Halide perovskites are highly promising, emerging semiconductors for a variety of applications. Here I will delve into their nanoscale properties including the effect on macroscopic optoelectronic properties.

First, I will show quantum picosecond transients appearing in formamidinium-rich halide perovskite absorbers at low temperature. From correlative local measurements, we show that these quantum transients originate from local nano-twinning structures.

Next, I will show the presence of dynamic, lower-symmetry local nanodomains embedded within the higher-symmetry average phase in various perovskite compositions. The properties of these nanodomains are tunable via the A-site cation selection: methylammonium induces a high density of anisotropic, planar nanodomains of out-of-phase octahedral tilts, while formamidinium favours sparsely distributed isotropic, spherical nanodomains with in-phase tilting, even when crystallography reveals cubic symmetry on average. By demonstrating the influence of A-site cation on local nanodomains and consequently, on macroscopic properties, we show the connection between these local dynamic domains and optoelectronic properties.

Finally, I will show recent results of epitaxial, layer-by-layer growth of halide perovskites by vapour deposition, leading to a new library of high-quality materials.

Atomic-scale microstructure of lead halide perovskite thin films

Understanding the atomic-scale crystallographic properties of photovoltaic semiconductor materials such as silicon, GaAs, and CdTe has been essential in their development from interesting materials to large-scale energy conversion industries. However, studying photoactive hybrid perovskites by transmission electron microscopy (TEM) has proved particularly challenging due to the large electron energies typically employed in these studies. [1,2] In particular, the very close structural relationship between a number of crystallographic orientations of the pristine perovskite and lead iodide has resulted in severe ambiguity in the interpretation of EM-derived information, severely impeding the advance of atomic resolution understanding of the materials.

In this talk, I will outline how to reliably study hybrid organic-inorganic perovskite materials using electron microscopy. With the ability to image the pristine phase of these beam-sensitive materials, we are able to obtain highly localised crystallographic information about technologically relevant materials. Using low-dose selected area electron diffraction, I will show how mixing the archetypal CH(NH₂)₂PbI₃ (FAPbI₃) and CH₃NH₃PbI₃ (MAPbI₃­) improves solar cell device performance through the elimination of twin domains and stacking faults. [3]

Using a careful low-dose scanning TEM (STEM) protocol, we are also able to image these materials in their thin-film form with atomic resolution. [4] Our images enable a wide range of previously undescribed phenomena to be observed, including a remarkably highly ordered atomic arrangement of sharp grain boundaries and coherent perovskite/PbI₂ interfaces, with a striking absence of long-range disorder in the crystal. These findings explain why inter-grain interfaces are not necessarily detrimental to perovskite solar cell performance, in contrast to what is commonly observed for other polycrystalline semiconductors.

Combining broad beam electron diffraction with 4D STEM (or scanning electron diffraction), I will also show why certain 2D perovskite materials are better than others at improving device performance and stability.

Our findings thus provide a significant shift in our atomic-level understanding of this technologically important class of lead-halide perovskites.

[1] Adv. Mater. 2018, 30, 1800629

[2] Nature communications 8 (1), 14547

[3] Nature Energy 6 (6), 624-632

[4] Science 370, eabb5940 (2020)

[5] ACS Energy Letters 9 (4), 1455-1465

Time-resolved optoelectronic measurements, including time-resolved photoluminescence (trPL) and transient surface photovoltage (trSPV), have emerged as powerful tools for investigating charge-carrier dynamics in photovoltaic materials and especially in perovskite research [1]. While trPL has been extensively utilized to study recombination and quantify losses, its interpretation remains challenging, particularly for interlayer stacks [2]. Similarly, trSPV provides complementary insights by tracking the photoinduced dipole [3]  and has received notable attention within the perovskite community.

Levine et al. [1] demonstrated that combining these techniques effectively distinguishes between charge extraction and recombination at the buried interface of perovskite and the selective contact in a half-cell device configuration. This distinction is vital for optimizing perovskite solar cells, as interfacial losses are the primary factor limiting their power conversion efficiency. Although their results are promising, their model requires further refinement, and a deeper fundamental understanding is necessary to develop a more advanced analysis routine. Such a routine would identify interface-limiting factors, enabling differentiation between transport-related issues and defects.

