PCEs of four types of SHJ-TSCs are advancing significantly in past six years. So far, the record PCEs of mechanical stacking GaInP/SHJ and GaAs/SHJ are 32.5 and 32.8%, respectively. By detailed balance model with precise ab initio calculations, the optimal top cell bandgaps of 2T and 4T SHJ-TSCs are 1.73 and 1.81 eV (Figure 2(d)–(e)), respectively . And the PCE of 2T and 4T SHJ-TSCs can exceed 45%, which is close to theoretical maximum of 46%. As shown in Figure 2(f), the PCE increase with the number of pn junctions. However, once the number of subcells exceeds three, the gain of PCE for additional sub-cell will become smaller. For triple-junction (3J) tandem devices with SHJ bottom cell, the optimal bandgap combination of top and middle cell is 2.01 and 1.5 eV, respectively. Theoretically, the PCE of triple-junction (3J) Si based TSCs can exceed 50% . However, high quality materials with bandgaps of 2.01 and 1.5 eV are difficult to fabricate, and Ga0.51In0.49P(1.91 eV)/Al0.06Ga0.94As (1.51 eV) is a good alternative to be the top and middle cell on SHJ. According to the theoretical simulation, a maximum PCE of GaInP/GaAs/Si is close to 44%. The world-record PCE of GaInP/GaAs/Si was lower than 36%, far from reaching the theoretical limit . Further improvements in the future are crucial.
3 SHJ bottom cells
As shown in Figure 3(a), typical SHJ solar cell is c-Si/a-Si heterojunction solar cell, also named heterojunction with intrinsic thin-layer cell (HIT) , , , , , . HIT solar cell is commonly used bottom cells in perovskite/Si TSCs. This selection can be attributed to a number of scenarios that may include: (1) An ultrathin hydrogenated amorphous silicon layer is used in HIT to realize excellent surface passivation (Figure 3(b)) . The open-circuit voltage of HIT is as high as 750 mV , which is very close to Auger-limited 761 mV with the similar thickness. (2) The transparent electrode of HIT is usually tin-doped indium oxide (ITO), which is a good substrate and recombination layer for perovskite top cell . (3) The HIT cell has good near infrared (NIR) response, which can reduce the optical loss of TSCs. However, HIT cell has temperature limits . A HIT solar cell is not suitable for 2T III–V/Si TSCs because the GaAs substrate is chemically etched. When encountering sulfuric acid, hydrochloric acid and ammonia solution etching, HIT solar cell will degrade. Therefore, other heterojunction structures that replace the c-Si/a-Si heterojunction have attracted the interest of researchers.
New efficient SHJ solar cells may include double side tunnel oxide passivating contacts(TOPCon) solar cell , , polycrystalline silicon on oxide (POLO) passivating contacts solar cell , , Silevo Triex solar cell , and dopant-free asymmetric heterocontacts (DASH) solar cells , . Among this, TOPCon may replace HIT with its high efficiency and temperature stability. So far, most high efficiency 2T III–V/Si TSCs use TOPCon as bottom cell , , . For instance, a record PCE of 34.1% has been achieved for a wafer bonding GaInP/GaAs/Si TSC with a TOPCon bottom cell . Polysilicon/SiOx passivating contacts provide an ideal design to obtain full-area carrier-collecting junctions . This full-area contact avoids the lateral transport of charge carrier, enables the use of higher resistivity wafers (Figure 3(b)). TOPCon is resistant to light-activated degradation. The annealing temperature above 800 °C form polysilicon contacts make it compatible with the high temperature process of III–V or perovskite top cells. However, the p+ polysilicon contacts of TOPCon are difficult to optimize. This may be due to excessively large valence band offsets between p+ polysilicon and the silicon wafer .
4T SHJ-TSCs usually use polycrystalline silicon on oxide-interdigitated back contact (POLO-IBC) cell as bottom cell. Carrier selective junctions are formed by creating polycrystalline silicon-rich layers on ultrathin silicon oxide films , . The pinhole density in interfacial oxide films of POLO junction plays a key role in determining the electrical properties , . As shown in Figure 3(c), Haase et al.  optimized the layer-selective laser ablation process of SiO2 and obtained a saturation current density of 2 fA ⋅ cm−2 on n-type POLO. In this approach, the device exhibited a PCE of 26.1% with a Voc of 733 mV. Rienacker et al. prepared GaInP/GaAs//Si triple-junction cell with POLO-IBC bottom cell and achieved a certified PCE of 35.4% .
Compared to HIT concept, which suffers parasitic absorption in the window layer, the DASH cells, featuring wide band gap carrier-selective materials on opposite sides, delivers lower current losses in the blue spectrum (Figure 3(e)). Besides, DASH removes the limitations of doped amorphous silicon layers and reduces the recombination-active defects caused by excessive dopant. Therefore, DASH opens a wider optical and electrical parameter space to optimize different solar cell components. Using MoOx and LiFx based heterocontacts with a-Si:H passivation layers, Bullock et al. obtained a PCE of 20.7% efficient DASH solar cell (Figure 3(d)) . Based on the concept of metal–insulator–semiconductor, Silevo Triex cell is a proprietary configuration that combines a tunnel oxide junction with doped a-Si:H film emitter layers on silicon wafers. With high quality passivation of ultrathin SiOx layers, excess carriers can be collected at the emitter by quantum tunneling (Figure 3(f)). By optimizing key processes to improve passivation, junction quality, and shading effects, Heng et al. achieved a PCE of 23.1% . Compared with HIT bottom cell, Silevo Triex cell has higher FF and Jsc. Because of less parasitic absorption and carrier recombination, Silevo Triex can be used as bottom cell for TSCs. However, the PCE of Silevo Triex cell is still lower than that of HIT, which limits the application in TSCs.
4 Perovskite/SHJ TSCs
Perovskite/SHJ TSCs have shown PCEs soaring from 13.7% to over 29.5% within merely six years’ research and development. Tables 1 and 2 show the evolution of the 2T and 4T perovskite/SHJ TSCs, respectively. Perovskite is typically originated from a 3D crystal structure consisting of three primary ions with a stoichiometry of ABX3, where A = methylammonium (MA), formamidinium (FA), cesium (Cs), or rubidium (Rb); B = lead (Pb) or tin (Sn); X = iodine (I), bromine (Br), or chlorine (Cl). The crystalline ABX3 structure can be described as a specifical 3D sublattice in which A+ is surrounded by BX6 octahedra. So far, the hybrid perovskite compounds are prepared with either MA or FA as the A cation. By interchanging the above cations , , , , , , , , , , metals , and halides , , the band gap of perovskite can be adjusted in the range of 1.15–3.06 eV. Organic–inorganic lead halide perovskite solar cells exhibit many advantages that make them attractive for using as top cells in SHJ-TSCs [42, 51–67, 86–89].
|Si bottom cell||Perovskite top cell||Eg (eV)||Recombination layer||Area (cm2)||VOC (V)||JSC (mA/cm2)||FF (%)||Eff. (%)||Research institution||Year|
|Front planar SHJ||Cs0.05(FA0.77MA0.23)0.95Pb(I0.77Br0.23)3||1.68||ITO||1.0599||1.9||19.26||79.52||29.15||HZB/KTU ||2020|
|Textured SHJ||(Cs0.06FA0.78MA0.16) Pb (Br0.17I0.83)3||1.64||ITO||1.43||1.8||18.81||77.5||26.3||Tor Vergata- ||2020|
|Front planar SHJ||Cs0.75MA0.22Pb(I0.82Br0.15Cl0.03)3||1.67||ITO||1||1.876||18.55||74.85||26.05||CUB/NREL/USTC/SU/ASU ||2020|
|Textured SHJ||Cs0.1MA0.9Pb(I0.9Br0.1)3||–||ITO||0.49||1.82||19.2||75.3||26.2||UNC/ASU- ||2020|
|Front planar SHJ||FA0.65MA0.2Cs0.15) Pb(I0.8Br0.2)3||1.68||ITO||0.188||1.756||19.1||79.2||26.7||KAIST/NU/SNU/NREL- ||2020|
|Front planar SHJ||Cs0.05(MA0.83FA0.17) Pb(I0.83Br0.17)3||–||ITO||0.7709||1.77||19.22||76.6||26||HZB/TU/e/TUB- ||2020|
|Textured SHJ||Cs0.05MA0.15FA0.8PbI2.25Br0.75)||1.68||ITO||0.832||1.7932||19.08||75.1||25.7||UToronto/KAUST/NREL/USD ||2020|
|Front planar SHJ||Cs0.05(MA0.17FA0.83) Pb1.1(I0.83Br0.17)3||1.66||ITO||0.81||1.76||19.4||77||26.5||HZB/ULJ ||2019|
|Front planar SHJ||Cs0.05(FA0.83MA0.17)0.95Pb(I1−xBrx)3||1.63||ITO||0.092||1.79||19.04||74.6||25.43||HZB ||2019|
|Front planar SHJ||Cs0.15(FA0.83MA0.17)0.85Pb(I0.8Br0.2)3||1.64||ITO||0.49||1.8||17.8||79.4||25.4||UNL/UNC/ASU/CAS ||2019|
|Front Planar SHJ||Cs0.17FA0.83PbI0.83Br0.17||1.63||ITO||57.4||1.723||17.48||75||22.6||CSEM/EPFL- ||2019|
|Textured SHJ||FA0.9Cs0.1PbBr0.325I2.675||1.6||nc-Si||1.42||1.788||19.5||73.1||25.2||EPFL/CSEM ||2018|
|Front planar SHJ||FA0.75Cs0.25Pb(I0.8Br0.2)3||1.68||ITO||1||1.77||18.4||77||25||SU/ASU/TCD- ||2018|
|Planar SHJ||Cs0.08FA0.74MA0.18Pb(I0.88Br0.12)3||1.65||ITO||0.13||1.78||17.82||75||23.73||NKU ||2018|
|Front Planar SHJ||Cs0.19MA0.81PbI3||1.58||nc-Si||0.25||1.75||16.8||77.5||22.8||EPFL/CSEM ||2018|
|Front planar SHJ||Cs0.19MA0.81PbI3||1.58||nc-Si||0.25||1.751||16.8||77.5||22||EPFL/CSEM||2018|
|Front planar SHJ||CsFAPbI3−xBrx||1.63||nc-Si||12.96||1.769||16.5||65.4||19.1||EPFL/CSEM ||2018|
|Front planar SHJ||FA0.83Cs0.17Pb(I0.83Br0.17)3||1.63||ITO||1||1.65||18.1||79||23.6||Stanford/ASU/MIT/Oxford ||2017|
|Front planar SHJ||MAPbI3||1.55||IZO||1.43||1.717||16.4||73.1||20.5||EPFL/CSEM ||2016|
|Double side planar SHJ||MAPbI3||1.55||ITO nc-Si||0.12||1.76||14||77||18.1||HZB/EPFL ||2016|
|Double side planar SHJ||MAPbI3||1.55||IZO||0.17||1.69||15.9||79.9||21.2||EPFL/CSEM ||2015|
|Si bottom cell||Perovskite top cell||Eg (eV)||Area (top; bottom) cm2||VOC (V)||JSC (mA/cm2)||FF (%)||Eff. (%)||Research institution||Year|
|Textured SHJ||Cs0.05FA0.81MA0.14PbI2.55Br0.45||–||0.049, 0.049||1.12, 0.7||22.3, 17.2||77.7, 76.6||19 + 9.2 = 28.2||UToronto/KAUST ||2020|
|Textured SHJ||MAPbI3||1.55||0.096;1||1.156, 0.698||19.8, 15.6||79.9, 80.1||18.3 + 8.7 = 27||UCAS/SNU ||2020|
|Topcon||Cs0.05FA0.8MA0.15PbI2.55Br0.45||1.63||0.059, 0.1||1.1, 0.675||20.6, 16.6||78.9, 79.6||17.8 + 8.9 = 26.7||GT/NREL ||2020|
|Textured SHJ||MAPbI3||1.55||0.16, 4.2||1.008, 0.708||19.8, 18.8||78, 79.3||10.6 + 15.6 = 26.2||KAUST ||2019|
|Textured SHJ + IBC||CsMAFAPbI3−xBrx||1.63||0.36;4||1.1, 0.69||21.5, 14.2||73.5, 81||16.6 + 7.9 = 24.5||ANU/SYSU ||2017|
|Textured SHJ||MAPbI3||1.55||0.25;4||1.069, 0.693||20.1, 15.98||76.1, 79.5||16.4 + 8.8 = 25.2||EPFL/CSEM ||2016|
|Textured SHJ||MAPbI3||1.55||1;4||1.057, 0.692||19.3, 15.5||71.6, 79.4||14.5 + 8.5 = 23.0||EPFL/CSEM ||2016|
|Textured SHJ||MAPbI3||1.55||0.075;4||1.08, 0.679||20.6, 12.3||74.1, 77.9||16.5 + 6.5 = 23.0||UNL/ASU ||2016|
|Textured SHJ||FA0.83Cs0.17Pb(I0.6Br0.4)3||1.74||0.09;na||1.10, 0.69||19.9, 13.9||71, 76||12.5 + 7.3 = 19.8||Oxford/HZB ||2016|
|Textured SHJ||MAPbI3||1.55||0.25;4||0.938, 0.693||17.36, 15.81||59.6, 77.4||10.4 + 7.8 = 18.2||EPFL/CSEM ||2015|
|Textured SHJ||MAPbI3||1.55||0.25;4||0.821, 0.689||14.5, 13.7||51.9, 76.7||6.2 + 7.2 = 13.4||EPFL/CSEM/ASCR ||2015|
As shown in Figure 4(a), early designs for 2T perovskite/SHJ TSCs relied on double side planar SHJ bottom cell, which resulted in significant optical losses. Werner et al. employed double side planar SHJ as bottom cell, yielding a PCE of 19.2% for 2T TSC (Figure 4(b)). Because the reflectance and parasitic absorption losses, the external quantum efficiency (EQE) current density of bottom cell was only 15.6 mA ⋅ cm−2. An improved structure based on rear side textured SHJ bottom cell can mitigate these losses. SHJ bottom cell with front side planar is compatible with existing solution-based spin-coating perovskite fabrication processes. However, this configuration leads to strong front surface reflection losses and poor light trapping, especially in the NIR area where SHJ bottom cell absorbs weakly . Currently, the best device with rear side textured SHJ exhibits a total cumulative current of 39.6 mA ⋅ cm−2  (Figure 4(c)), a value that is lower than that of double side textured SHJ-TSCs (40.4 mA ⋅ cm−2).
