The escalating demand for solar energy highlights the critical need for advancing more efficient next-generation photovoltaic (PV) technologies. Currently, perovskite-based single junction and tandem solar cells are reaching or even exceeding the performance of established technologies like three-five tandem and GaAs solar cells. In the case of perovskite versus GaAs, employing nanophotonics to fine-tune the PV bandgap emerges as a viable strategy, especially when reaching the limits of composition engineering of FAPbI3. Tandem solar cells present another practical approach to surpass the single-junction Shockley-Queisser limit. Our initial efforts concentrated on solution-coating thick perovskite layers onto small, textured silicon substrates, successfully proving the concept's viability. In Singapore, we are actively working to shift this technology from laboratory research to manufacturing, focusing on applying perovskite layers to industrial Czochralski (CZ) silicon wafers. With cooperative efforts, our goal is to propel this technology from lab-scale to full-scale fabrication, aiming to surpass the performance of three-five PVs in full module size, thereby addressing the growing global energy demands with more efficient solar solutions.

Integrating metal halide perovskite top cells with bottom cells formed by crystalline silicon or low band gap perovskites into monolithic tandem devices has recently attracted increased attention due to the high efficiency potential and application relevance of these cell architectures. Here we present our recent results on monolithic tandem combinations of perovskite top-cells with crystalline silicon, and Sn-Pb perovskites as well as tandem relevant aspects of perovskite single junction solar cells. 

In 2020, we have shown that self-assembled monolayers (SAM) could be implemented as appropriate hole selective contacts. The implementation of new generation SAM molecules enabled further reduction of non-radiative recombination losses with high open circuit voltages and fill factor. By fine-tuning the SAM molecular structure even further, the photostability of perovskite composition with tandem-ideal band gaps of 1.68 eV could be enhanced by reduction of defect density and fast hole extraction. That enabled a certified efficiency for perovskite/silicon tandems at 29.15% [1].

By optical optimizations, we could further improve this value to 29.80% in 2021. Periodic nanotextures were used that show a reduction in reflection losses in comparison to planar tandems, with the new devices being less sensitive to deviations from optimum layer thicknesses. The nanotextures also enable a greatly increased fabrication yield from 50% to 95%. Moreover, the open-circuit voltage is improved by 15 mV due to the enhanced optoelectronic properties of the perovskite top cell on top of the nanotexture [2].

In the end of 2022, we enabled a new world record for perovskite/silicon tandem solar cells at 32.5% efficiency. We demonstrated that an additional surface treatment strongly reduces interface recombination and improves the band alignment with the C60 electron transporting material. With these modifications, single junction solar cells show high open circuit voltages of up to 1.28 V in a p-i-n configuration, and we achieve 2.00 V in monolithic tandem solar cells [3]. A comparable surface treatment was also applied to 1.80 eV band gap perovskites to enable Voc values of 1.35 V and these were integrated into monolithic all-perovskite tandem solar cells enabling a certified efficiency of 27.5% [4].

In addition to the experimental material and device development, also main scientific and technological challenges and empirical efficiency limits as well as advanced analysis methods will be discussed for perovskite based tandem solar cells. In addition, results on upscaling and stability of these industrial relevant tandem solar cells by thermal evaporation will be shown.

Perovskite-based tandem solar cells, with their compelling performance and cost-effective fabrication, stand at the forefront of photovoltaic technologies. Nevertheless, reduction in energy loss and enhancements in robustness of the front wide-bandgap perovskite sub-cell remain crucial, aiming for an efficient and stable perovskite-based tandem device. Surface passivation using large organic spacer cations on the top of perovskite films is a promising strategy to address the issues mentioned above, as it can effectively diminish the defect density. Simultaneously, the introduction of low-dimensional phases into perovskites has been demonstrated to decrease the minority concentration at the interface, effectively reducing nonradiative recombination and mitigating the mismatch between the open-circuit voltage (V_OC) and the quasi-Fermi level splitting (QFLS), particularly for wide-bandgap-based perovskites.[1-4] In some other instances, the enhancement in VOC following surface passivation was attributed to the dipole moment of specific spacer cations that did not give rise to low-dimensional phases.[5,6] Moreover, the reduction in energy offsets of conduction bands (CBs) between the perovskite layer and the transporting layers also directly results in a V_OC enhancement.[6,7] Uniquely separating passivation from energy alignment remains challenging, especially for multifunctional cations that can form 2D phases but also generate a dipole moment. Therefore, to achieve further enhancements in V_OC and fill factor (FF) for wide bandgap perovskite sub-cells, it is essential to develop a comprehensive understanding of the underlying mechanisms.