In this work, we use drift-diffusion simulations to refine the analysis of both techniques, ensuring a more accurate characterization of charge transport and recombination processes in half-cell devices. The simulations are performed with SIMsalabim [4] and incorporate critical physical parameters, such as charge carrier mobility and energetic band bending, overcoming limitations in previous models. Our results demonstrate that laser repetition rates significantly influence carrier dynamics, impacting transient signal interpretation. Furthermore, we highlight discrepancies between previous kinetic models and our findings, offering an improved theoretical framework for correlating these transient measurements with solar cell performance metrics. This approach allows a rigorous analysis of interfacial losses and efficiency optimization in emerging photovoltaic technologies.

In recent years, organic-inorganic lead halide perovskite-based solar cells (PSCs) have emerged as a potentially disruptive photovoltaic technology, offering high power conversion efficiencies (PCE) with low manufacturing costs and simple fabrication via solution processing. Despite the mixed ionic-electronic conductivity of perovskites and the high defect densities in the semiconductor expected from solution processing, PSCs can easily achieve high PCE values. The defects most likely to form mostly induce shallow energy levels close to the band gap edges, which relates to why perovskites are often said to have a notorious defect tolerance. However, abundant ionic defects are mobile and often critically affect the performance and stability of PSCs.

To improve stability and scalability of PSCs, fully mesoscopic carbon-based PSCs (CPSCs) have emerged as a promising architecture, which avoids the need for precious metal electrodes and organic hole transporting layers. However, their PCE still lags behind the record efficiencies achieved by other single-junction planar PSC stacks, highlighting the need for a deeper understanding of their limitations.

The effects of mobile ions in CPSCs infiltrated with methylammonium lead triiodide (MAPbI₃) are investigated through an approach combining experiments and simulations. The electric field screening effect of ionic charges is found to be visible in external quantum efficiency (EQE) measurements, which show how current collection losses depend on the illumination wavelength. By analyzing temperature and voltage dependent EQE spectra, we reveal how the internal energy landscape is shaped by the ionic distribution – which in turn determines the response observed in the current density-voltage (J-V) curve. The experimental results are reproduced by drift-diffusion simulations (including mobile ionic charges) coupled with an optical model. The findings shed light on the device physics of PSCs and provide new characterization approaches useful towards device optimization.

A substantial effort has been made over the past decade to develop lead-free alternatives to halide perovskites that could emulate their exceptional optoelectronic properties, but overcome their stability and toxicity limitations. Many of these perovskite-inspired materials are visible light absorbers, and a promising application is indoor light harvesting [1]. This talk discusses our recent work on two such materials for indoor photovoltaics: BiSBr and Sb₂S₃. BiSBr is novel light absorber [2]; Sb₂S₃ has been developed for solar cells, but its development for indoor photovoltaics hasjust begun [3]. We show both materials to be stable in air, and for Sb₂S₃, we achieve large-grained films with compact morphology through hydrothermal growth with monoethanoamine additives. By lowering the grain boundary density, we achieve increases in open-circuit voltage to 0.8 V under 1-sun illumination [2]. For IPVs, we achieve 17.55% power conversion efficiency under 3000 K white light emitting diode illumination (1000 lx), and create 5 cm²-area minimodules to demonstrate their applications in powering a multisensory platform used for IoT [3]. This places Sb₂S₃ IPVs among the most efficient for pnictogen-based semiconductors, and the devices were also stable for at least a month of testing without encapsulation [3].

To finish this talk, I will discuss the debate in the growing community working on emerging materials for IPV around a consensus in performance measurement. I focus on the open questions and possible solutions being proposed, as well as future pathways.

Halide perovskites have emerged as a versatile class of materials in optoelectronics and quantum materials research, owing to their exceptional tunability and rich interplay of photonic, electronic, spin, and lattice degrees of freedom. Their chemical flexibility enables the design of tailored interactions across a broad range of applications, from solar cells to spintronic devices. My team integrates first-principles approaches—including density functional theory, tight-binding models, and machine learning–accelerated molecular dynamics—to investigate the intricate structure-property relationships that govern perovskite behavior.