Various textured light management structures on the front side have been employed to fully harvest the high photocurrent. For example, McGehee et al.  used pyramidal textured polydimethylsiloxane (PDMS) to reduce reflection loss. The EQE current density of perovskite top cell increased by 1 mA ⋅ cm−2. But the EQE current density of SHJ bottom cell decreased due to PDMS layer was incapable in reducing reflection at longer wavelengths. Textured antireflection foil is another effective way to increase the photocurrent of tandem devices. With antireflection foils, the EQE current density gain of perovskite top cell exceeds 2 mA ⋅ cm−2 . The reflection loss is reduced to 2.5 mA ⋅ cm−2. However, the effectiveness of antireflection foils will be impaired during encapsulation . The high processing costs of atomically smooth surfaces makes it difficult as candidate for practical application in SHJ-TSCs.
To fully harvest the incident photons and reduce costs, double side textured c-Si bottom cell has been used (Figure 4(f)). Sahli et al. used fully textured c-Si bottom cell with random pyramids, resulting in a certified PCE of 25.2% (Figure 4(g)) . The reflection loss was only 1.64 mA ⋅ cm−2. A total photocurrent density was 40.4 mA ⋅ cm−2, which was close to that of the record Si device (42.5 mA ⋅ cm−2) (Figure 4(h)). However, to fabricate conformal and shunt-free perovskite cell, most layers were thermally evaporated in vacuum. Wide-band-gap perovskite materials have mixed cations and mixed halides. This composition requires four or more organic and inorganic precursor sources. Some trace additives, such as KI, PEAI, and Pb(SCN)2, butanethiol, exacerbate the difficulties of control over the deposition rates during the evaporation process . It seems difficult to build up efficient tandems using thermal evaporation. Another approach is to shrink the size of the pyramids and to increase the perovskite charge-carrier diffusion length. Because the thickness of perovskite layer exceeds the height of the pyramid, to be conformal is not required and spin coating can be used. For example, Huang et al. used solution-based blading of perovskites to deposit on the fully textured c-Si bottom cell with pyramids with height less than 1 μm. The tandem device exhibited a PCE of 26.2% with Voc, Jsc and FF of 1.82 V, 19.2 mA ⋅ cm−2 and 75.3% . They found that pyramids with height over 1 μm are not suitable for spin coating method, because perovskite films with 0.5–1 μm thickness cannot completely cover the pyramid. And the thickness of perovskite films is limited by the diffusion length of charge carrier. Meanwhile, pyramids with height of less than 0.43 μm increase the reflectance to 14.2%. Therefore, pyramids with a height ranging from 0.43 to 2.5 μm is a good choice. As shown in Figure 4(d), Hou et al.  used spin-coating micrometer-thick perovskites with fully textured SHJ bottom cell, reaching a certified PCE of 25.7% (Figure 4(d)). To improve collection of photogenerated carriers, they increased the depletion width in the perovskites layer . They demonstrated that the depletion region in the perovskite layer at the valley of Si pyramids was wider (380 nm) than at the top (120 nm) (Figure 4(e)). This result indicates that the depletion width increases according to the thickness of perovskite layer, which is useful for charge collection.
The performance of mechanically stacked 4T tandem cells also increased [65, 68–76]. The first report of perovskite/SHJ mechanically stacked tandem cell was published in 2016 by Löper et al. , showing a PCE of 13.4%. The latest record was held by Sargent et al.  at the UToronto, with a 28.2% PCE obtained from a 4T tandem measurement combining a 0.049 cm2 perovskite top cell with a SHJ bottom cell. Up to now, most published 4T performance results were based on indirect measurements involving a small perovskite top cell (<1 cm2) and a lager silicon bottom cell [65, 68–72, 90]. However, some progress has already been made in upscaling the perovskite top cell. For example, semitransparent modules with a larger aperture area (4 cm2) have been demonstrated, reaching a PCE of 12%, which resulted in a module-on-cell 4T with 20.2% PCE .
4.2 Performance optimization
The performance optimization of perovskite/SHJ TSCs mainly include four aspects: transparent conducting electrode [75, 92–98], electron and hole transport layers (ETLs and HTLs) , , , , perovskite layer , , , , , , , , , , and recombination layer , .
4.2.1 Transparent conducting electrode
In perovskite-based tandems, the top cell should be highly NIR transparent (800–1200 nm) and fully harvest the solar spectrum of blue part (300–800 nm) . Therefore, the front transparent electrode of top cell should avoid parasitic absorption in the blue part of the solar spectrum. The rear transparent electrode of top cell in 4T tandems should minimize the free-carrier absorption in the red part. The losses of parasitic absorption can be reduced by decreasing the free-carrier charge. However, decrease in the carrier charge means bad conductivity for charge collection. A trade-off between the transparency and conductivity against carrier density is observed in these materials. Based on these criteria, transparent conductive oxide (TCO) films have been explored for perovskite-based tandems, which include Sn-doped In2O3 (ITO) , H-doped In2O3(In2O3:H) , zinc oxide-doped indium oxide (IZO) , tungsten-doped indium oxide (IWO) , Zr-doped indium oxide (IZRO)  and so on. Among this, ITO is commonly used one due to its wide bandgap of 3.5–4.3 eV and high work function of 4.8 eV . Besides, ITO has high available transmittance (above 80%) in visible light and low electrical resistivity (1.5–2.0 × 10−4 Ω cm). However, it has high parasitic absorption in NIR range (Figure 5(a)). To reduce resistivity, it requires high deposition temperatures or post annealing to achieve ideal crystalline structure. This would degrade the perovskite cell. Magnetron sputtering process can also damage the underlying interface layers and perovskite absorbers. In2O3:H has high mobility and optical transparency. But the water vapor effusion would degrade the device performance. Similarly, IZO can be deposited at low temperature and low power, without annealing and the addition of oxygen. Unfortunately, the band edge is blurry, which decreases the transmittance in blue part of spectrum. Besides, IWO can also be used as transparent electric anode, which exhibits 0.6% parasitic absorption in near infrared spectrum. Zheng et al.  employed IWO transparent electric anode, yielding average reflectance over 88% in the wavelength range of 750–1100 nm. As shown in Figure 5(a), IZRO has been a good choice due to high NIR transparency and conductivity. Aydin et al. used IZRO to replace the ITO, improving the Jsc from 15.6 to 17.4 mA ⋅ cm−2, achieving a PCE of 26.2% for 4T perovskite/SHJ solar cell. They demonstrated that the electron mobility could increase to as high as 77 cm2 V−1 s−1 and the sheet resistance is only 18 Ω sq−1 .
Generally, there are two ways to reduce the sputtering damage of TCO: (1) depositing sputter barrier layers , ; (2) using ultrathin metal electrode instead of TCO. For n–i–p cells, thermally evaporated MoOX is a commonly used sputter barrier layer . Because it interacts with the Ar plasma in the sputter deposition, it shows that parasitic absorption increases with the increase of oxygen vacancies. Other metal oxides, such as NiOX, VOX and tungsten oxide (WOX) can still be suitably used as the sputtering buffer layer. For the p–i–n structure, zinc oxide (ZnO) nanoparticles ,  and atomic layer deposited (ALD) SnO2  layer are used as sputtered buffer layer to avoid bombardment. For example, Bush et al.  used a 4 nm ALD SnO2 and 2 nm zinc tin oxide (ZTO) window layer between the front ITO and LiF/PCBM electron transport layer, yielding a PCE of 23.6% for a 2T tandem device. They found that an SnO2 layer of 4 nm is necessary to prevent Zn in reaching underneath layers and improve stability for tandems. The only problem of ALD SnO2 is the effect of water during ALD process . Ultrathin metal electrodes, such as Au, Ag, Al, and Cu have been employed. Huang et al.  found that Cu and Au have better stability than Ag and Al under ambient atmosphere. In addition, Cu/Au  and MoOX/Au  have also been demonstrated to be transparent conducting electrode with simple manufacturing processes. Yang et al. used MoO3/Au/MoO3 sandwiched gold nanomesh as electrode to achieve a PCE of 27.0% for 4T tandem device. The sheet resistance is 19.6 Ω sq−1 . However, the average transmission is only 56% in the wavelength range of 800–1100 nm (Figure 5(a)). The major problem for ultrathin metal is strong parasitic absorption in the top cell’s subbandgap spectral range.
4.2.2 Electron and hole transport layers
The separation and transport capability of free charge carriers are critical for ETLs and HTLs in perovskite/Si tandem device. For efficient tandem devices, HTLs should meet the following criteria: (1) The highest occupied molecular orbital (HOMO) energy or the valence band values should be slightly higher than the valence band of the perovskite layer to improve the hole transfer and electron blocking (Figure 5(b)). (2) High hole mobility is desirable for better transporting . (3) High transparency is required to avoid parasitic absorption. (4) They should be thermally and photochemically stable. (5) They should be processed easily and cost-effective. Generally, HTLs can be divided into three categories: polymeric, small organic molecule, and inorganic. The widely used small molecule for HTL, Spiro-OMeTAD shows strong parasitic absorption in the blue and UV spectral region. Another drawback of Spiro-OMeTAD is high thermal degradation rate. Therefore, polymeric poly(bis(4-phenyl) (2,4-bimethylphenyl) amine) (PTAA) has been used instead of Spiro-OMeTAD. For example, Huang et al. used PTAA with a thickness of 15 nm as HTL to achieve a PCE of 26.2% for 2T tandem device. The spin coated PTAA is thin, uniform and conformal on the pyramids of SHJ bottom cell . Saliba et al. demonstrated that PTAA has a better stability than Spiro-OMeTAD at 85 °C. However, the HOMO energy value of PTAA is about −5.1 eV, which is mismatched to the valance band maximums of perovskite with a range of −5.43 to −5.71 eV . Stolterfoht found that PolyTPD has better HOMO energy value to improve the Voc of SHJ-TSCs (Figure 5(c)). Other polymeric HTLs, P3HT, and PEDOT:PSS also have drawbacks, such as complex purification process, tricky characterization with broad molar mass distribution, and poor batch-to-batch reproducibility . Therefore, many groups have manufactured alternative HTLs with high hole mobility and low cost, such as inorganic NiO, CuI, and CuSCN . NiO is a wide band gap HTL that has been exploited successfully in perovskite/SHJ TSCs. Xu et al.  used NiOx/polyTPD bilayer HTLs to achieve a Voc of 1.886 V for 2T tandem device. They demonstrated that 20-nm-thick sputtered NiOx could reduce the shunting. However, the photostability of NiOx based TSCs is poorly explored .