As it will be discussed at the conference, the impacts of surface treatments with guanidinium bromide (GABr), 4-fluorophenylammonium iodide (F-PEAI) and their mixture on the formation of low-dimensional phases, device performances, as well as underlying loss mechanisms. Based on a hybrid experimental (QFLS) - theoretical (drift-diffusion simulation) approach, it reveals that the reductions in surface recombination velocity and energy level offsets between perovskite and electron transporting layer are the major contributors to a record V_OC × FF production for 1.80 eV perovskite solar cell, which exhibited notably reproducibility transferring to another laboratory. After integrating with a narrow-bandgap perovskite rear cell, we demonstrated an efficient all-perovskite tandem solar cell with stable output of 27.2% under maximum power point.

Tandem technology is the most promising route to further enhance the power conversion efficiency of solar cells. In recent years perovskite/silicon tandem solar cells have demonstrated an unprecedented rise of efficiencies far beyond the silicon single-junction efficiency limit. Further, all-perovskite multijunction solar cells gained increasing interest due to its compatibility with a fully printable, flexible, and lightweight technology platform.

In this contribution, optical aspects of perovskite tandem solar cells are discussed. This comprises 1) “classical” light management issues such as the minimization of reflection losses and the improvement of light trapping in thin absorber layers for the full spectral range. To achieve this, usually nano- and or micro-textures are implemented into the device coming along with special technological challenges concerning the compatibility of the textures with the respective perovskite deposition method. Further, we discuss 2) current-matching constraints in monolithic perovskite tandem devices. While under standard texting conditions there is little tolerance concerning the engineering of the perovskite bandgap(s), there is evidence that outdoor weather conditions, bifacial illumination as well as luminescent coupling can strongly relax current-matching constraints. Finally, we will give an outline how 3) optical engineering of luminescence and re-absorption processes can not only influence the short-circuit current density but also the electronic performance of the tandem solar cell device.

Halide perovskites overwhelmed the field of photovoltaics with unprecedented progress in efficiency. Their facile bandgap tunability renders perovskite solar cells excellent building-blocks for multi-junction architectures, that provide the prospect to overcome fundamental efficiency limits of single-junctions. Combinations of perovskite wide-gap cells exist with a large variety of narrow-gap technologies, such as silicon or CIGS. Furthermore, low-cost tandem technologies are particularly interesting, such as all-perovskite tandems. However, stability concerns exist especially for narrow-gap perovskite cells, that typically contain large amounts of tin (instead of lead). The notorious oxidation of the Sn²⁺ to Sn⁴⁺ infers a detrimental self-doping [1]. Since this issue might impose a fundamental stability limit, organic solar cells are an attractive alternative as a narrow-gap subcell to form perovskite/organic tandems. Since the introduction of non-fullerene acceptors has revived the field, organic solar cells now reach efficiencies >19% and express absorption spectra extending well into the infrared. Organic and perovskite semiconductors share similar processing technologies, which makes them attractive partners in multi-junction architectures. I will discuss the prospects and challenges of perovskite-organic tandem solar cells by highlighting the key aspects of the individual building blocks and their interplay in the tandem. Specifically, the role of non-fullerene acceptors in efficient narrow-gap organic solar cells with high operational stability is discussed. A further focus is the wide-gap perovskite solar cell, where long term stability is the most pressing issue that needs attention. Eventually, the design and functionality of high-quality interconnects are outlined along with a view on its impacts on the characteristics of the tandem. I will present a specific example of a perovskite/organic tandem using an ALD grown InO_x interconnect yielding a very promising efficiency of 24% with prospects reaching well beyond 30%.[2] In the end, I will benchmark perovskite-organic tandem solar cells against other emerging tandem solar cell technologies.