A major focus of our work lies in understanding and controlling defect chemistry, a key factor affecting the efficiency and long-term stability of perovskite solar cells. Through detailed electronic structure analysis and predictive modeling, we identify defect pathways and propose mitigation strategies, including compositional engineering and surface passivation techniques.

In parallel, we are advancing a novel research direction into the chirality of hybrid perovskites. By incorporating chiral organic ligands, we investigate how structural asymmetry can induce unique spin-dependent phenomena, such as chiral-induced spin selectivity (CISS), and enhance chiroptical responses. These effects open promising avenues for next-generation devices, including spin-polarized light-emitting diodes (spin-LEDs) and chiral photodetectors.

Hybrid organic–inorganic perovskites (HOIPs) have emerged as excellent materials for solar cell applications. Indeed, their extreme tunability and facile synthesis have opened the door to many new applications. Chiral HOIPs are attracting great interest as promising frameworks for chiroptoelectronics as well as spintronics applications.[1,2] The chiroptical properties observed in chiral HOIPs can be explained understanding the chirality transfer from the chiral organic molecules to the achiral inorganic octahedra. A key element of the chirality transfer mechanism involves the distortion of the coordination geometry of the inorganic octahedra induced by the presence of chiral ligands.

In this study, we propose a tailored simulation workflow based on Density Functional Theory (DFT) and Time-Dependent Density Functional Theory (TD-DFT)[3] to theoretically explore the chirality transfer mechanism and the chiroptical properties of chiral HOIPs. To this aim, we investigate the chiroptical response of lead- and tin-based 2D chiral perovskites, specifically 2D R- and S-(MBA+)2PbI4[4] and R- and S-(MBA+)2SnI4.[5] We explore the most impactful factors influencing their Circular Dichroism (CD) signals through ab-initio molecular dynamics simulations and the analysis of the density of electronic states (DOSs). Our findings reveal that the relevant chiroptical features are linked to a chirality transfer event driven by a metal–ligand overlap of electronic levels. This effect is more evident for tin-based chiral perovskites showing higher excitonic coupling.

Single crystal hybrid perovskites (SCHP) are promising materials for optoelectronic applications due to excellent properties such as homogeneity, reduced defects, lack of grain boundaries ^[1]. 2D SCHP show great potential for future applications as they offer advantages such as improved stability, improved carrier lifetimes, tuneable optical, electrical properties, significant light absorption coefficient ^[2,3]. Additionally, these materials offer multiple quantum well structures that results in improved exciton binding energy and strong quantum confinement effects, while large organic A cation further improves chemical and environmental stability ^[4].

In this presentation, we discuss recent results on the growth of lead-free thin 2D PEA₂SnI₄ and F-PEA₂SnI₄ single crystals grown using the space-confined approach. We investigate the structural, optical properties, and stability of the materials under environmental conditions by studying the change in the photoluminescence intensity as a function of time. Optical microscope images confirm the formation of 100-150 μm thick single crystal films of both PEA₂SnI₄ and F-PEA₂SnI₄ with emission band around ~632 nm. XRD analysis of both SCHP films showed regularly spaced diffraction peaks suggesting a homogenous crystal orientation. The stability studies as observed by the change of photoluminescence intensity as a function of time, revealed the instability of Sn perovskites, which could be due to oxidation under environmental conditions, more prominent in case of continuous light illumination. Further improvement in the stability could be achieved by implementing strategies for improving the materials stability, such as encapsulation or additive engineering.