For efficient perovskite cells, the ETLs extract the electrons from the perovskite layer to the cathode. Therefore, good energy-level alignment and high electron mobility are essential to guarantee the electron transport. The deeper valence band edge (VB) is needed to block injection of holes. High transmittance can reduce the parasitic absorption losses. Besides, the stability and cost also influence the application of ETLs . There are several different types of ETLs include TiO2, SnO2, ZnO, phenyl-C61-butyric acid methyl ester (PCBM), and C60. Historically, a TiO2 mesoporous scaffold layer was firstly used as ETL with a sintering step at 500 °C in perovskite solar cell . However, this relatively high temperature is not suitable for c-Si/a-Si heterojunction bottom cell. ZnO owns high electron mobility of 205–300 cm2 V−1 s−1 and low temperature process. But it will react with perovskite when the annealing temperature exceeds 100 °C . As shown in Figure 5(b), SnO2 owns deep valence band level with a range of −4.2 to 4.5 eV and high electron mobility of 240 cm2 V−1 s−1. However, the solution processing of SnO2 would dissolve the already deposited perovskite layer for pin perovskite/SHJ TSC. Therefore, SnO2 buffer layer is commonly prepared by low temperature ALD. SnO2 also has some drawbacks such as interface recombination and inefficient electron extraction . Therefore, fullerene (C60) and its derivatives such as PCBM were used to passivate the perovskite surface defects and suppress hysteresis . Xu et al.  used C60/bathocuproine (BCP) as ETLs to achieve a PCE of 27.13% for 2T perovskite/SHJ TSC. Because the parasitic absorption of C60, the current density is only 19.12 mA·cm−2. The morphology control and frontier orbital position also need to be optimized to enhance the performance.
4.2.3 Perovskite absorption layer
At first, MAPbI3 was used as top cell for perovskite/SHJ TSCs. However, the band gap of MAPbI3 is about 1.55–1.63 eV, which is below the optimize range of 1.67–1.75 eV . The band gap of MAPbI3 can be increased by replacing part of iodine with bromine at the halogen anion site X. Unfortunately, wide band gap perovskite with high Br content is limited by photoinduced phase segregation . Under illumination, the perovskite decomposes into Br- and I-rich regions. The I-rich regions own narrow band gap, and act as gap defect or trap states. Light induced phase segregation leads to significant Voc loss for perovskite cells, so wide band gap perovskite solar cell exhibits larger Voc deficit (Eg/q − Voc) than narrow phase.
As shown in Figure 5(d), mixed cations of FA, Cs, and MA at A-site is the best choice for the perovskite absorption layer. The mixed cations of FA, Cs, or Rb can reduce phase segregation and improve material quality . However, the FA/Cs based perovskite film has small grains, which leads to high density of grain boundaries and high surface recombination . Therefore, several approaches have been proposed, such as increasing grain size, surface passivation, 2D/3D heterojunction engineering, and ion compensation . For example, Al–Ashouri et al.  used a composition of Cs0.05(FA0.77MA0.23)0.95Pb(I0.77Br0.23)3 and obtained a PCE of 29.15% and a maximum Voc of 1.9 V in 2T perovskite/SHJ TSC. Performance improvements in wide band gap perovskite solar cell was made by increasing the Cl content in lattice and replacing I partially with Br . Huang et al. used a Cs0.15(FA0.83MA0.17)0.85Pb(I0.7Br0.3)3 perovskite top cell with band gap of 1.70 eV and achieved a PCE of 25.4% for 2T perovskite/SHJ TSC . They found that the addition of MACl could increase the grain size, while the addition of MAH2PO2 could reduce nonradiative recombination by passivation of grain boundary . Kim et al.  used 2D/3D mixed wide band gap perovskite-(FA0.65MA0.2Cs0.15)Pb(I0.8Br0.2)3-as top cell, and obtained a PCE of 26.7% for 2T perovskite/SHJ TSC. PEA-based additives have been proven to be critical to form the 2D passivation layer and improve the Voc and FF. Sargent et al. used a Lewis base as additive to reduce the trap density and increase the electron-diffusion length, yielding a PCE of 19% for semitransparent perovskite top cell. By combining this perovskite cell with a textured SHJ cell, the best PCE of 4T tandem device is 28.2% .
4.2.4 Recombination layer
For high efficiency TSCs, the tunnel junction should meet the following standards: (1) minimal resistive losses and maximal optical transparency; (2) a barrier to prevent the solvent and ions from penetrating into bottom cell; (3) poor in-plane conductivity and good out-plane conductivity to reduce shunts; and (4) compatible with the above and underneath layer fabrication processes. As shown in Figure 5(e), the first tunnel junction of 2T TSC was based on crystallized p++-Si/n++-Si recombination layer. In essence, the quasi-Fermi level (EF) on both side of the p–n-junction transfers to the valence (EV) and conduction (EC) band due to the high doping levels. Once a positive voltage is applied to the p–n-junction, full states of the n-side align with empty states of the p-side. To activate the dopants, the a-Si:H layer was annealed at 680 °C for 15 min . This high temperature is not suitable for the c-Si/a-Si heterojunction bottom cell. To address this issue, TCO-based tunnel junctions have been fabricated by low-temperature process such as ITO, IZO, and ZTO. For example, Albrecht et al. used 20 nm ITO layer as tunnel junction and achieved a certified PCE of 25% for 2T perovskite/SHJ TSC . The EQE current density of perovskite and c-Si were 20.69 and 17.85 mA ⋅ cm−2, respectively, indicating that the current was limited by the bottom cell of c-Si, which was mainly due to the refractive index mismatch between ITO and its adjacent layers. To address this issue, Mazzarella et al. employed nanocrystalline silicon oxide (nc-SiOx:H) interlayer under the ITO, resulting in 1.4 mA ⋅ cm−2Jsc gain in the bottom cell. As shown in Figure 5(f), the 110 nm thick nc-SiOx:H with refractive index of 2.6 (at 800 nm) was the best choice for infrared light management . At present, the lowest electrical resistivity of ITO is 1.5 × 10−4 Ω cm. However, the lateral conductivity and longitudinal conductivity of ITO are the same, which leads to shunt paths for top cell. Therefore, nc-Si:H tunnel junction was fabricated to replace ITO. The Eg of nc-Si:H is narrower than that of ITO, so it only absorbs the blue region of the spectrum that the top cell has absorbed. Besides, the high-density grain boundary of nc-Si:H impedes the lateral transport of charge carriers. The large single grains span across the tunnel junction to improve the longitudinal conductivity. By implementing nc-Si:H tunnel junction in a 2T tandem device, Sahli et al. obtained a PCE of 25.2%. The EQE current density of the tandem device was improved to 19.5 mA ⋅ cm−2. This work represents a key step toward controlled fabrication of efficient large area perovskite/SHJ TSCsusing the knowledge of nc-Si:H recombination layer.
4.3 Toward market
For market application, perovskite/SHJ TSCs are facing several challenges such as long-term stability, device area upscaling, and material toxicity . The stability of perovskite solar cells depends on the surrounding environment including humidity , , oxygen , temperature , , and light illumination [81, 120, 123]. The halogen ions at X-site influence the photoinstability of wide band gap perovskite solar cells. As discussed before, using partial Cl to replace I could improve the photostability of perovskite material . The instability of moisture and oxygen is closely related to the A-site cation. Cs-doped mixed-cation perovskite shows long time stability under heat stress and humidity environments . For example, Xu et al.  used Cs0.75MA0.22Pb(I0.82Br0.15Cl0.03)3 perovskite top cell to achieve less than 4% degradation after 1000 h of operation at 60 °C under 100-sun illumination. Further stability improvement can be seen through appropriate glass/polymer encapsulation. Encapsulating with glass and butyl rubber, Hou et al.  demonstrated that the TSCs retained their original performance at 85 °C for 400 h in the dark.
At present, the fabrication of large area perovskite/SHJ TSCs (>1 cm2) and modules (>10 cm2) faces many challenges, such as fabrication method, film quality, and long-time stability. Due to poor uniformity and material waste, spin-coating is not suitable for upscaling perovskite/SHJ TSCs. Large area deposition methods have been presented such as inkjet printing, blade coating, spray coating, slot-die coating. However, commercial c-Si cells are textured, which hinders the use of solution methods. Therefore, vapor deposition has been developed to make uniform large area perovskite TSCs . First demonstration of large area perovskite/SHJ TSC was presented by Sahli et al. , with 18% PCE of 12.96 cm2. Encouragingly, Oxford PV obtained a PCE of 28% for 2T perovskite/SHJ TSC (1 cm2) by a vapour deposition. This device has been passed the 2000 h damp heat reliability test (IEC 61215). To achieve the commercial application of perovskite/SHJ TSCs, it is necessary to put more effort into large area cells while maintaining the high PCE.
Lead toxicity is another issue related to the photovoltaic application of perovskite materials. To this end, several groups have demonstrated the ability to fabricate lead-free, or lead-less perovskite solar cells . However, the PCE of these devices are much lower than that of current lead-containing devices. In addition, environmentally friendly solvents should be considered to replace toxic solvents such as dimethylformamide and dimethyl sulfoxide. It is no exaggeration to say that the research to put perovskite/SHJ solar cells into practical application is still at a rudimentary stage.
4.4 PCE prospect and improvement strategy
Currently, 2T perovskite/SHJ TSCs have reached a high PCE of 29.5%, and further improvement strategy should be proposed. The record-efficiency top cell parameters are shown in Figure 6(a)–(d) for 2T and 4T SHJ tandem devices. Compared to GaInP top cell, the main limitations of wide band gap perovskite cells are Voc and FF. A minimum Voc deficit of perovskite top cell is 0.45 V , which is far below than the optimum 0.34 V . It is noticed that the Voc deficit of GaInP (1.81 eV) cell is only 0.32 V (Figure 6(b)). This indicates that there is large potential for improving the Voc of perovskite/SHJ TSCs. By passivating defects to reduce nonradiative recombination, the Voc of 1.67 V of the perovskite top cell can be enhanced to 1.35 V, while the Voc of SHJ cell can be increased to 0.75 V by reducing the thickness of Si wafer. The total Voc can be increased to 2.08 V for 2T perovskite/SHJ TSCs. Currently, the record FF of 2T perovskite/SHJ TSC is only 79.9%, which is much lower than the 89.8% of 2T III–V/Si TSC (Figure 6(d)) . A main FF limitation of perovskite solar cells is the ideality factor (n) . The optimal value of the ideality factor is about 1–1.3. However, the high efficiency perovskite solar cells exhibit high values of 1.4 to 1.8. Therefore, it is essential to reduce the ideality factor by minimizing nonradiative recombination. Noticing that point defect such as organic cation vacancies and halide anion vacancies in grain boundaries is closely tied to the nonradiative recombination . Thus, charged components compensation by quaternary ammonium and halide ions may be a good choice to reduce nonradiative recombination. In addition, the contact resistance between the transparent electrode and the grid also limits the FF. Therefore, reducing contact resistance and improving carrier extraction are beneficial to obtain the ideal FF. The theoretical Jsc can reach to 22 mA ⋅ cm−2 for perovskite top cell with a band gap of 1.75 eV . Up to now, the record Jsc of perovskite/SHJ TSCs is 19.22 mA ⋅ cm−2, while the record Jsc of semitransparent perovskite top cell is 19.6 mA ⋅ cm−2 with a band gap of 1.68 eV (Figure 6(c)) . By depositing conformal perovskite layer on textured Si cell, the Jsc is predicted to achieve 20 mA ⋅ cm−2 . By these optimizations, a tandem efficiency of 35% could be expected a Voc of 2.08 V, a JSC of 20 mA ⋅ cm−2 and an FF of 85%. Then the PCE reaches the 77% of their SQ-limit of 45.7% (Figure 6(a)).
5 III–V/SHJ TSCs
Compared with perovskite materials, III–V compound semiconductor materials, with their excellent proven reliability and adjustable band gap, exhibit high efficiency and promising potential for industrial application. In many cases, III–V semiconductor multijunction solar cells are superior to other solar cells. However, the high costs preclude their use for nonconcentrator terrestrial application. To address this issue, III–V/SHJ TSCs have appeared on the road map of many PV companies. At present, many strategies have been used to fabrication III–V/SHJ TSCs, including heteroepitaxial growth, wafer bonding and mechanical stacking. The purpose of this section is to summarize all these studies, with an emphasis on examples demonstrated by wafer bonding.
