Among transparent electrode architectures, TCO/metal/TCO multilayers can deliver optical transparency and electrical conductivity that match or surpass those of single TCO films and ultrathin metal layers. In this work, tin-doped indium oxide (ITO) thin films were first optimized on large-area substrates (156 × 156 mm²) by systematically varying the target-to-substrate distance, sputtering pressure, and applied RF power. We report an optimized 50-nm-thick ITO layer deposited at a target-to-substrate distance of 115 mm, which exhibited a thickness non-uniformity of ∼10%, along with a carrier mobility of 47 cm² V⁻¹ s⁻¹, a carrier concentration of 2.60 × 10²⁰ cm⁻³, and a resistivity of 9.24 × 10⁻⁴ Ω cm. Subsequently, oxide/metal/oxide (OMO) transparent electrode configurations were optimized using the optimized ITO as the oxide layer, investigating different combinations of oxide thickness and metal (Ag and Al) thicknesses. Among the studied structures, we report the ITO/Ag/ITO (25/10/25 nm) stack demonstrated superior performance, achieving a low sheet resistance of 12 Ω sq⁻¹ and a high transmittance of 82.5% at 550 nm. When integrated into front/back-contacted silicon heterojunction (FBC-SHJ) solar cells, the optimized OMO electrode improved device performance compared with the other reference electrodes. With respect to the optimized 50-nm-thick ITO electrode, the ITO/Ag/ITO (25/10/25 nm) structure shows clear improvements in device performance, with increases of 16 mV in V_OC, 0.20 mA cm⁻² in J_SC, and 1.23% in FF. These gains represent an absolute efficiency improvement of 1.00%, leading to a maximum efficiency of 23.20%. Interestingly, when a much thinner Ag layer is used (ITO/Ag/ITO, 25/2/25 nm), the device performance drops noticeably efficiency of 16.20%. We report that increasing the Ag thickness to 10 nm, however, leads to a substantial recovery and enhancement, emphasizing how sensitive device performance is to the thickness of the metallic interlayer.

Partial shading remains a critical challenge for perovskite photovoltaics, as shaded cells in otherwise illuminated modules operate in reverse bias, which can accelerate degradation. In other solar cell technologies, reverse bias electroluminescence (ReBEL) imaging has been successfully employed to investigate the reverse bias breakdown mechanism. Here, ReBEL imaging is proposed as technique for imaging the local reverse bias current flow on perovskite solar cells. To that end, the ReBEL signal is affirmed to originate from radiative recombination in the absorber layer. ReBEL imaging is compared to established thermal current imaging techniques and reveals superior sensitivity and spatial resolution while showing similar features in the investigated samples. Further experiments reveal a dependence on the reverse bias current: The spatial distribution of the ReBEL signal differs significantly between small and large current injection levels. This could point towards two different breakdown mechanisms. These results show that, despite some open questions, ReBEL imaging could help understand the reverse bias behavior of perovskite solar cells better.

This article introduces a parallel differential power processing (PDPP) architecture for photovoltaic (PV)/battery applications. The PV to Virtual Bus (PV2VB) architecture enables the integration of a battery and manages its power while performing maximum power point tracking on the PV strings. In the proposed PV2VB PDPP architecture, the battery is positioned at the virtual bus, acting as the input for all string-level converters (SLCs). By selecting a lower voltage for the battery at the virtual bus compared to the PV string or the main bus voltages, component voltage ratings can be reduced. The architecture employs dual active bridge converters connected to bridgeless (BL) converters as SLCs to generate both positive and negative output voltages while providing isolation. These SLCs track the maximum power point of each PV string, while the central converter manages battery charging and discharging. Experimental results confirm the performance and effectiveness of the proposed PV2VB PDPP architecture, achieving efficiencies between 95.5% and 99%.

The degradation of perovskite solar cells due to reverse bias (RB) is one of the remaining challenges hindering the commercialization of the technology. To overcome this challenge, a thorough understanding of and control over the breakdown (BD) voltage are crucial. A prerequisite for this is that the community “speaks the same language,” that is, that the reported BD voltages are comparable. A review of literature data shows that the impact of measurement parameters is often unknown and seems to depend strongly on sample properties. It follows that standardization is the only way to reach comparability. Here, a set of measurement parameters to fill this gap is proposed. Additionally, various definitions of a “BD voltage” are used in parallel without any way of relating them to each other; this metric and its determination need to be considered as well. After a thorough discussion of the available definitions, the use of the point of maximum curvature is introduced. Its main advantage is the possible connection to an analytical description of the BD mechanism. In this way, a starting point for scientists new to the field of RB stability is provided, and the ground for a broader discussion in the community is prepared.