Keywords: Sn single crystal films, 2D Sn perovskites, lead-free, stability

Mixed tin-lead (Sn-Pb) halide perovskites, with their tuneable bandgaps (1.2–1.4 eV), show great promise for the development of highly efficient all-perovskite tandem solar cells. However, achieving commercial viability and stabilized high efficiency for Sn-Pb perovskite solar cells (PSCs) presents numerous challenges.  Among various optimization strategies, incorporating additives has proven critical in modulating the crystallization of Sn-Pb halide perovskites. Despite the widespread use of additives to improve performance, the detailed photophysical mechanisms remain unclear. In this work, we elucidate the mechanistic role of guanidinium thiocyanate, a chaotropic agent, in the crystallization of Sn-Pb halide perovskites. We combine hyperspectral imaging with real-time in-situ photoluminescence spectroscopy, to study the crystallization process of Sn-Pb halide perovskites. Our findings reveal that the chaotropic agent modulates the crystal growth rate during perovskite crystallization, resulting in more homogeneous films with reduced non-radiative recombination. We challenge the common assumption that crystallization stops once the solvent evaporates by identifying photoluminescence variations during the cooldown process. The resulting films exhibit a photoluminescence quantum yield of 7.28% and a charge carrier lifetime exceeding 11 µs, leading to a device efficiency of 22.34% and a fill factor of over 80%. This work provides a fundamental understanding of additive-mediated crystal growth and transient cooldown dynamics, advancing the design of high-quality Sn-Pb perovskites for efficient and stable optoelectronics.

Undestranding the structure-property relationships at multiple length scales is of primary importance for rational design of new materials. Solid State Nulcear Magnetic Resonance (SSNMR), probing shot range (atomic scale) propeties, is complementary to diffraction techniques, which on the other hand give are sentitive to long-range structure. In recent years SSNMR emerged as characterization technique for Lead Halide Perovskites (LHPs) for its ability to study ion dynamics, compositional variations and ion incorporation, chemical interactions, and degradation mechanisms [1].

In this contribution, recent studies on multiple-cation lead mixed-halide perovskite Cs0.05FA0.81MA0.14PbI2.55Br0.45 [2], mono and multi layers 2D Ruddlesden-Popper (RP) phases containing butylammonium as spacer (BA2MAn-1PbnI3n+1 with n=1-4) [3], and CsPbBr3 nanocrystals [4] will be presented and discussed, showing how multinuclear SSNMR (133Cs, 207Pb, 13C, 1H) can provide detailed information on structural and compositional disorder, organic cation dynamics and surface properties of nanocrystals. Finally new developments on how isotopic enrichment can be used to study surface properties of powder and single crystal 3D LHPs will be shown.

The unprecedented fast rise of power conversion efficiency (PCE) of perovskite-based solar cells (PSC) in recent years has created a vast worldwide research activity in this material class for photovoltaic and other opto-electronic applications. Several materials compositions and device architectures have been described and best reported PCE’s yield recently more than 26%. Also improved stability under specific conditions has been shown for specific architectures. Whereas all these results indicate a high potential for this novel solar technology, further steps must be taken to convince industry and even the whole PV community that perovskite-based photovoltaics can really emerge from the lab into industrially applicable solar module processing. Our R&D program works actively on the upscaling of perovskite solar modules with scalable processes up to sizes of 35x35 cm².

Similarly, the perovskite PV technology has boosted the tandem research whereby perovskite cells and modules are placed on top of other PV devices like Si or CIGS solar cells. Impressive lab scale results exceeding 34% PCE have been reported. New challenges arise when this needs to be upscaled to full wafer or module size. It will be discussed how we approach these challenges.

Interface engineering via large organic ammonium cations has become a key strategy to enhance both efficiency and stability in perovskite solar cells (PSCs). These cations can form either thin molecular passivation layers or induce low-dimensional (2D) perovskite capping layers atop 3D perovskite absorbers—two structurally and electronically distinct configurations that differently affect device performance.

Self-assembled 2D perovskite layers effectively suppress interfacial recombination, boosting open-circuit voltage and operational stability. Bromide-based 2D perovskites further promote Br/I interhalide exchange, generating a compositional gradient from surface to bulk. This dual passivation—surface and bulk—enhances charge extraction and reduces non-radiative losses, enabling power conversion efficiencies up to 24.4% and excellent outdoor stability over 800 hours without degradation.

These results underscore the potential of ammonium cation-based strategies to go beyond defect passivation, enabling favorable structural and compositional tuning of the active layer. Further mechanistic insight and process optimization are critical to unlocking their full potential for scalable photovoltaic applications.