5.1 Heteroepitaxial growth
Heteroepitaxial integration approach is probably the earliest explored approach towards the fabrication of III–V/Si TSCs such as AlGaAs/Si , GaAsP/Si . In principle, it is possible to fabricate high efficiency III–V/Si tandem devices by controlling the density of threading dislocations. The growth of a polar III–V compound semiconductor on nonpolar Si would create defects such as antiphase domains (APDs). These APDs can be effectively removed by using a (100) Si wafer with a 4°–6° offcut in the (111) direction . Another obstacle is the 4% lattice-mismatch between GaAs (αGaAs = 0.56 nm) and Si (αSi = 0.54 nm), which can lead to threading dislocations and misfit dislocations (Figure 7(a)). Two possible approaches have been proposed, one is the thermal-cycle annealing (TCA), and the other is low growth rate and low temperature process during the primal GaAs nucleation on Si. For instance, Akiyama et al.  have demonstrated that GaAs layer can be directly grown on (100) Si substrates without buffer layers. Firstly, the 10 nm thick GaAs layer was deposited at 400 °C to enhance the nucleation. Then, the temperature was increased to 650–700 °C to obtain a better and uniform epitaxial layer. Yamaguchi et al. found that TCA could effectively increase the dislocations movement and reduce the stress in the crystal, thus reducing dislocation densities. The difference of thermal expansion coefficients between GaAs (5.73 × 10−6 °C−1) and Si (2.6 × 10−6 °C−1) lead to residual strain in the material, which will result in poor crystalline quality. Inserting buffer layers between the III–V material and the Si substrate is a promising approach to limit the impact of crystal defects. Optical transmissivity, electrical conductivity, and thickness are three important parameters of the buffer layer. Currently, common buffer layers include SiGe , GaAsP , AlGaAs , and GaAs . Compared to low-bandgap SiGe buffers and wide bandgap GaAsP would be a better choice to minimize the dislocation . However, the decrease of threading dislocations density to 107 cm−2 is still deleterious for minority carrier devices, so this approach remains challenging.
So far, the highest reported AM1.5G efficiency of 22.3% is achieved by Feifel and coworkers  for a GaInP/GaAs/Si TSC in 2019. The 200 nm thick GaN0.024P0.976 deposited on the GaP nucleation layer to reduce defect density at the GaP/Si interface. Then, the temperature was increased to 700 °C during the GaAsyP1−y buffer growth. Due to the high temperature, silicon bottom solar cells are mainly silicon homojunction solar cells. Due to stable and high PCE, it is expected that new SHJ solar cells will replace silicon homojunction solar cells.
5.2 Wafer bonding
Wafer bonding approach seems to be the most successful approach for fabricating high-performance tandem device with relatively larger size. Compared with heteroepitaxial, wafer bonding is not limited by the lattice matching and thermal expansion coefficients. The defect network required for lattice mismatch is confined at bonded interfaces . This process was originally developed by Lasky in 1985s, and reexamined by Cariou, Shigekawa, and many other research groups to fabricate Si/Si  Ge/Ge , Si/GaAs , AlGaAs-Si , Si/InP , and Si/InGaP  junctions. A typical wafer bonding process starts with wafer preprocessing by a chemical solution or plasma, followed by mechanical compression, and ends with postannealing of wafers to form strong bonds. The wafer bonding is mainly induced and dictated by hydrogen bridge bonds or van der Waals interactions, which are two orders of magnitude weaker than covalent bonds. In this sense, the electrically conductive bond layers should have very low surface roughness (<0.2 nm) and must be free from void . As a major requirement, there should be a good chemical mechanical polishing step capable of forming two mirror surfaces, ideally they should be able to form compact bonding layer with higher bonding energy.
At present, wet-chemical etching and surface activated bonding (SAB) are two main methods of wafer bonding. For the former, a hydrogen-terminated surface is generated by wet-chemical etching. After compression, two hydrophobic surfaces are connected by weak hydrogen-bridge bonds . Finally, hydrogen is desorbed to form covalent bonds at the interface by postannealing (>800 °C). Another technique is the SAB in which the surfaces are activated by plasma or a fast argon atom or highly energetic ion-beam . Postannealing is not required because it lowers the bonding energy. All major steps involved in a SAB process are schematically illustrated in Figure 7(b), with the fabrication of GaInP/GaAs/Si TSCs as an example. Based on the schematic diagram of GaInP/GaAs/Si tandem device, Si, and GaAs wafers formed tandem structure when they were loaded on top and bottom electrodes into a bond chamber . An argon fast atom beam was employed to remove the surface oxide, and no re oxidation occurred in the bonding layer under high vacuum ambient (<3 × 10−8 mbar). The wafers were brought in contact under the force of 10 kN for a few minutes. The GaAs growth substrate was then etched away by H2O2 and NH4OH solution. In order to obtain a good bonding mechanical quality, nearly perfect surface must be bonded together with low surface roughness and small particle pollution. Cariou et al.  have demonstrated a 2T GaInP/GaAs//Si solar cell with AlGaAs/GaAs tunnel junction reaching a notable PCE of 33.3% (1-sun, AM1.5G) (Figure 7(d)). Due to a diffractive crossed grating made of a polymeric resist at the rear side of TOPcon bottom cell, internal light trapping was enhanced and the device was close to current matching (Figure 7(e)) . Furthermore, Lackner et al.  used Ga0.51In0.49P/Al0.06Ga0.94As (1.91eV/1.51eV) top cell to replace the former GaInP/GaAs (1.92/1.42 eV), yielding an increase in 50 mV for Voc and 2.6% for FF (Figure 7(c)).
One of the challenges faced by SAB is to select the appropriate tunnel junction . As shown in Figure 8(a), there are two types of tunneling mechanisms in the tunnel junction: direct transition (band to band) and trap assisted tunneling (TAT). Direct transition even takes place in intrinsic materials where no trap exists. According to quantum mechanical principles, if energy and momentum are conserved, charge carriers can pass through potential barriers. However, after reaching the characteristic peak of tunnel current density, the current drops sharply (Figure 8(b)). This property can be attributed to occupied states of the n-side which are aligned longer with empty states on the p-side. Similar to buffer layer, optical transmissivity, electrical conductivity, and thickness are three key parameters of tunnel junctions. The AlGaAs/GaAs tunnel junction is widely used in III–V/Si TSC. The major advantage of AlGaAs/GaAs tunnel junction is that AlGaAs could be more easily doped with carbon than GaAs. Besides, a high aluminum content layer has a diffusion suppressing effect . In contrast, GaAs/GaAs tunnel junction is limited by the optical absorption due to the Moss Burstein effect of p-type layer . AlGaAs/GaInP tunnel junction is more difficult to fabricate due to tellurium segregation on the growing surface during high doping .
Another challenge faced by SAB is to select the appropriate heterointerface to form an ideal interface resistance also. So far, different heterointerfaces have been extensively studied, such as p+-GaAs/p+-Si , p+-GaAs/n+-Si , n+-GaAs/n+-Si . For instance, Liang et al. have demonstrated that the interface resistances of p+-GaAs/n+-Si, p+-GaAs/p+-Si, n+-GaAs/n+ Si, n+-GaAs/p+-Si were 0.196, 1.887, 0.112 and 0.132 Ω cm2, respectively (Figure 8(d)) . In essence, different polarity and doping concentration play an important role in the value of interfacial resistances. Using this information, one can obtain ideal interface resistances as small as a few milliohms by optimizing the thickness and doping concentration in the proper ranges. For example, Essig et al. proved that the lowest interface resistance of n-Si and n-GaAs could be reduced to 2.5 × 10−3 Ω·cm2 by adjusting the carrier concentration at the surface. However, fast atom beam (FAB) can cause damage to the surfaces of bonded substrates. Shigekawa et al.  attempted to insert an ITO layer between Si and GaAs to reduce the damage. They found that the interface resistances of p+-GaAs/ITO/n+-Si and n+-GaAs/ITO/n+-Si were 0.61 and 0.12 Ω·cm2 (Figure 8(c)), respectively. The difference of the interface resistances was likely due to a potential barrier generated by the midgap states at the bonding interfaces. Noting that InGaP/GaAs/Si triple-junction (3J) solar cells with an n+-GaAs/ITO/n+-Si showed an efficiency of 18% with a Voc of 2.69 V, a JSC of 10.6 mA ⋅ cm−2 and an FF of 84.1%.
5.3 Mechanical stacking
Mechanical stacking integration represents a straightforward route to III–V/Si tandem device. In this approach, a glass slide with an III–V top cell is attached to the bottom cell SHJ solar cell with a transparent and insulating adhesive. Epoxy has been successfully demonstrated by several groups, with notable examples including 4T GaInP/GaAs/Si , GaAs/Si , and GaInP/Si ,  tandem devices. It is generally accepted that mechanical stacking provides a simple, high-performance procedure that also allows the variety of top-cell materials present on the silicon bottom cell, to be stacked in a single step. For example, Essig et al.  fabricated mechanically stacked 4T GaInP/Si tandem device using a transparent epoxy between the subcells. A PCE of 27.08% is achieved by optimizing the long-wavelength light and luminescent coupling between the two junctions. Moreover, Essig et al.  have demonstrated that by mechanical stacking GaInP/GaAs top cell on SHJ bottom cells, the PCE of tandem device could reach to 35.9%.
Tables 3 and 4 show the evolution of the 2T and 4T III–V/Si TSCs technology efficiency, respectively. In addition to insulating adhesive, metal, and metal nanoparticle can serve as a class of transparent conductive adhesive for fabricating III–V/Si tandem device . In this regard, Pd nanoparticle array that could be conveniently patterned on the top of silicon bottom cell using a selfassembled template of polystyrene-block-poly(2-vinylpyridine) (PS-b-P2VP) to fabricate tandem device (Figure 8(e)) . Synthesis of these nanoparticle arrays with different materials, for instance, provides a powerful route to the fabrication of III–V/Si tandem device from various subcells. As shown by Mizuno et al. , Pd nanoparticle array as thin as 37 nm could be prepared by copolymer-templated fabrication. This leaded to a Pd nanoparticle density of 60 ng/cm2 and surface coverage ratio of 12%. The 3J InGaP/GaAs//Si tandem device with a Pd nanoparticle array showed a PCE of 25.1% with a Voc of 2.88 V, a JSC of 10.51 mA·cm−2 and an FF of 83% . Moreover, Makita et al. used InGaP/AlGaAs two junction cell and TOPCon bottom cell to replace the former InGaP/GaAs top cell and Al-BSF Si bottom cell, improving the Jsc from 10.51 to 12.72 mA·cm−2, resulting in a PCE of 30.3% for 2T III–V/Si TSC (Figure 8(f)). By using Al0.06Ga0.94As (1.50 eV) middle cell, the absorption edge decreased from 870 to 830 nm. Therefore, the current density of Si bottom cell increased by 2.71 mA·cm−2.
|Si bottom cell||Top cell||Area (cm2)||VOC (V)||Jsc(mA/cm2)||FF (%)||Eff (%)||Research institution||Year|
|Si bottom cell||Top cell||Area (cm2)||Voc (V)||Jsc (mA/cm2)||FF||Efficiency||Research institution||Year|
|Textured SHJ||GaInP/GaAs||1||2.520, 0.681||13.61, 22.03||87.5, 78.5||30.01 + 5.9 = 35.91||EPFL ||2017|
|POLO-IBC||GaInP/GaAs||1||2.535, 0.67||13.43, 10.4||87.9, 18.8||29.94 + 5.49 = 35.43||ISFH ||2017|
|Textured SHJ||GaAs||1||1.092, 0.683||28.9, 11.07||85.0, 79.2||26.83 + 5.99 = 32.82||EPFL ||2017|
|POLO-IBC||GaInP||1||1.43, 0.3687||15.4, 24.2||86.6, 75.1||19.1 + 12.5 = 31.5||ISFH ||2017|
|Textured SHJ||GaInP||1||1.456, 0.667||14.15, 22.7||87.9, 76.2||18.1 + 11.7 = 29.8||NREL ||2016|
|TOPcon||GaInP||1||1.444, 0.605||14.6, 23.3||87.9, 60.5||18.54 + 8.54 = 27.08||NREL ||2015|
6 Conclusion and outlook
This review provides an overview on different methods that have been developed for fabricating TSCs. We briefly discussed a series of methods for fabricating TSCs that are based on the upscaling of small area solar cells, with an emphasis on perovskite/SHJ and III–V/SHJ tandem device. The big challenge for perovskite/SHJ TSCs is the Voc loss. The main reason is nonradiative recombination at the grain boundary and interface. To address this issue, a range of strategies have been applied including increasing grain size, surface passivation, heterojunction engineering, and ion compensation. Defects passivation is the most powerful means to improve Voc. Intense studies are focusing on Lewis bases/acids, alkali metal ions (Na, K, Ru, Cs), ligand passivation, halogen ion (Cl, Br), PbI2, 2D perovskite, insulating polymers, and guanidinium-based additives. The band alignment of charge transporting layer also plays a key role for Voc improvement. Another issue is fabricating perovskite top cell on textured c-Si cell, vacuum deposition method may be a good option. To further improve the TSCs performance, optical and electron losses should be minimized. If these issues are solved, a PCE over 35% can be expected. For up-scaling fabrication and commercialization, long-term stability and material toxicity are also needed to be optimized.
As for III–V/Si TSCs, heteroepitaxial integration approach based on the control of dislocation densities at the interface are able to produce devices, but in poor quality. So far, approaches based on the SAB seem to be most versatile in fabricating high efficiency III–V/SHJ TSCs. This method, however, is not appropriate for using with textured SHJ solar cells, and the requirement of clean room environment and tunnel junction may also cause high cost. Therefore, mechanical stacking by transparent conductive adhesive and metal nanoparticle array may be a new research direction for high performance III–V/SHJ TSCs. Judged against high performance and low cost, all the approaches described in this review need to be improved before they can be widely used in commercial applications.