Offshore floating photovoltaics (OFPVs) emerge as a promising solution to overcome land constraints associated with inland renewable energy deployment. However, as OFPVs are still a developing technology, several performance-related uncertainties persist. The reduction in energy yield caused by wave-induced losses (WIL) is one such critical uncertainty that needs to be understood, quantified and minimised. To address this need, this work introduces a physics-based modelling framework that couples validated hydrodynamic simulations with opto-electrical analysis to accurately estimate WIL. An extensive sensitivity analysis is then carried out, performing over 100 simulations by systematically varying both design and environmental parameters. The results show that WIL ranges between 1%–30% on an hourly basis and exhibits a nonlinear dependence on both parameter groups. The resulting dataset is then used to develop S [Figure presented] IFT 1.0 - a surrogate model capable of predicting WIL across a wide range of design and operating conditions, achieving an average absolute RMSE of 3% relative to the physics-based model. The insights from S [Figure presented] IFT 1.0 are finally used to provide practical measures that minimise WIL at a system design level. Overall, this work provides a complete pathway to model, quantify, predict, and minimise WIL, promoting confident and scalable OFPV deployment.

Photovoltaic (PV) system performance is linked to climatic conditions in which the system operates. This leads to the Köppen-Geiger-Photovoltaic (KGPV) climate classification. KGPV is created by overlaying four irradiation levels with the commonly used Köppen-Geiger climate zones. Potential drawbacks of this approach are that the climate features are not considered in a combined manner in the sorting process and that the KGPV zones inherent a dependence on precipitation. We propose a machine-learning approach to address this deficiencies and improve PV climate classification. First, supervised learning is used to evaluate the correlation between climate features and a PV system's specific energy yield. We find that the inclusion of the darkest and brightest irradiation months as well as UV irradiation improves accuracy, while wind speed, relative humidity, precipitation and annual mean daily temperature difference have little impact on accuracy. Subsequently, k-means clustering combined with comprehensive qualitative analysis, identifies a PV classification based on seven climate features and 21 clusters. A mountainous climate characterized by moderate to low temperature and high irradiation is uncovered compared to KGPV. Moreover, this new PV climate classification reduces the sum of squared errors by 58 % compared to KGPV clearly signifying a more accurate PV climate classification approach.

As crystalline silicon (c-Si) solar cells approach their theoretical efficiency limit, the perovskite/silicon (PerSi) tandem technology offers a promising solution for further improving the efficiency of photovoltaic (PV) modules. However, as perovskite cells are facing stability issues, it is unclear whether PerSi modules will have a larger lifetime energy yield (LEY) than c-Si modules. In this work, we present a novel methodology to simulate the LEY of PerSi tandem devices, accounting for environmental stress factor-dependent degradation across four different climates. Our approach combines a physics-based analytical degradation model for components shared with c-Si modules and a scenario-based degradation model for the perovskite top cell. This method enables us to identify the tolerable degradation rate (k_tol) of the perovskite cell under different scenarios and climatic conditions. We find that k_tol is lowest when degradation occurs in the short-circuit current, reaching a minimum value of 1.2% per year in Delft (the Netherlands). Additionally, we demonstrate that k_tol inversely depends on the module lifetime, reaching values up 7.6% per year in Lagos (Nigeria). Moreover, we show that module efficiency (η_mod) significantly impacts k_tol. For instance, increasing η_mod from 28.0% to 32.9% raises k_tol by approximately 50%. Additionally, we propose a simplified model that can predict k_tol without the computationally intensive simulations, which has a root-mean-square error of 0.34% per year. Lastly, environmental impact assessments reveal that PerSi modules are more sustainable in all impact categories when the degradation rate is 80% of k_tol for LEY.

Perovskite/silicon (PS) technology includes three main configurations: two-terminal (2T), three-terminal (3T), and four-terminal (4T). Previous studies have made various comparisons between these configurations, significantly advancing our understanding of these devices. While these studies mostly focus on simulations on cell level, we perform bandgap energy ((Formula presented.)) optimization at the module level for different configurations under outdoor conditions. Using opto-electrical simulations, we predict the energy yield of each module at four geographical locations, with varying values of (Formula presented.). The optimal (Formula presented.) for the 2T, 3T, and 4T modules are 1.62, 1.80, and 1.82 eV, respectively. We also perform a loss analysis to explore the differences in power losses among the configurations. These loss differences can be attributed to the configurations having different optimal (Formula presented.) values (affecting the thermalization losses) or different module designs (affecting the interconnection losses). Among all losses, mismatch losses play the most critical role in optimizing the bandgap. Overall, all optimized configurations have similar energy yields (all differences within 1.5%) across all locations. Finally, we compare the robustness of the different configurations against different scenarios of perovskite degradation. Our results show that the 4T module is the least sensitive to degradation in the perovskite subcell.