The production of halide perovskite layers under ultra-high vacuum (UHV) conditions, with real-time monitoring of their growth dynamics, is a promising way for addressing their longstanding instability issue. It is indeed expected that pristine layers, with high purity and near perfect crystalline order, represent an ideal playground for understanding the relative importance of intrinsic and extrinsic instabilities. For the vacuum deposition MAPbI₃, the volatility of methylammonium iodide (MAI) represents a significant challenge in achieving the right stoichiometry. In our study, we emphasized the crucial role of lead iodide (PbI₂) flux in the process. Using molecular beam epitaxy (MBE), we achieved the controlled growth of MAPbI₃ films on a graphene/SiC(0001) substrate; the growing layer is characterized in real time by correlating the optical properties and the growth dynamics.

To achieve the best control of the growth process, the growing layer is characterized in real-time by GIFAD (Grazing Incidence Fast Atom Diffraction), QMS (Quadrupole Mass Spectrometry), and SDRS (Surface Differential Reflectance Spectroscopy). In particular, GIFAD, a non-destructive technique provides fine information on the organization dynamics and can monitor the growth of the most fragile materials for hours without any damage [1]. SDRS yields information on the optical properties, which can then be correlated to the changes of the growing layer provided by GIFAD. GIFAD data reveal a layer-by-layer growth mode without evidence of crystallization in the early stage. SDRS measurements show the evolution of the optical features as a function of film thcikness. Notably, threee primary electronic transitions chracteristic of MAPbI₃ are observed at approximately 375 nm, 480nm, and 760 nm.

[1] A. Momeni et al., J. Phys. Chem. Lett. 9, 908, 2018

Atomic layer deposition of aluminum oxide (ALD-Al₂O₃) layers has recently been studied for stabilizing perovskite solar cells (PSCs) against environmental stressors and mitigation of pernicious halide ion migration from the perovskite towards the hole transport interface. [1] However, its effectiveness in preventing the infiltration of ions and additives from the hole-transport layer into perovskites remains insufficiently understood. [2]

Herein, we demonstrate the deposition of a compact ultrathin (<0.75 nm) ALD-Al₂O₃ layer that conformally coats the morphology of a triple-cation perovskite layer (see TOC). This promotes an effective contact of the hole transporter layer on top of the perovskite, thereby improving the charge carrier collection between these two layers. Upon systematically investigating the layer-by-layer structure of the PSC, we discovered that ALD-Al₂O₃ also acts as a diffusion barrier for the degraded species from the adjacent transport layer into the perovskite. In addition to these protective considerations, ALD-Al₂O₃ impedes the transition of crystalline perovskites to an undesired amorphous phase. [3] Using a combination of spectroscopy, diffraction, and microscopy techniques, our study contributes to a holistic understanding of decoupling the final device degradation from the instability caused by ion migration from the perovskite material and by the residues from the degraded charge extraction layer into the perovskite. [3] Consequently, the dual functionality (i.e., enhanced contact and diffusion barrier) of the ALD-Al₂O₃ protection enhanced the device stability, retaining 98% of its initial performance compared to <10% for pristine devices after 1500 h of outdoor testing (ISOS-O-L) under ambient conditions. Finally, this study deepens our understanding of the mechanism of ALD-Al₂O₃ as a two-way diffusion barrier, highlighting the multifaceted role of buffer layers in interfacial engineering for the long-term stability of PSCs.

Self-assembled molecules (SAMs) have been proven to be effective hole selective contacts in inverted perovskite solar cells (PSCs). Understanding the effect of molecular structure on the perovskite crystallization is vital to design SAMs for high-performance devices. In this work, we comparatively study four kinds of anchoring groups of carboxylic acid (-COOH), phosphonic acid (-PO(OH)₂), cyanoacetic acid (CN/COOH) and cyano/cyano (CN/CN) by modifying the EADR03 molecule to elucidate their effect on the perovskite crystallization and charge transfer at interface. We find that carboxylic acid, phosphonic acid and cyanoacetic acid could effectively bind with ITO to form an ultra-thin layer. Moreover, these anchoring groups can passivate the perovskite at the buried interface and affect the crystal growth. Among them, phosphonic acid promotes the growth of high crystallinity perovskite, resulting in a champion device with over 21% efficiency based on 1.61 eV bandgap perovskite and 24.0% efficiency for 1.55 eV bandgap perovskite without extra bulk and surface passivation. The transient optoelectronic characterizations under operational devices reveal that the high crystallinity perovskite reduces the density of trap states to suppress the interfacial non-radiative recombination, leading to a high V_OC in phosphonic acid SAM-based devices. This work reveals the passivation role of the anchoring group in SAMs and guides the rational design of the molecule for high-performance SAM-based devices.