Funding source: the National Natural Science Foundation of China
Award Identifier / Grant number: Grant Nos.61674084
Funding source: Key R&D Program of Hebei Province
Award Identifier / Grant number: No. 19214301D
Funding source: National Key Research and Development Program of China
Award Identifier / Grant number: Grant No.2018YFB1500103
Funding source: Tianjin Science and Technology Project
Award Identifier / Grant number: Grant No. 18ZXJMTG00220
Funding source: Overseas Expertise Introduction Project for Discipline Innovation of Higher Education of China
Award Identifier / Grant number: Grant No. B16027
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: The authors gratefully acknowledge the supports from National Key Research and Development Program of China (Grant No. 2018YFB1500103), the National Natural Science Foundation of China (Grant Nos. 61674084), the Overseas Expertise Introduction Project for Discipline Innovation of Higher Education of China (Grant No. B16027), Tianjin Science and Technology Project (Grant No. 18ZXJMTG00220). Key R&D Program of Hebei Province (No. 19214301D).
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.
 K. Masuko, M. Shigematsu, T. Hashiguchi, et al., “Achievement of more than 25% conversion efficiency with crystalline silicon heterojunction solar cell,” IEEE J. Photovoltaics, vol. 4, no. 6, pp. 1433–1435, 2014. Search in Google Scholar
 J. Nakamura, N. Asano, T. Hieda, C. Okamoto, H. Katayama, and K. Nakamura, “Development of heterojunction back contact Si solar cells,” IEEE J. Photovoltaics, vol. 4, no. 6, pp. 1491–1495, 2014, https://doi.org/10.1109/jphotov.2014.2358377. Search in Google Scholar
 M. Taguchi, A. Yano, S. Tohoda, et al., “24.7% record efficiency HIT solar cell on thin silicon wafer,” IEEE J. Photovoltaics, vol. 4, no. 1, pp. 96–99, 2014. Search in Google Scholar
 D. Adachi, J. L. Hernández, and K. Yamamoto, “Impact of carrier recombination on fill factor for large area heterojunction crystalline silicon solar cell with 25.1% efficiency,” Appl. Phys. Lett., vol. 107, no. 23, p. 233506, 2015. Search in Google Scholar
 K. Yoshikawa, H. Kawasaki, W. Yoshida, et al., “Silicon heterojunction solar cell with interdigitated back contacts for a photoconversion efficiency over 26%,” Nat. Energy, vol. 2, no. 5, 2017. Search in Google Scholar
 K. Yoshikawa, W. Yoshida, T. Irie, et al., “Exceeding conversion efficiency of 26% by heterojunction interdigitated back contact solar cell with thin film Si technology,” Sol. Energy Mater. Sol. Cells, vol. 173, pp. 37–42, 2017. Search in Google Scholar
 J. B. Heng, J. Fu, B. Kong, et al., “23% high-efficiency tunnel oxide junction bifacial solar cell with electroplated Cu gridlines,” IEEE J. Photovoltaics, vol. 5, no. 1, pp. 82–86, 2015. Search in Google Scholar
 F. Feldmann, M. Simon, M. Bivour, C. Reichel, M. Hermle, and S. W. Glunz, “Efficient carrier-selective p- and n-contacts for Si solar cells,” Sol. Energy Mater. Sol. Cells, vol. 131, pp. 100–104, 2014, https://doi.org/10.1016/j.solmat.2014.05.039. Search in Google Scholar
 A. Richter, J. Benick, F. Feldmann, A. Fell, M. Hermle, and S. W. Glunz, “n-Type Si solar cells with passivating electron contact: identifying sources for efficiency limitations by wafer thickness and resistivity variation,” Sol. Energy Mater. Sol. Cells, vol. 173, pp. 96–105, 2017, https://doi.org/10.1016/j.solmat.2017.05.042. Search in Google Scholar
 F. Haase, C. Hollemann, S. Schäfer, et al., “Laser contact openings for local poly-Si-metal contacts enabling 26.1%-efficient POLO-IBC solar cells,” Sol. Energy Mater. Sol. Cells, vol. 186, pp. 184–193, 2018. Search in Google Scholar
 F. Haase, S. Schafer, C. Klamt, et al., “Perimeter recombination in 25%-efficient IBC solar cells with passivating POLO contacts for both polarities,” IEEE J. Photovoltaics, vol. 8, no. 1, pp. 23–29, 2018. Search in Google Scholar
 K. Ding, U. Aeberhard, F. Finger, and U. Rau, “Silicon heterojunction solar cell with amorphous silicon oxide buffer and microcrystalline silicon oxide contact layers,” Phys. Status Solidi Rapid Res. Lett., vol. 6, no. 5, pp. 193–195, 2012, https://doi.org/10.1002/pssr.201206030. Search in Google Scholar
 K. Ding, U. Aeberhard, F. Finger, and U. Rau, “Optimized amorphous silicon oxide buffer layers for silicon heterojunction solar cells with microcrystalline silicon oxide contact layers,” J. Appl. Phys., vol. 113, no. 13, p. 134501, 2013, https://doi.org/10.1063/1.4798603. Search in Google Scholar
 K. Ding, U. Aeberhard, V. Smirnov, B. Holländer, F. Finger, and U. Rau, “Wide gap microcrystalline silicon oxide emitter for a-SiOx:H/c-Si heterojunction solar cells,” Jpn. J. Appl. Phys., vol. 52, no. 12R, p. 122304, 2013, https://doi.org/10.7567/jjap.52.122304. Search in Google Scholar
 A. Richter, F. Lentz, M. Meier, F. Finger, and K. Ding, “Light management in planar silicon heterojunction solar cells via nanocrystalline silicon oxide films and nano-imprint textures,” Phys. Status Solidi, vol. 213, no. 7, pp. 1976–1982, 2016, https://doi.org/10.1002/pssa.201533024. Search in Google Scholar
 M. Pomaska, A. Richter, F. Lentz, et al., “Wide gap microcrystalline silicon carbide emitter for amorphous silicon oxide passivated heterojunction solar cells,” Jpn. J. Appl. Phys., vol. 56, no. 2, p. 022302, 2017, https://doi.org/10.7567/jjap.56.022302. Search in Google Scholar
 T. Krajangsang, S. Inthisang, J. Sritharathikhun, et al., “An intrinsic amorphous silicon oxide and amorphous silicon stack passivation layer for crystalline silicon heterojunction solar cells,” Thin Solid Films, vol. 628, pp. 107–111, 2017, https://doi.org/10.1016/j.tsf.2017.03.010. Search in Google Scholar
 C. Battaglia, S. M. de Nicolás, S. De Wolf, et al., “Silicon heterojunction solar cell with passivated hole selective MoOx contact,” Appl. Phys. Lett., vol. 104, no. 11, p. 113902, 2014, https://doi.org/10.1063/1.4868880. Search in Google Scholar
 J. Geissbühler, J. Werner, S. Martin de Nicolas, et al., “22.5% efficient silicon heterojunction solar cell with molybdenum oxide hole collector,” Appl. Phys. Lett., vol. 107, no. 8, p. 081601, 2015. Search in Google Scholar
 A. Richter, M. Hermle, and S. W. Glunz, “Reassessment of the limiting efficiency for crystalline silicon solar cells,” IEEE J. Photovoltaics, vol. 3, no. 4, pp. 1184–1191, 2013, https://doi.org/10.1109/jphotov.2013.2270351. Search in Google Scholar
 I. Almansouri, A. Ho-Baillie, S. P. Bremner, and M. A. Green, “Supercharging silicon solar cell performance by means of multijunction concept,” IEEE J. Photovoltaics, vol. 5, no. 3, pp. 968–976, 2015, https://doi.org/10.1109/jphotov.2015.2395140. Search in Google Scholar
 M. Yamaguchi, K.-H. Lee, K. Araki, and N. Kojima, “A review of recent progress in heterogeneous silicon tandem solar cells,” J. Phys. D Appl. Phys., vol. 51, no. 13, p. 133002, 2018, https://doi.org/10.1088/1361-6463/aaaf08. Search in Google Scholar
 S. Essig, C. Allebé, T. Remo, et al., “Raising the one-sun conversion efficiency of III–V/Si solar cells to 32.8% for two junctions and 35.9% for three junctions,” Nat. Energy, vol. 2, no. 9, 2017. Search in Google Scholar
 M. Green, E. Dunlop, J. Hohl‐Ebinger, M. Yoshita, N. Kopidakis, and X. Hao, “Solar cell efficiency tables (version 57),” Prog. Photovoltaics Res. Appl., vol. 29, no. 1, pp. 3–15, 2020, https://doi.org/10.1002/pip.3371. Search in Google Scholar
 T. Leijtens, K. A. Bush, R. Prasanna, and M. D. McGehee, “Opportunities and challenges for tandem solar cells using metal halide perovskite semiconductors,” Nat. Energy, vol. 3, no. 10, pp. 828–838, 2018, https://doi.org/10.1038/s41560-018-0190-4. Search in Google Scholar
 J. Werner, B. Niesen, and C. Ballif, “Perovskite/Silicon tandem solar cells: marriage of convenience or true love story? – An overview,” Adv. Mater. Interfaces, vol. 5, no. 1, p. 1700731, 2017, https://doi.org/10.1002/admi.201700731. Search in Google Scholar
 I. J. Park, J. H. Park, S. G. Ji, M.-A. Park, J. H. Jang, and J. Y. Kim, “A three-terminal monolithic perovskite/Si tandem solar cell characterization platform,” Joule, vol. 3, no. 3, pp. 807–818, 2019, https://doi.org/10.1016/j.joule.2018.11.017. Search in Google Scholar
 F. Gota, M. Langenhorst, R. Schmager, J. Lehr, and U. W. Paetzold, “Energy yield advantages of three-terminal perovskite-silicon tandem photovoltaics,” Joule, vol. 4, no. 11, pp. 2387–2403, 2020, https://doi.org/10.1016/j.joule.2020.08.021. Search in Google Scholar
 T. Tayagaki, K. Makita, T. Tachibana, et al., “Three-Terminal tandem solar cells with a back-contact-type bottom cell bonded using conductive metal nanoparticle arrays,” IEEE J. Photovoltaics, vol. 10, no. 2, pp. 358–362, 2020, https://doi.org/10.1109/jphotov.2019.2957661. Search in Google Scholar
 P. Tockhorn, P. Wagner, L. Kegelmann, et al., “Three-Terminal perovskite/silicon tandem solar cells with top and interdigitated rear contacts,” ACS Appl. Energy Mater., vol. 3, no. 2, pp. 1381–1392, 2020, https://doi.org/10.1021/acsaem.9b01800. Search in Google Scholar
 M. Schnabel, H. Schulte-Huxel, M. Rienäcker, et al., “Three-terminal III–V/Si tandem solar cells enabled by a transparent conductive adhesive,” Sustain. Energy Fuels, vol. 4, no. 2, pp. 549–558, 2020, https://doi.org/10.1039/c9se00893d. Search in Google Scholar
 H. Li and W. Zhang, “Perovskite tandem solar cells: from fundamentals to commercial deployment,” Chem. Rev., vol. 120, no. 18, pp. 9835–9950, 2020, https://doi.org/10.1021/acs.chemrev.9b00780. Search in Google Scholar
 A. Polman, M. Knight, E. C. Garnett, B. Ehrler, and W. C. Sinke, “Photovoltaic materials: present efficiencies and future challenges,” Science, vol. 352, no. 6283, p. aad4424, 2016, https://doi.org/10.1126/science.aad4424. Search in Google Scholar
 D. Luo, R. Su, W. Zhang, Q. Gong, and R. Zhu, “Minimizing non-radiative recombination losses in perovskite solar cells,” Nat. Rev. Mater., vol. 5, no. 1, pp. 44–60, 2019, https://doi.org/10.1038/s41578-019-0151-y. Search in Google Scholar
 S. P. Philipps, F. Dimroth and A. W. Bett, High-efficiency III–V Multijunction Solar Cells, 2018, pp. 439–472, https://doi.org/10.1016/b978-0-12-809921-6.00012-4. Search in Google Scholar
 C. Chen, Z. Song, C. Xiao, et al., “Achieving a high open-circuit voltage in inverted wide-bandgap perovskite solar cells with a graded perovskite homojunction,” Nanomater. Energy, vol. 61, pp. 141–147, 2019, https://doi.org/10.1016/j.nanoen.2019.04.069. Search in Google Scholar