In urban settings, residential photovoltaic (PV) systems typically involve conversion from direct to alternating current either at the string or module level. Generally, conversion at the module level offers better performance in shading scenarios induced by the built environment but is usually more expensive. This study explores the potential of ‘Smart’ PV modules, which include built-in buck converters that allow substrings within a PV module to operate independently and thus improve PV energy yield in shaded urban scenarios. Eight PV system configurations, varying in inverter topology and cell architecture, are simulated across three locations. Results show that systems using string inverters with Smart modules achieve the highest annual energy yield. Financial analysis further demonstrates that for the considered scenario’s up to 50% additional module costs are tolerable for a similar financial performance. In addition to energy yield improvements, Smart modules reduce the maximum temperatures observed within the system.

Advanced and emerging photovoltaic (PV) technologies play a crucial role in meeting the increasing global energy demand sustainably. Simulations are essential for predicting system behavior and improving our understanding of complex PV architectures. This work extends an existing modeling framework designed for novel PV systems, offering a modular and flexible workflow suitable for diverse research applications. The framework computes PV performance from first-principles physics, removing the need for module datasheets. It comprises two pre-processing steps and six simulation steps. The first steps determine the optical behavior of the modules, followed by irradiance modeling and temperature calculations. The final steps evaluate the electrical characteristics and the conversion to alternating current at the full-system level. The framework incorporates detailed energy loss analysis and includes advanced features such as partial shading, reverse-bias effects, and photon recycling. Two applications demonstrate its capabilities: comparing module configurations in urban settings and optimizing multi-junction PV system design. Results show that Smart modules enhance shade resilience, delivering approximately (Formula presented.) higher energy yields. Additionally, the optimal perovskite bandgap for perovskite/silicon tandem devices is found to be 1.60–1.62 eV. These outcomes highlight the framework's value for future PV system research and development. The developed software can be found at: https://github.com/YBlom1999/PVMD_Toolbox.

Light-simulation tools—exemplified by Radiance—are widely used for quantitative daylight studies and are increasingly adopted in agrivoltaics (agri-PV) to handle complex geometry via ray tracing. Yet, beyond typical workflows three practical limitations persist: spectrally resolved skies are restricted to the visible band; soft-shadow (penumbra) rendering relies on runtime-intensive solar-disk sampling; and fast, integrated canopy models remain scarce. We present a Radiance-compatible Python framework that adds: (i) atmosphere-specific sun–sky generation across the solar spectrum; (ii) efficient, equal-area sampling of the solar disk; and (iii) a simple canopy reconstruction tailored to narrow-trained orchards. To improve spectral fidelity, resolution, and range, we couple SMARTS-derived spectra to a Perez-based sky, leveraging Radiance's multispectral rendering. We deterministically sample the sun's finite extent using a Fibonacci lattice, yielding stable penumbra without prohibitive runtimes. The canopy model parameterizes porosity and seasonal development at a daily rate. Canopy representation matters: opaque–static models, common in agri-PV simulations, systematically underestimate light levels and miss spatiotemporal patterns needed to diagnose suboptimal conditions. Comparatively, a porous–dynamic model led to ≈26% higher seasonal light levels, with gains attaining ≈100% early in the season and converging to ≈16% after foliage matured. While penumbra is limited under conventional PV modules, penumbra-capable renderings enable exploration of design pathways—narrower cell layouts (half-cell and beyond) with greater module–canopy separation—that smooth lighting extremes.

The monolithic integration of perovskite top cells on textured crystalline silicon affords efficient tandem devices with strong prospects for large-scale applications. Such integration has primarily relied on state-of-the-art recombination junctions, which typically comprise transparent conductive oxides and molecular self-assembled monolayer (SAM) contacts. However, the potential influence of bottom cell nanoroughness, which may vary based on specific processing routes and technologies, has received far less attention. Here, we systematically engineered the top surface nanoroughness of silicon heterojunction solar cells to examine its impact on monolithic perovskite–silicon tandem solar cells. We employed two approaches: (i) varying the thickness of (n)-type hydrogenated nanocrystalline silicon ((n)nc-Si:H) layers or (ii) applying a plasma treatment using a hydrogen and carbon dioxide gas mixture before the deposition of (n)nc-Si:H layers. Both methods enhanced the conductivity and crystallinity of (n)nc-Si:H layers and increased the surface nanoroughness, with plasma treatment enabling the efficient realization of distinct nanoroughness in thin (n)nc-Si:H (15-nm-thick) layers. Our results reveal that the surface nanoroughness imposed by (n)nc-Si:H layers influences the SAM anchoring, leading to increased work function shifts and improved SAM/perovskite interface quality, thereby impacting the overall tandem device performance. Notably, tandem devices incorporating higher-nanoroughness bottom cells achieved increased fill factors, dominating the observed tandem efficiency enhancements, with a peak efficiency of 32.6% enabled by a 30-second-long plasma treatment.