One of the main components of p-i-n perovskite solar cells is the hole transport layer, for which self-assembled molecule called SAMs are used because due to its ability to create thin films on the nanometer scale, reducing losses that affect device performance. Different structures have been developed, among which those based on carboxylic acids have shown a good performance, this layer interacts with the ITO substrate through its anchor group to achieve a good performance of the device. In this work, we present a p-i-n device in which a new SAM was employed as a hole transport layer HTL deposited via spin coating. This molecule composed of triphenylamine and carboxylic acid has less hysteresis effect and high performance than the reference structure EADR03, with a high efficiency of 22.4% and fill factor of 82%. It is therefore capable of being considered as a reference for future photovoltaic devices due to its stability, high optical sensitivity and easy manufacturing.

Metal-halide perovskites (MHP) are highly promising optoelectronic materials due to their exceptional properties and versatile fabrication methods [1]. These materials are crucial for solar cells and optoelectronics devices to quantum emitters.

Thermal evaporation has emerged as a pivotal method to enable precise control over film thickness, composition tuning, stress-free deposition, surface modification capabilities, and large-area [2] devices—key factors in the advancement of perovskite optoelectronics.
My recent work has focused on overcoming the challenges of scalability and reproducibility in perovskite solar cell (PSC) production, critical for their commercial viability. In a recent work, we have demonstrated the possibility of achieving a sixfold increase in speed while maintaining film quality and high-power conversion efficiencies. This accelerated co-evaporation process demonstrates the potential for large-scale, cost-effective PSC production without needing post-annealing, further simplifying the manufacturing process [3].
Moreover, leveraging on the possibility of fabricating high-quality films with high thickness control by thermal evaporation we have explored the promising field of thermally evaporated perovskite-based Multiple Quantum Wells (MQWs) [4]. MQWs structures can enhance the optoelectronic properties of nanoscale thin films, through control over electronic energy levels and quantum mechanical phenomena, and opening up avenues for unconventional optoelectronic functionalities.
We demonstrated how both MAPbI3-based MQWs offer significant advancements in light-emitting and photodetection technologies, expanding their sensitivity into the near-infrared range and enhancing photoluminescence and charge separation efficiency.[5].
These recent works not only address critical challenges in scaling perovskite solar cell production but also open new avenues for optoelectronic device design, highlighting the versatility and promise of thermally evaporated perovskite materials for next-generation energy solutions.

References:
1) Min, H., et al., Nature, 2021. 598, 444.; Yoo, J.J., et al., Nature, 2021. 590 587.
2) J. Li et al., Joule 2020, 4, 1035; H.A. Dewi et al., Adv. Funct. Mater. 2021, 11, 2100557; J. Li et al., Adv. Funct. Mater. 2021, 11, 2103252;
3) Dewi et al. ACS Energy Lett. 2024, 9, 4319−4322
4) Advanced Materials 2021, 33, 2005166; L. White et al. ACS Energy Lett. 2024, 9, 83;
5) L. White, ACS Energy Lett. 2024, 9, 4450.

              Vapor phase deposition of perovskites is often considered a suitable deposition method for uniform deposition on large-area substrates with minimal upscaling losses. For this reason, it is assumed to be suitable for industrial-scale perovskite deposition. However, concerns regarding their practical deposition rates have entered the discourse, since deposition rates of ~1000 nm.min^-1 are typically considered necessary for the technology to be commercially viable. This is of particular concern for co-evaporation of perovskites, which is one of the most common vapor-based fabrication techniques, with typical deposition rates of 6 – 10 nm.min^-1 reported in literature.[1] Some research into MAPbI₃ has aimed to increase deposition speed without compromising device power conversion efficiency (PCE), achieving maximum deposition rates of ~26 nm.min^-1.[2,3] However, there has been no such research on formamidinium (FA)-based perovskites, or for wide bandgap perovskites suitable for tandem applications. A significant reason for this deficit is the propensity of FA to decompose at high temperatures, such as those required for elevated deposition rates, and a lack of understanding how this will impact device PCE.[4]