 M. Saliba, T. Matsui, K. Domanski, et al., “Incorporation of rubidium cations into perovskite solar cells improves photovoltaic performance,” Science, vol. 354, no. 6309, pp. 206–209, 2016, https://doi.org/10.1126/science.aah5557. Search in Google Scholar
 M. Lozac’h, S. Nunomura, and K. Matsubara, “Double-sided TOPCon solar cells on textured wafer with ALD SiOx layer,” Sol. Energy Mater. Sol. Cells, vol. 207, p. 110357, 2020, https://doi.org/10.1016/j.solmat.2019.110357. Search in Google Scholar
 J. Bullock, Y. Wan, Z. Xu, et al., “Stable dopant-free asymmetric heterocontact silicon solar cells with efficiencies above 20%,” ACS Energy Lett., vol. 3, no. 3, pp. 508–513, 2018. Search in Google Scholar
 C. Battaglia, A. Cuevas, and S. De Wolf, “High-efficiency crystalline silicon solar cells: status and perspectives,” Energy Environ. Sci., vol. 9, no. 5, pp. 1552–1576, 2016, https://doi.org/10.1039/c5ee03380b. Search in Google Scholar
 L. Mazzarella, Y. H. Lin, S. Kirner, et al., “Infrared light management using a nanocrystalline silicon oxide interlayer in monolithic perovskite/silicon heterojunction tandem solar cells with efficiency above 25%,” Adv. Energy Mater., vol. 9, no. 14, p. 1803241, 2019. Search in Google Scholar
 J. Bullock, M. Hettick, J. Geissbühler, et al., “Efficient silicon solar cells with dopant-free asymmetric heterocontacts,” Nat. Energy, vol. 1, no. 3, 2016, https://doi.org/10.1038/nenergy.2015.31. Search in Google Scholar
 X. Yang, K. Weber, Z. Hameiri, and S. De Wolf, “Industrially feasible, dopant-free, carrier-selective contacts for high-efficiency silicon solar cells,” Prog. Photovoltaics Res. Appl., vol. 25, no. 11, pp. 896–904, 2017, https://doi.org/10.1002/pip.2901. Search in Google Scholar
 R. Cariou, J. Benick, F. Feldmann, et al., “III–V-on-silicon solar cells reaching 33% photoconversion efficiency in two-terminal configuration,” Nat. Energy, vol. 3, no. 4, pp. 326–333, 2018. Search in Google Scholar
 K. Makita, H. Mizuno, T. Tayagaki, et al., “III‐V//Si multijunction solar cells with 30% efficiency using smart stack technology with Pd nanoparticle array,” Prog. Photovoltaics Res. Appl., vol. 28, no. 1, pp. 16–24, 2019. Search in Google Scholar
 D. Lackner, O. Höhn, R. Müller, et al., “Two‐Terminal direct wafer‐bonded GaInP/AlGaAs//Si triple‐junction solar cell with AM1.5g efficiency of 34.1%,” Sol. RRL, vol. 4, no. 9, p. 2000210, 2020. Search in Google Scholar
 A. Ingenito, G. Nogay, Q. Jeangros, et al., “A passivating contact for silicon solar cells formed during a single firing thermal annealing,” Nat. Energy, vol. 3, no. 9, pp. 800–808, 2018, https://doi.org/10.1038/s41560-018-0239-4. Search in Google Scholar
 M. Rienäcker, M. Schnabel, E. L. Warren, A. Merkle, and R. Peibst, “Mechanically stacked dual-junction and triple-junction III–V/Si-IBC cells with efficiencies of 31.5 % and 35.4 %,” in 33rd European Photovoltaic Solar Energy Conference and Exhibition, 2017. Search in Google Scholar
 A. Al-Ashouri, E. Khnen, B. Li, A. Magomedov, and S. Albrecht, “Monolithic perovskite/silicon tandem solar cell with >29% efficiency by enhanced hole extraction,” Science, vol. 370, no. 6522, 2020, https://doi.org/10.1126/science.abd4016. Search in Google Scholar
 E. Lamanna, F. Matteocci, E. Calabrò, et al., “Mechanically stacked, two-terminal graphene-based perovskite/silicon tandem solar cell with efficiency over 26%,” Joule, vol. 4, no. 4, pp. 865–881, 2020. Search in Google Scholar
 J. Xu, C. C. Boyd, Z. J. Yu, A. F. Palmstrom, and M. D. Mcgehee, “Triple-halide wide–band gap perovskites with suppressed phase segregation for efficient tandems,” Science, vol. 367, no. 6482, pp. 1097–1104, 2020, https://doi.org/10.1126/science.aaz5074. Search in Google Scholar
 B. Chen, Z. J. Yu, S. Manzoor, et al., “Blade-coated perovskites on textured silicon for 26%-efficient monolithic perovskite/silicon tandem solar cells,” Joule, vol. 4, no. 4, pp. 850–864, 2020. Search in Google Scholar
 D. Kim, H. J. Jung, I. J. Park, et al., “Efficient, stable silicon tandem cells enabled by anion-engineered wide-bandgap perovskites,” Science, vol. 368, no. 6487, pp. 155–160, 2020. https://doi.org/10.1126/science.aba3433. Search in Google Scholar
 E. Köhnen, M. Jošt, A. B. Morales-Vilches, et al., “Highly efficient monolithic perovskite silicon tandem solar cells: analyzing the influence of current mismatch on device performance,” Sustain. Energy Fuels, vol. 3, no. 8, pp. 1995–2005, 2019, https://doi.org/10.1039/c9se00120d. Search in Google Scholar
 Y. Hou, E. Aydin, M. D. Bastiani, et al., “Efficient tandem solar cells with solution-processed perovskite on textured crystalline silicon,” Science, vol. 367, no. 6482, pp. 1135–1140, 2020. Search in Google Scholar
 M. Jošt, E. Köhnen, A. B. Morales-Vilches, et al., “Textured interfaces in monolithic perovskite/silicon tandem solar cells: advanced light management for improved efficiency and energy yield,” Energy Environ. Sci., vol. 11, no. 12, pp. 3511–3523, 2018. Search in Google Scholar
 B. Chen, Z. Yu, K. Liu, et al., “Grain engineering for perovskite/silicon monolithic tandem solar cells with efficiency of 25.4%,” Joule, vol. 3, no. 1, pp. 177–190, 2019. Search in Google Scholar
 B. A. Kamino, B. Paviet-Salomon, S.-J. Moon, et al., “Low-Temperature screen-printed metallization for the scale-up of two-terminal perovskite–silicon tandems,” ACS Appl. Energy Mater., vol. 2, no. 5, pp. 3815–3821, 2019, https://doi.org/10.1021/acsaem.9b00502. Search in Google Scholar
 F. Sahli, J. Werner, B. A. Kamino, et al., “Fully textured monolithic perovskite/silicon tandem solar cells with 25.2% power conversion efficiency,” Nat. Mater., vol. 17, no. 9, pp. 820–826, 2018, https://doi.org/10.1038/s41563-018-0115-4. Search in Google Scholar
 K. A. Bush, S. Manzoor, K. Frohna, et al., “Minimizing current and voltage losses to reach 25% efficient monolithic two-terminal perovskite–silicon tandem solar cells,” ACS Energy Lett., vol. 3, no. 9, pp. 2173–2180, 2018. Search in Google Scholar
 S. Zhu, F. Hou, W. Huang, et al., “Solvent engineering to balance light absorbance and transmittance in perovskite for tandem solar cells,” Sol. RRL, vol. 2, no. 11, p. 1800176, 2018, https://doi.org/10.1002/solr.201800176. Search in Google Scholar
 F. Sahli, B. A. Kamino, J. Werner, et al., “Improved optics in monolithic perovskite/silicon tandem solar cells with a nanocrystalline silicon recombination junction,” Adv. Energy Mater., vol. 8, no. 6, p. 1701609, 2018, https://doi.org/10.1002/aenm.201701609. Search in Google Scholar
 K. A. Bush, A. F. Palmstrom, Z. J. Yu, et al., “23.6%-efficient monolithic perovskite/silicon tandem solar cells with improved stability,” Nat. Energy, vol. 2, no. 4, 2017. Search in Google Scholar
 J. Werner, L. Barraud, A. Walter, et al., “Efficient near-infrared-transparent perovskite solar cells enabling direct comparison of 4-terminal and monolithic perovskite/silicon tandem cells,” ACS Energy Lett., vol. 1, no. 2, pp. 474–480, 2016, https://doi.org/10.1021/acsenergylett.6b00254. Search in Google Scholar
 S. Albrecht, M. Saliba, J. P. Correa Baena, et al., “Monolithic perovskite/silicon-heterojunction tandem solar cells processed at low temperature,” Energy Environ. Sci., vol. 9, no. 1, pp. 81–88, 2016, https://doi.org/10.1039/c5ee02965a. Search in Google Scholar
 J. Werner, C. H. Weng, A. Walter, et al., “Efficient monolithic perovskite/silicon tandem solar cell with cell area >1 cm(2),” J. Phys. Chem. Lett., vol. 7, no. 1, pp. 161–6, 2016, https://doi.org/10.1021/acs.jpclett.5b02686. Search in Google Scholar
 B. Chen, S. W. Baek, Y. Hou, et al., “Enhanced optical path and electron diffusion length enable high-efficiency perovskite tandems,” Nat. Commun., vol. 11, no. 1, p. 1257, 2020, https://doi.org/10.1038/s41467-020-15077-3. Search in Google Scholar
 Z. Wang, X. Zhu, S. Zuo, et al., “27%‐Efficiency four‐terminal perovskite/silicon tandem solar cells by sandwiched gold nanomesh,” Adv. Funct. Mater., vol. 30, no. 4, p. 1908298, 2019, https://doi.org/10.1002/adfm.201908298. Search in Google Scholar
 A. Rohatgi, K. Zhu, J. Tong, et al., “26.7% efficient 4-terminal perovskite–silicon tandem solar cell composed of a high-performance semitransparent perovskite cell and a doped poly-Si/SiOx passivating contact silicon cell,” IEEE J. Photovoltaics, vol. 10, no. 2, pp. 417–422, 2020. Search in Google Scholar
 E. Aydin, M. De Bastiani, X. Yang, et al., “Zr‐Doped indium oxide (IZRO) transparent electrodes for perovskite‐based tandem solar cells,” Adv. Funct. Mater., vol. 29, no. 25, p. 1901741, 2019, https://doi.org/10.1002/adfm.201901741. Search in Google Scholar
 J. Peng, T. Duong, X. Zhou, et al., “Efficient indium-doped TiOxElectron transport layers for high-performance perovskite solar cells and perovskite-silicon tandems,” Adv. Energy Mater., vol. 7, no. 4, p. 1601768, 2017, https://doi.org/10.1002/aenm.201601768. Search in Google Scholar
 B. Chen, Y. Bai, Z. Yu, et al., “Efficient semitransparent perovskite solar cells for 23.0%-efficiency perovskite/silicon four-terminal tandem cells,” Adv. Energy Mater., vol. 6, no. 19, p. 1601128, 2016. Search in Google Scholar
 D. P. Mcmeekin, G. Sadoughi, W. Rehman, et al., “A mixed-cation lead mixed-halide perovskite absorber for tandem solar cells,” Science, vol. 351, no. 6269, pp. 151–155, 2016, https://doi.org/10.1126/science.aad5845. Search in Google Scholar
 J. Werner, G. Dubuis, A. Walter, et al., “Sputtered rear electrode with broadband transparency for perovskite solar cells,” Sol. Energy Mater. Sol. Cells, vol. 141, pp. 407–413, 2015, https://doi.org/10.1016/j.solmat.2015.06.024. Search in Google Scholar
 P. Loper, S. J. Moon, S. M. de Nicolas, et al., “Organic-inorganic halide perovskite/crystalline silicon four-terminal tandem solar cells,” Phys. Chem. Chem. Phys., vol. 17, no. 3, pp. 1619–29, 2015, https://doi.org/10.1039/c4cp03788j. Search in Google Scholar
 G. E. Eperon, S. D. Stranks, C. Menelaou, M. B. Johnston, L. M. Herz, and H. J. Snaith, “Formamidinium lead trihalide: A broadly tunable perovskite for efficient planar heterojunction solar cells,” Energy Environ. Sci., vol. 7, no. 3, p. 982, 2014, https://doi.org/10.1039/c3ee43822h. Search in Google Scholar
 J. W. Lee, D. J. Seol, A. N. Cho, and N. G. Park, “High-efficiency perovskite solar cells based on the black polymorph of HC(NH2)2 PbI3,” Adv. Mater., vol. 26, no. 29, pp. 4991–8, 2014, https://doi.org/10.1002/adma.201401137. Search in Google Scholar