Silicon is a promising alternative to the conventional graphite anodes due to its high theoretical capacity and favorable lithiation potential for lithium-ion batteries (LIBs) with liquid as well as solid-state electrolytes. However, lithiation-induced extreme volume change causes severe mechanochemical deformation and continuous formation of solid-electrolyte interphase leads to cell failure. One of the strategies to mitigate this problem is alloying silicon with a suitable element that can alter the surface electrochemistry and/or lithiation pathways, and acts as mechanical buffer. Nonetheless, these benefits come with a compromise on the specific capacity, which strongly influences the mass loading of the electrodes, highlighting the need to deconvolute the intertwined influence of composition and mass loading when designing high performance electrodes. In this work, we systematically studied the influence of composition and mass loading in monolithic amorphous silicon and non-stoichiometric silicon nitride (SiN_x) electrodes on their electrochemical performance as LIB anodes. The incorporation of nitrogen in the electrode matrix clearly improves the electrochemical stability at the expense of reduced specific capacity, while higher mass loading accelerates capacity fading, most critically in amorphous silicon electrodes. Postmortem analysis reveals that such capacity fading in the electrodes with higher mass loading can be related to delamination due to evolved tensile stress during the charge–discharge cycle. Yet, nitrogen-rich SiN_x monolithic electrodes accommodate strain more effectively. These findings demonstrate that while pristine Si delivers high specific capacity and long-term stability in thin films, thicker (>1 µm) monolithic electrodes benefit from higher nitrogen content in SiN_x, which provides more stable cycling and sustained capacity.

Bifacial perovskite/silicon solar cells can combine the advantages of tandem technology (high efficiencies) and bifacial modules (additional received irradiance from the rear) to increase the energy yield of photovoltaic (PV) systems further. In literature, it has already been shown that for two-terminal tandems this would require a lower bandgap energy (E_g) for the perovskite cell, as the rear irradiance increases the current in the bottom cell creating a current mismatch, if this is not considered during optimization. This work expands on bifacial two-terminal tandem optimization by considering aspects not included before. Besides the E_g, the thickness (d) of the perovskite is also optimized, as this also affects the current matching. Additionally, this work studies the trends in different energy losses of the PV module to better understand what affects the optimal perovskite cell. Our simulations show that the optimal E_g is 1.61–1.65 eV and the optimal d is 650–750 nm, which agrees with the observations in literature. The optimal E_g and d are mostly a trade-off between mismatch and thermalization losses, meaning that the mismatch losses should not be fully minimized. Additionally, the irradiance from the rear side is converted less efficiently than the front side irradiance due to larger thermalization and reflection losses. Therefore, the energy yield of bifacial tandem modules, compared to monofacial tandem ones, only increases for large ground albedo. Finally, our results show that the bifacial tandems have over a 25% gain in energy yield compared to bifacial single junction modules and up to 5% gain compared to monofacial tandem modules.

The uncertainty surrounding land availability for renewable energy deployment is a growing concern, creating a strong need for alternative solutions. In recent years, offshore floating photovoltaic (OFPV) systems have shown great promise in meeting global energy demands without competing for land resources. With ambitious targets like 3 GW in The Netherlands by 2030 and global projections exceeding 20 GW, OFPVs are emerging as a key solution at this critical juncture in energy transition. The significance of this technology is also reflected in the 95% increase in research outputs over the past five years. Despite this growth, insights remain scattered, with limited understanding of both the technology and performance. This review fills this gap by providing a comprehensive overview of OFPV systems, addressing both technical and performance aspects. Specifically, the objectives are to: provide detailed information about technology readiness levels, real-world deployments, and a new classification matrix to categorize different OFPV designs; identify key processes like dynamic motion, cooling, optical changes, and long-term degradation that impact energy yield (EY); and quantify the impact of each process on EY based on reported data. The findings reveal that dynamic motion (-0.4% to -15%) and long-term degradation (-2% to -20%) generally reduce EY, while cooling (-4% to +20%) and optical effects (-40% to +5%) can enhance or reduce EY depending on operating conditions. While these insights are crucial, several challenges remain, with the most pressing being the need to standardize measurement and modeling techniques for EY prediction to propel OFPVs towards large-scale commercialization.