              In response, we present the first attempt to significantly increase FA-based perovskite deposition speed and analyze the impact of increasing deposition rate on device PCE. Our investigation reveals a negative correlation between deposition rate of FA based co-evaporated perovskites and device power conversion efficiency. Furthermore, high deposition rates will also negatively impact process repeatability. On a device level, the cause of these PCE drops are the emergence of μm-scale inhomogeneities in our perovskite film. Our investigation explores the root cause of these inhomogeneities, demonstrating that they are not the result of the classical degradation products of FAI, which we demonstrate using in situ mass spectrometry. Instead, temperature variations within the residual crucible material leads to the formation of coke products, which are subsequently deposited onto the perovskite as spit defects. 

              During our investigation, we analyzed multiple solutions to improve sample homogeneity and reduce PCE losses when increasing deposition rate, separated into two independent approaches. The first approach utilizes vertical scaling methods, which reduces the loss in PCE from ~23%_rel to ~9%_rel for devices with a ~18 nm.min^-1 deposition rate compared to a baseline of 5 nm.min^-1. The second approach utilizes preconditioning, reducing PCE loss from ~31%_rel to ~26%_rel for devices with ~21 nm.min^-1 deposition rate compared to a baseline 6 nm.min^-1. Our work represents the first attempt to meaningfully increase FA-based wide bandgap perovskite deposition rate while maintaining high efficiency and will facilitate future developments in this field.

To date, the stability of perovskite-based photodiodes under reverse bias has primarily been investigated in the context of solar cell applications. This condition commonly arises during partial shading, where a shaded cell is forced to conduct the current generated by its unshaded neighbors. The choice of electron transport layer (ETL) has been shown to play a crucial role in the breaking down mechanism. Unlike solar cells—where reverse biasing falls outside the normal operating range—photodetectors are designed to operate in this regime to achieve optimal signal-to-noise ratio and faster carrier extraction. However, the effects of prolonged reverse biasing on the performance of perovskite-based photodetectors remain largely unexplored, with most studies limiting their operating range to relatively safe levels, typically up to -0.5 V.

In this work, we investigate the impact of the ETL on the performance of all-inorganic p-i-n perovskite-based photodiodes (PePDs), fabricated exclusively through physical vapor deposition methods to ensure high thermal stability and scalable production. Specifically, we demonstrate that the use of a fullerene–metal oxide bilayer, along with careful tuning of their respective thicknesses, not only enhances reverse-bias stability, but also reduces performance variability and improves carrier extraction speed. The optimized photodiode exhibits a dark current below 0.1 μA/cm² even after hour-long biasing at -2 V, along with a >70% improvement in extraction speed—promising sub-μs rise times for further scaled-down pixels. These results pave the way for the development of reliable, all-evaporated perovskite-based imagers integrated atop silicon read-out circuits, offering ultra-high-speed performance and compatibility with fabrication processes that require a high thermal budget.

Furthermore, we utilize novel characterization techniques to gain insights into the mechanisms associated with the observed performance improvements. For instance, the enhancement in reverse-bias stability is attributed to a reduction in defect states at the perovskite/ETL interface, as revealed through a combination of transient photocurrent measurements with various excitation wavelengths and transfer-matrix algorithm simulations. Simultaneously, the improvements in response speed are further investigated through capacitance spectroscopy of both the photodiode stack itself and a simpler metal-oxide-semiconductor structure. Ultimately, the enhanced response speed is attributed to an extension of the depletion width into the ETL and a corresponding reduction of the equivalent RC constant.

The presented enhancements in the reliability and speed of all-inorganic, vacuum-deposited PePDs, coupled with the comprehensive understanding of their multifaceted performance in the reverse bias regime open new pathways for the development of CMOS-compatible perovskite photodetectors.