 D. Bi, C. Yi, J. Luo, et al., “Polymer-templated nucleation and crystal growth of perovskite films for solar cells with efficiency greater than 21%,” Nat. Energy, vol. 1, no. 10, 2016. Search in Google Scholar
 T. M. Koh, K. Fu, Y. Fang, et al., “Formamidinium-Containing metal-halide: An alternative material for near-IR absorption perovskite solar cells,” J. Phys. Chem. C, vol. 118, no. 30, pp. 16458–16462, 2013, https://doi.org/10.1021/jp411112k. Search in Google Scholar
 Z. Li, M. Yang, J.-S. Park, S.-H. Wei, J. J. Berry, and K. Zhu, “Stabilizing perovskite structures by tuning tolerance factor: formation of formamidinium and cesium lead iodide solid-state alloys,” Chem. Mater., vol. 28, no. 1, pp. 284–292, 2015, https://doi.org/10.1021/acs.chemmater.5b04107. Search in Google Scholar
 B. Conings, A. Babayigit, M. T. Klug, et al., “A universal deposition protocol for planar heterojunction solar cells with high efficiency based on hybrid lead halide perovskite families,” Adv. Mater., vol. 28, no. 48, pp. 10701–10709, 2016, https://doi.org/10.1002/adma.201603747. Search in Google Scholar
 M. Saliba, T. Matsui, J. Y. Seo, et al., “Cesium-containing triple cation perovskite solar cells: improved stability, reproducibility and high efficiency,” Energy Environ. Sci., vol. 9, no. 6, pp. 1989–1997, 2016, https://doi.org/10.1039/c5ee03874j. Search in Google Scholar
 C. Yi, J. Luo, S. Meloni, et al., “Entropic stabilization of mixed A-cation ABX3 metal halide perovskites for high performance perovskite solar cells,” Energy Environ. Sci., vol. 9, no. 2, pp. 656–662, 2016, https://doi.org/10.1039/c5ee03255e. Search in Google Scholar
 D. Bi, W. Tress, M. I. Dar, P. Gao, and A. Hagfeldt, “Efficient luminescent solar cells based on tailored mixed-cation perovskites,” Sci. Adv., vol. 2, no. 1, pp. e1501170, 2016, https://doi.org/10.1126/sciadv.1501170. Search in Google Scholar
 A. R. Yusoff, H. P. Kim, X. Li, J. Kim, J. Jang, and M. K. Nazeeruddin, “Ambipolar triple cation perovskite field effect transistors and inverters,” Adv. Mater., vol. 29, no. 8, 2017, https://doi.org/10.1002/adma.201602940. Search in Google Scholar
 Y. Ogomi, A. Morita, S. Tsukamoto, et al., “CH3NH3SnxPb(1-x)I3 perovskite solar cells covering up to 1060 nm,” J. Phys. Chem. Lett., vol. 5, no. 6, pp. 1004–11, 2014, https://doi.org/10.1021/jz5002117. Search in Google Scholar
 Q. Chen, N. De Marco, Y. Yang, et al., “Under the spotlight: the organic–inorganic hybrid halide perovskite for optoelectronic applications,” Nano Today, vol. 10, no. 3, pp. 355–396, 2015, https://doi.org/10.1016/j.nantod.2015.04.009. Search in Google Scholar
 H. J. Snaith, “Present status and future prospects of perovskite photovoltaics,” Nat. Mater., vol. 17, no. 5, pp. 372–376, 2018, https://doi.org/10.1038/s41563-018-0071-z. Search in Google Scholar
 J. P. Mailoa, C. D. Bailie, E. C. Johlin, et al., “A 2-terminal perovskite/silicon multijunction solar cell enabled by a silicon tunnel junction,” Appl. Phys. Lett., vol. 106, no. 12, p. 121105, 2015, https://doi.org/10.1063/1.4914179. Search in Google Scholar
 M. Jaysankar, W. Qiu, M. van Eerden, et al., “Four-Terminal perovskite/silicon multijunction solar modules,” Adv. Energy Mater., vol. 7, no. 15, p. 1602807, 2017, https://doi.org/10.1002/aenm.201602807. Search in Google Scholar
 C. D. Bailie, M. G. Christoforo, J. P. Mailoa, et al., “Semi-transparent perovskite solar cells for tandems with silicon and CIGS,” Energy Environ. Sci., vol. 8, no. 3, pp. 956–963, 2015, https://doi.org/10.1039/c4ee03322a. Search in Google Scholar
 F. Guo, H. Azimi, Y. Hou, et al., “High-performance semitransparent perovskite solar cells with solution-processed silver nanowires as top electrodes,” Nanoscale, vol. 7, no. 5, pp. 1642–9, 2015, https://doi.org/10.1039/c4nr06033d. Search in Google Scholar
 H. Kanda, A. Uzum, H. Nishino, et al., “Interface optoelectronics engineering for mechanically stacked tandem solar cells based on perovskite and silicon,” ACS Appl. Mater. Interfaces, vol. 8, no. 49, pp. 33553–33561, 2016, https://doi.org/10.1021/acsami.6b07781. Search in Google Scholar
 P. You, Z. Liu, Q. Tai, S. Liu, and F. Yan, “Efficient semitransparent perovskite solar cells with graphene electrodes,” Adv. Mater., vol. 27, no. 24, pp. 3632–8, 2015, https://doi.org/10.1002/adma.201501145. Search in Google Scholar
 T. Duong, N. Lal, D. Grant, et al., “Semitransparent perovskite solar cell with sputtered front and rear electrodes for a four-terminal tandem,” IEEE J. Photovoltaics, vol. 6, no. 3, pp. 679–687, 2016, https://doi.org/10.1109/jphotov.2016.2521479. Search in Google Scholar
 L. Felix, G. Marc A, A. Steve, et al., “In situ graphene doping as a route toward efficient perovskite tandem solar cells,” Phys. Status Solidi, vol. 213, no. 7, pp. 1989–1996, 2016. https://doi.org/10.1002/pssa.201532944. Search in Google Scholar
 Y. Yang, Q. Chen, Y. T. Hsieh, et al., “Multilayer transparent top electrode for solution processed perovskite/Cu(in,Ga)(Se,S)2 four terminal tandem solar cells,” ACS Nano, vol. 9, no. 7, pp. 7714–21, 2015, https://doi.org/10.1021/acsnano.5b03189. Search in Google Scholar
 K. A. Bush, C. D. Bailie, Y. Chen, et al., “Thermal and environmental stability of semi-transparent perovskite solar cells for tandems enabled by a solution-processed nanoparticle buffer layer and sputtered ITO electrode,” Adv. Mater., vol. 28, no. 20, pp. 3937–43, 2016, https://doi.org/10.1002/adma.201505279. Search in Google Scholar
 F. Fu, T. Feurer, T. P. Weiss, et al., “High-efficiency inverted semi-transparent planar perovskite solar cells in substrate configuration,” Nat. Energy, vol. 2, no. 1, 2016, https://doi.org/10.1038/nenergy.2016.190. Search in Google Scholar
 L. Xiong, Y. Guo, J. Wen, et al., “Review on the application of SnO2 in perovskite solar cells,” Adv. Funct. Mater., vol. 28, no. 35, p. 1802757, 2018, https://doi.org/10.1002/adfm.201802757. Search in Google Scholar
 L. Calió, S. Kazim, P. M. Grtzel, and S. Ahmad, “Hole-transport materials for perovskite solar cells,” Angew. Chem. Int. Ed., vol. 55, no. 47, pp. 14522–14545, 2016. https://doi.org/10.1002/anie.201601757. Search in Google Scholar
 F. Jiang, T. Liu, B. Luo, et al., “A two-terminal perovskite/perovskite tandem solar cell,” J. Mater. Chem., vol. 4, no. 4, pp. 1208–1213, 2016, https://doi.org/10.1039/c5ta08744a. Search in Google Scholar
 L. Zheng, J. Wang, Y. Xuan, et al., “A perovskite/silicon hybrid system with a solar-to-electric power conversion efficiency of 25.5%,” J. Mater. Chem., vol. 7, no. 46, pp. 26479–26489, 2019. Search in Google Scholar
 Y. Bai, X. Meng, and S. Yang, “Interface engineering for highly efficient and stable planar p-i-n perovskite solar cells,” Adv. Energy Mater., vol. 8, no. 5, p. 1701883, 2018, https://doi.org/10.1002/aenm.201701883. Search in Google Scholar
 M. Stolterfoht, P. Caprioglio, C. M. Wolff, et al., “The impact of energy alignment and interfacial recombination on the internal and external open-circuit voltage of perovskite solar cells,” Energy Environ. Sci., vol. 12, no. 9, pp. 2778–2788, 2019, https://doi.org/10.1039/c9ee02020a. Search in Google Scholar
 Z. Zhang, Z. Li, L. Meng, S. Y. Lien, and P. Gao, “Perovskite‐Based tandem solar cells: get the most out of the sun,” Adv. Funct. Mater., vol. 30, no. 38, p. 2001904, 2020, https://doi.org/10.1002/adfm.202001904. Search in Google Scholar
 G. E. Eperon, T. Leijtens, K. A. Bush, et al., “Perovskite-perovskite tandem photovoltaics with optimized bandgaps,” Science, vol. 354, no. 6314, pp. 861–865, 2016. https://doi.org/10.1126/science.aaf9717. Search in Google Scholar
 F. Fu, T. Feurer, T. Jager, et al., “Low-temperature-processed efficient semi-transparent planar perovskite solar cells for bifacial and tandem applications,” Nat. Commun., vol. 6, p. 8932, 2015, https://doi.org/10.1038/ncomms9932. Search in Google Scholar
 J. Urieta-Mora, I. Garcia-Benito, A. Molina-Ontoria, and N. Martin, “Hole transporting materials for perovskite solar cells: A chemical approach,” Chem. Soc. Rev., vol. 47, no. 23, pp. 8541–8571, 2018, https://doi.org/10.1039/c8cs00262b. Search in Google Scholar
 Y. Kim, E. H. Jung, G. Kim, D. Kim, B. J. Kim, and J. Seo, “Sequentially fluorinated PTAA polymers for enhancing VOC of high-performance perovskite solar cells,” Adv. Energy Mater., vol. 8, no. 29, p. 1801668, 2018, https://doi.org/10.1002/aenm.201801668. Search in Google Scholar
 A. G. Boldyreva, I. S. Zhidkov, S. Tsarev, et al., “Unraveling the impact of hole transport materials on photostability of perovskite films and p-i-n solar cells,” ACS Appl. Mater. Interfaces, vol. 12, no. 16, pp. 19161–19173, 2020, https://doi.org/10.1021/acsami.0c01027. Search in Google Scholar
 Q. Jiang, X. Zhang, and J. You, “SnO2 : A wonderful electron transport layer for perovskite solar cells,” Small, vol. 14, no. 31, p. e1801154, 2018. https://doi.org/10.1002/smll.201801154. Search in Google Scholar
 A. A. Said, J. Xie, and Q. Zhang, “Recent progress in organic electron transport materials in inverted perovskite solar cells,” Small, vol. 15, no. 27, p. e1900854, 2019, https://doi.org/10.1002/smll.201900854. Search in Google Scholar
 S. Gharibzadeh, B. Abdollahi Nejand, M. Jakoby, et al., “Record open‐circuit voltage wide‐bandgap perovskite solar cells utilizing 2D/3D perovskite heterostructure,” Adv. Energy Mater., vol. 9, no. 21, p. 1803699, 2019, https://doi.org/10.1002/aenm.201803699. Search in Google Scholar
 L. Deying, Y. Wenqiang, W. Zhiping, et al., “Enhanced photovoltage for inverted planar heterojunction perovskite solar cells,” Science, vol. 360, no. 6396, pp. 1442–1446, 2018. Search in Google Scholar
 M. Saliba, J. P. Correa-Baena, M. Gratzel, A. Hagfeldt, and A. Abate, “Perovskite solar cells: from the atomic level to film quality and device performance,” Angew Chem. Int. Ed. Engl., vol. 57, no. 10, pp. 2554–2569, 2018, https://doi.org/10.1002/anie.201703226. Search in Google Scholar
 Y. S. Kwon, J. Lim, H.-J. Yun, Y.-H. Kim, and T. Park, “A diketopyrrolopyrrole-containing hole transporting conjugated polymer for use in efficient stable organic–inorganic hybrid solar cells based on a perovskite,” Energy Environ. Sci., vol. 7, no. 4, p. 1454, 2014, https://doi.org/10.1039/c3ee44174a. Search in Google Scholar
 B. Philippe, B.-W. Park, R. Lindblad, et al., “Chemical and electronic structure characterization of lead halide perovskites and stability behavior under different exposures—a photoelectron spectroscopy investigation,” Chem. Mater., vol. 27, no. 5, pp. 1720–1731, 2015, https://doi.org/10.1021/acs.chemmater.5b00348. Search in Google Scholar