Tandem photovoltaic with complementary absorbing sub-cells is a promising pathway to overcome the thermalization loss and surpass the detailed balance limit of single junctions, stimulating intense interest. In recent years, the integration of wide-bandgap perovskite solar cells with narrow-bandgap organic solar cells represents a more prospective route in constructing tandem solar cells. However, for n-i-p type perovskite/organic tandems, the crucial interconnection layer (ICL) has been limited to the MoO_x/metal nanoparticles/ZnO structure. As is well-known, this ICL has several limitations, leading to undesired optical and energy losses within the tandem's ICL, as well as high costs and time consumption from a fabrication perspective. In our study, we introduced a solution-processed hole transporting material to replace the traditional thermally evaporated MoO_x in building the ICL for inorganic-perovskite/organic tandems. The optimized ICL exhibited significantly improved transmittance, particularly in the near-infrared region, which was conducive to enhancing the efficiency of the rear organic cell in harnessing light and achieved an increase in short-circuit current density (J_SC) of approximately 1.5 mA/cm². Additionally, owing to the favorable energy level alignment in the optimized ICL, the tandem device achieved a recorded open-circuit voltage of up to 2.3 V, along with an efficiency of 21% and enhanced stability.

The efficiency of perovskite solar cells has surpassed 26%, approaching the fundamental efficiency limits for single-junction cells. Tandem technology, combining two solar cells with different bandgaps, offers a solution to overcome detailed-balance limits by reducing thermalization loss. Although the best perovskite/silicon and all-perovskite tandem devices have exceeded 33% and 29% efficiencies respectively, the high energy consumption of silicon wafers leads to significant carbon emissions, and the self-doping oxidation effect (Sn²⁺ to Sn⁴⁺) in narrow-bandgap perovskites poses an instability issue. These challenges can be addressed by changing the narrow bandgap cells to non-fullerene organic photovoltaics (OPVs). In such structures, solution-processed OPVs are energy-friendly and do not have stability concerns related to self-doping oxidation. The recent emergence of near-infrared acceptor Y6 further positions perovskite/organic tandem solar cells as a promising technology.

In previous work, we integrated PM6:Y6:PC₆₁BM ternary OPVs into a perovskite/organic tandem devices, yielding a record efficiency of 24%. Notably, OPVs maintained approximately 95% original efficiency after 5000 hours of continuous operation under irradiation with low-energy photons (850 nm), but degraded rapidly when illuminated with a white light-emitting diode (LED), indicating that the visible spectral region could be responsible for device degradation^[1]. In this work, we utilize monochromatic LED sources covering ultraviolet, visible, and near-infrared spectral region to systematically explore the photostability of Y6 ternary OPVs. Our results reveal that under continuous operation in the maximum-power point under irradiation with low-energy photons (λ > 590 nm), the devices show long-term stability (>1000 hours), while high-energy photons (λ < 530 nm) infer degradation, with the device's T90/T80 lifetime strongly correlated with the photon energy. Additionally, the distinct degradation behaviors of single polymers or NFAs under monochromatic light exposure elucidate their respective contributions to degradation in bulk heterojunction devices.

Interestingly, ternary OPVs featuring other Y-family NFAs with various  energy gaps (Y18 (1.31 eV), CH1007 (1.30 eV), mBzS-4F (1.25 eV)) exhibit similar degradation behaviors under the same illumination conditions,  verifying that the photo-degradation behavior evidenced above is generally valid. The distinctive photostable property of OPVs under low-energy photons can eliminate concerns regarding the stability of the rear cell in perovskite-based tandem structures, because the front perovskite cells function as spectral low-pass filters. We integrates these OPVs as sub-cells in a tandem structure with a wide-bandgap perovskite cells, the corresponding photostability of perovskite/organic tandem devices strongly depends on that of the wide-bandgap perovskite sub-cells, highlighting the advantage of organic rear cells and directing community focus more towards the stability of wide-bandgap perovskite in the future.

Stacking two perovskite layers with complementary bandgap in tandem device promise to exceed the efficiency limit of single-junction solar cells. In addition to high-efficiency potential, perovskite-perovskite (all-perovskite) thin-film tandems can be fabricated on flexible and lightweight substrates with very high power-to-weight ratios, thus opening numerous applications where flexibility and lightweight are important considerations. Leveraging low-temperature coating methods and high throughput roll-to-roll manufacturing, perovskite-based thin-film tandems promise very low manufacturing costs and CO₂ footprint. Working towards our vision of roll-to-roll (R2R) manufacturing of flexible all-perovskite tandem solar modules, we present our latest progress on the development of highly efficient flexible all-perovskite tandem devices (Fig. 1). With interface and additive engineering, we demonstrate 25.4 % power conversion efficiency in a flexible 2-terminal all-perovskite tandem solar cell. Furthermore, we will present the first flexible all-perovskite tandem mini-modules and discuss the challenges of homogenous coating of perovskite on large areas using both scalable solution processes and vapor processes. Finally, we will discuss the potential and future challenges of developing highly efficient, lightweight, and stable all-perovskite tandem photovoltaics.