 N. Aristidou, I. Sanchez‐MolinaDr., T. Chotchuangchutchaval, et al., “The role of oxygen in the degradation of methylammonium lead trihalide perovskite photoactive layers,” Angew. Chem., vol. 54, no. 28, pp. 8208–8212, 2015. https://doi.org/10.1002/anie.201503153. Search in Google Scholar
 B. Conings, J. Drijkoningen, N. Gauquelin, et al., “Intrinsic thermal instability of methylammonium lead trihalide perovskite,” Adv. Energy Mater., vol. 5, no. 15, p. 1500477, 2015, https://doi.org/10.1002/aenm.201500477. Search in Google Scholar
 S. R. Raga, M.-C. Jung, M. V. Lee, M. R. Leyden, Y. Kato, and Y. Qi, “Influence of air annealing on high efficiency planar structure perovskite solar cells,” Chem. Mater., vol. 27, no. 5, pp. 1597–1603, 2015, https://doi.org/10.1021/cm5041997. Search in Google Scholar
 Z. Wang, D. P. McMeekin, N. Sakai, et al., “Efficient and air-stable mixed-cation lead mixed-halide perovskite solar cells with n-doped organic electron extraction layers,” Adv. Mater., vol. 29, no. 5, 2017, https://doi.org/10.1002/adma.201604186. Search in Google Scholar
 S. Yang, W. Fu, Z. Zhang, H. Chen, and C.-Z. Li, “Recent advances in perovskite solar cells: efficiency, stability and lead-free perovskite,” J. Mater. Chem., vol. 5, no. 23, pp. 11462–11482, 2017, https://doi.org/10.1039/c7ta00366h. Search in Google Scholar
 J. J. Yoo, S. Wieghold, M. C. Sponseller, et al., “An interface stabilized perovskite solar cell with high stabilized efficiency and low voltage loss,” Energy Environ. Sci., vol. 12, no. 7, pp. 2192–2199, 2019, https://doi.org/10.1039/c9ee00751b. Search in Google Scholar
 R. Cariou, J. Benick, M. Hermle, D. Lackner, and F. Dimroth, “Development of highly-efficient III–V//Si wafer-bonded triple-junction solar cells,” in 43rd IEEE Photovoltaic Specialists Conference (PVSC), IEEE, 2016. Search in Google Scholar
 H. Shen, D. Walter, Y. Wu, et al., “Monolithic perovskite/Si tandem solar cells: pathways to over 30% efficiency,” Adv. Energy Mater., vol. 10, no. 13, p. 1902840, 2020. Search in Google Scholar
 D. H. Kim, C. P. Muzzillo, J. Tong, et al., “Bimolecular additives improve wide-band-gap perovskites for efficient tandem solar cells with CIGS,” Joule, vol. 3, no. 7, pp. 1734–1745, 2019, https://doi.org/10.1016/j.joule.2019.04.012. Search in Google Scholar
 J. Rijcken, P. H. M. Bovendeerd, A. J. G. Schoofs, D. H. V. Campen, and T. Jimbo, “MOCVD growth of high efficiency current-matched AlGaAs/Si tandem solar cell,” J. Cryst. Growth, vol. 174, no. 1, 1997. Search in Google Scholar
 T. J. Grassman, D. J. Chmielewski, S. D. Carnevale, J. A. Carlin, and S. A. Ringel, “GaAsP/Si dual-junction solar cells grown by MBE and MOCVD,” IEEE, pp. 326–331, 2017. Search in Google Scholar
 T. Soga, T. Jimbo, and M. Umeno, “TEM characterization of antiphase domain in GaP on Si grown by MOCVD,” J. Cryst. Growth, pp. 171–174, 1994. Search in Google Scholar
 R. Cariou, J. Benick, P. Beutel, et al., “Monolithic two-terminal III–V//Si triple-junction solar cells with 30.2% efficiency under 1-sun AM1.5g,” IEEE J. Photovoltaics, vol. 7, no. 1, pp. 367–373, 2017. Search in Google Scholar
 M. T. Currie, S. B. Samavedam, T. A. Langdo, C. W. Leitz, and E. A. Fitzgerald, “Controlling threading dislocation densities in Ge on Si using graded SiGe layers and chemical-mechanical polishing,” Appl. Phys. Lett., vol. 72, no. 14, pp. 1718–1720, 1998, https://doi.org/10.1063/1.121162. Search in Google Scholar
 T. J. Grassman, J. A. Carlin, B. Galiana, et al., “Nucleation-related defect-free GaP/Si(100) heteroepitaxy via metal-organic chemical vapor deposition,” Appl. Phys. Lett., vol. 102, no. 14, p. 142102, 2013, https://doi.org/10.1063/1.4801498. Search in Google Scholar
 A. Freundlich, L. Lombez, M. Sugiyama, et al., “1.7eV Al0.2Ga0.8As solar cells epitaxially grown on silicon by SSMBE using a superlattice and dislocation filters,” Spie Opto, vol. 9743, p. 974310, 2016. Search in Google Scholar
 N. Jain and M. K. Hudait, “III–V multijunction solar cell integration with silicon: present status, challenges and future outlook,” Energy Harvest. Syst., vol. 1, pp. 3–4, 2014, https://doi.org/10.1515/ehs-2014-0012. Search in Google Scholar
 M. Feifel, D. Lackner, J. Ohlmann, J. Benick, M. Hermle, and F. Dimroth, “Direct growth of a GaInP/GaAs/Si triple‐junction solar cell with 22.3% AM1.5g efficiency,” Sol. RRL, vol. 3, no. 12, p. 1900313, 2019. Search in Google Scholar
 K. Petersen, P. Barth, J. Poydock, J. Brown, and J. Bryzek, “Silicon fusion bonding for pressure sensors,” in Solid-state Sensor and Actuator Workshop, Hilton Head, Technical Digest, IEEE, 1988. Search in Google Scholar
 K. Xiong, H. Mi, T.-H. Chang, et al., “AlGaAs/Si dual-junction tandem solar cells by epitaxial lift-off and print-transfer-assisted direct bonding,” Energy Sci. Eng., vol. 6, no. 1, pp. 47–55, 2018, https://doi.org/10.1002/ese3.182. Search in Google Scholar
 K. N. Yaung, M. Vaisman, J. Lang, and M. L. Lee, “GaAsP solar cells on GaP/Si with low threading dislocation density,” Appl. Phys. Lett., vol. 109, no. 3, p. 032107, 2016, https://doi.org/10.1063/1.4959825. Search in Google Scholar
 N. Shigekawa, T. Hara, T. Ogawa, et al., “GaAs/indium tin oxide/Si bonding junctions for III–V-on-Si hybrid multijunction cells with low series resistance,” IEEE J. Photovoltaics, vol. 8, no. 3, pp. 1–8, 2018. https://doi.org/10.1109/jphotov.2018.2802203. Search in Google Scholar
 P. Colter, B. Hagar, and S. Bedair, “Tunnel junctions for III–V multijunction solar cells review,” Crystals, vol. 8, no. 12, p. 445, 2018, https://doi.org/10.3390/cryst8120445. Search in Google Scholar
 M. Hermle, G. Létay, S. P. Philipps, and A. W. Bett, “Numerical simulation of tunnel diodes for multi‐junction solar cells,” Prog. Photovoltaics Res. Appl., vol. 16, no. 5, pp. 409–418, 2008, https://doi.org/10.1002/pip.824. Search in Google Scholar
 S. Wojtczuk, P. Chiu, X. Zhang, D. Pulver, C. Harris, and M. Timmons, “Bifacial growth InGaP/GaAs/InGaAs concentrator solar cells,” IEEE J. Photovoltaics, vol. 2, no. 3, pp. 371–376, 2012, https://doi.org/10.1109/jphotov.2012.2189369. Search in Google Scholar
 C. Ebert, Z. Pulwin, D. Byrnes, A. Paranjpe, and W. Zhang, “Tellurium doping of InGaP for tunnel junction applications in triple junction solar cells,” J. Cryst. Growth, vol. 315, no. 1, pp. 61–63, 2011, https://doi.org/10.1016/j.jcrysgro.2010.09.050. Search in Google Scholar
 J. Liang, S. Nishida, M. Morimoto, and N. Shigekawa, “Surface‐activating‐bonding‐based low‐resistance Si/III‐V junctions,” Electron. Lett., vol. 49, no. 13, pp. 830–832, 2013, https://doi.org/10.1049/el.2013.1553. Search in Google Scholar
 J. Liang, L. Chai, S. Nishida, M. Morimoto, and N. Shigekawa, “Investigation on the interface resistance of Si/GaAs heterojunctions fabricated by surface-activated bonding,” Jpn. J. Appl. Phys., vol. 54, no. 3, p. 030211, 2015, https://doi.org/10.7567/jjap.54.030211. Search in Google Scholar
 H. Mizuno, K. Makita, T. Tayagaki, T. Mochizuki, T. Sugaya, and H. Takato, “High-efficiency III–V//Si tandem solar cells enabled by the Pd nanoparticle array-mediated “smart stack” approach,” A, vol. 10, no. 7, p. 072301, 2017, https://doi.org/10.7567/apex.10.072301. Search in Google Scholar
 S. Essig, S. Ward, M. A. Steiner, et al., “Progress towards a 30% efficient GaInP/Si tandem solar cell,” Energy Procedia, vol. 77, pp. 464–469, 2015. Search in Google Scholar
 S. Essig, M. A. Steiner, C. Allebe, et al., “Realization of GaInP/Si dual-junction solar cells with 29.8% 1-sun efficiency,” IEEE J. Photovoltaics, vol. 6, no. 4, pp. 1012–1019, 2016. Search in Google Scholar
 E. Veinberg‐Vidal, L. Vauche, K. Medjoubi, et al., “Characterization of dual‐junction III‐V on Si tandem solar cells with 23.7% efficiency under low concentration,” Prog. Photovoltaics Res. Appl., vol. 27, no. 7, pp. 652–661, 2019. https://doi.org/10.1002/pip.3128. Search in Google Scholar
 M. Feifel, J. Ohlmann, J. Benick, et al., “Direct growth of III–V/silicon triple-junction solar cells with 19.7% efficiency,” IEEE J. Photovoltaics, vol. 8, no. 6, pp. 1590–1595, 2018. Search in Google Scholar
 T. Sugaya, T. Tayagaki, T. Aihara, et al., “Dual-junction GaAs solar cells and their application to smart stacked III–V//Si multijunction solar cells,” Appl. Phys. Express, vol. 11, no. 5, p. 052301, 2018, https://doi.org/10.7567/apex.11.052301. Search in Google Scholar
 H. Mizuno, K. Makita, T. Tayagaki, T. Mochizuki, and A. Ho-Baillie, “A “smart stack” triple-junction cell consisting of InGaP/GaAs and crystalline Si,” in 2017 IEEE 44th Photovoltaic Specialists Conference (PVSC), IEEE, 2017. Search in Google Scholar
 N. Shigekawa, J. Liang, R. Onitsuka, T. Agui, H. Juso, and T. Takamoto, “Current–voltage and spectral-response characteristics of surface-activated-bonding-based InGaP/GaAs/Si hybrid triple-junction cells,” Jpn. J. Appl. Phys., vol. 54, no. 8S1, p. 08KE03, 2015, https://doi.org/10.7567/jjap.54.08ke03. Search in Google Scholar
 S. Essig, J. Benick, M. Schachtner, A. Wekkeli, M. Hermle, and F. Dimroth, “Wafer-bonded GaInP/GaAs//Si solar cells with 30% efficiency under concentrated sunlight,” IEEE J. Photovoltaics, vol. 5, no. 3, pp. 977–981, 2015. Search in Google Scholar
 N. Shigekawa, L. Chai, M. Morimoto, J. Liang, and T. Takamoto, “Hybrid triple-junction solar cells by surface activate bonding of III–V double-junction-cell heterostructures to ion-implantation-based Si cells,” in Photovoltaic Specialist Conference, IEEE, 2014. Search in Google Scholar
 F. Dimroth, T. Roesener, S. Essig, et al., “Comparison of direct growth and wafer bonding for the fabrication of GaInP/GaAs dual-junction solar cells on silicon,” IEEE J. Photovoltaics, vol. 4, no. 2, pp. 620–625, 2014, https://doi.org/10.1109/jphotov.2014.2299406. Search in Google Scholar
 K. Derendorf, S. Essig, E. Oliva, et al., “Fabrication of GaInP/GaAs//Si solar cells by surface activated direct wafer bonding,” IEEE J. Photovoltaics, vol. 3, no. 4, pp. 1423–1428, 2013, https://doi.org/10.1109/jphotov.2013.2273097. Search in Google Scholar
 M. Baba, K. Makita, H. Mizuno, et al., “Feasibility study of two-terminal tandem solar cells integrated with smart stack, areal current matching, and low concentration,” Prog. Photovoltaics, vol. 25, no. 3, pp. 255–263, 2017. Search in Google Scholar
© 2021 Xingliang Li et al., published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.