Metal halide perovskite solar cells have advanced into a viable option for future renewable energy. Record single and tandem junction all-perovskite solar cells already provide power efficiencies of over ~26% and ~28%, respectively. A next target in photovoltaic energy conversion can possibly be met by developing perovskite triple or even quadruple junction solar cells. While these hold a promise to afford higher efficiencies, they require developing stable perovskite sub cells with bandgaps in the range of 1.8 to 2.3 eV, i.e., a range that has not received much attention so far. These wide-bandgap perovskites often suffer from more pronounced voltage losses due to non-radiative bulk and interfacial charge recombination. In developing new perovskite sub cells, photocurrent spectroscopy and absolute photoluminescence spectroscopy are used in combination with bulk and interface passivation strategies to eliminate these losses. This has enabled to reduce the voltage deficit over a wide range of bandgap. Guided by optical modeling, monolithic multi-junction solar cells have been fabricated by stacking two and three different bandgap perovskite sub cells in series using recombination junctions designed to provide near-zero electrical and optical losses. Collectively, these strategies enable monolithic tandem and triple junction solar cells with a power-conversion efficiency of over 26%.

Understanding the performance limiting factors of organic solar cells (OSCs) with very small optical bandgaps is crucial for the development of novel tandem photovoltaics (PVs), such as OSC/OSC or perovskite/OSC tandems. For example, by combining a low-bandgap OSC ~ 1.1 eV with a high-bandgap absorber ~ 1.7 eV into a tandem solar cell, a power conversion efficiency (PCE) as high as 31.0% can be theoretically achieved, according to the detailed balance (DB) limit.

In this study, we investigate the loss mechanisms of ultra-low-bandgap OSCs based on the blend of the donor polymer PTB7-Th with the non-fullerene acceptor (NFA) COTIC-4F, with a bandgap of 1.15 eV, a promising candidate for tandem PV. In a conventional device structure, we reach a record PCE value of 8.55% with an enhanced short-circuit current density (J_SC) and fill factor (FF) compared to previous records that relied on an inverted device geometry. While setting a record for this system, our PCE metrics still fall behind higher-bandgap OSC systems.

To guide further improvements, we investigate the various loss mechanisms and recombination processes using photoluminescence (PL) measurements, bias-dependent time-delayed collection field (TDCF) measurements, bias-assisted charge extraction (BACE), fluence-dependent photoinduced absorption (PIA), as well as light intensity-dependent open-circuit voltage (V_OC). Complemented with optical and electrical device simulations, we show that J_SC loss can be largely attributed to inefficient exciton dissociation in combination with geminate recombination of the charge transfer state. Further, we identify the fairly high bimolecular recombination coefficient as the main reason for the poor performance, while surface recombination is shown to mainly affect V_OC. Finally, our simulations show that the simultaneous reduction of bimolecular recombination (e.g. by ternary blends), surface recombination (e.g. by self-assembled monolayers), exciton and charge transfer recombination (e.g. by vapor annealing) would enable efficiencies of >15% in the PTB7-Th:COTIC-4F system.

Mobile ions play a significant role in perovskite photovoltaics (PV), yet their impact on the overall performance and stability of tandem solar cells (TSCs) remains largely unexplored. Moreover, the effects of hysteresis in the current-voltage (JV) characteristic and ionic field screening are usually not considered to be a major problem anymore in high-performance tandem cells due to the introduction of comparatively stable and well-performing pin-type perovskite cells. This conclusion is based on the established practice of using a relatively slow JV scanning rate for the characterization of these devices. Here, based on recent work,[1-4] I will present a comprehensive study that combines an experimental analysis of ionic losses in Si/perovskite and all-perovskite TSCs during device aging with drift-diffusion simulations. Our findings demonstrate that mobile ions have a significant influence on the hysteresis of both tandem cells at high JV scan speeds (e.g. 400 V/s) as well as on performance degradation due to field screening. Additionally, subcell-dominated “fast-hysteresis” measurements on all-perovskite tandems reveal more pronounced ionic losses in the wide-bandgap subcell during aging, which we attribute to its tendency for halide segregation. Drift-diffusion simulations fully corroborate the results. Finally, I will discuss how we can use the obtained ionic properties as an early fingerprint to predict the long-term stability of perovskite cells. Overall, our research provides valuable insights into how ion migration influences the energy-lifetime yield of perovskite PV and highlights new strategies to improve the stability of all perovskite-based single- and multi-junction cells.