The rapid progress of perovskite solar cells (PSCs) with a current power conversion efficiency (PCE) near to 27%[1] is mainly attributed to the optimization of optoelectronic properties of the perovskite layer and the interfacial engineering of the charge-selective contacts. Among several strategies, the use of self-assembled molecules (SAMs) as selective charge transport layers has attracted interest owing to their potential to improve charge extraction, suppress non-radiative recombination, and minimize current leakage, which results in enhanced efficiency and stability of devices [2]. In this work, we have synthesized and characterized four dipodal indolocarbazole SAMs, used as hole-selective contacts in inverted PSCs based on the CsFAMA perovskite absorber (Eg = 1.6 eV). We investigate the effects of SAMs' structural variations, such as methoxy substitution in terminal functional groups and the length of alkyl spacers, on interfacial properties and device performance. To do so, the ITO/SAM and ITO/SAM/PSCs interfaces were characterized in detail. The results reveal the effects of indolocarbazole SAMs on the crystallization of the perovskite and charge dynamics in the devices. The resulting iPSCs showed PCEs between 19.76% and 22.20%, with fill factor exceeding 82% and good stability under continuous illumination. Notably, iPSCs using SAM with unsubstituted indolocarbazole and pentyl spacer (5CPICZ) exhibited the highest PCE of 22.20%. In contrast, devices using analogous SAMs with propyl spacers (3CPICZ) achieved a PCE of 22.01%. On the other hand, the iPSCs with methoxy-substituted SAMs showed reduced performance. Devices with SAM 3CPICZ-M (methoxy groups and propyl spacers) achieve a PCE of 21.51%, %, while those with SAM 5CPICZ-M (methoxy groups and pentyl spacers) yielded 19.76%. The observed PCE trend is attributed to the electronic and structural differences caused by the functional groups and spacer length. The experimental results reveal that, compared to their methoxy-substituted counterparts, the unsubstituted CPICZ SAMs more effectively passivate interfacial defects, reducing nonradiative recombination, leading to improved PCE. Transient optoelectronic measurements demonstrated that improved PCE in devices with unsubstituted CPICZ SAMs is owing to longer carrier lifetimes and reduced charge recombination [3]. The improved energy level alignment, enhanced hole mobility, charge transfer dynamics, longer carrier lifetime, and suppressed bimolecular/trap-assisted recombination all contribute to the highest performance of 5CPICZ-based devices. Overall, this study highlights the importance of molecular design and provides a pathway for developing efficient indolocarbazole-based SAMs for perovskite solar cell applications.

Silicon solar cells have dominated the science and market for solar cells. Their efficiency was boosted after the passivation of their surface and minimised charge losses at the interface, allowing better contacts to extract the photoelectrical current.

In molecular solar cells, either organic or hybrid perovskite-based photoactive layers, selective contacts are paramount to obtain high efficiency. In the latter case, using self-assembled molecules is a must to push the solar cell efficiency up to 24% (under standard solar irradiation conditions). Our group pioneered the design and synthesis of self-assembled molecules ( SAMs) in perovskite solar cells and has focused on studying the relationship between the SAMs' molecular structure and device function. I will discuss the latest advances in SAMs for perovskite solar cells and how far we can achieve the maximum theoretical limit in perovskite solar cells and tandem devices.

To fabricate and eventually scale highly efficient perovskite-silicon tandem solar cells it is crucial to deposit perovskite thin-films of high quality reliably onto industrial silicon solar cells. To do this we employ several evaporation techniques such as co-evaporation, sequential evaporation, and the hybrid route which combines evaporation with wet-chemical deposition techniques. Each of these has its own challenges and opportunities. In this talk I will share the insights we have gained into the growth mechanisms, crystallization behavior, and the influencing factors that govern the evaporation of different perovskites.

We implemented these perovskite films into tandem solar cells reaching efficiencies above 31% on 1 cm² cells. I will discuss the modeling and optimization steps that were taken to improve the whole tandem stack and achieve these results.  These high laboratory efficiencies now motivate our efforts to take these devices from laboratory to large-scale production. For this purpose, we are also developing and optimizing large area deposition methods for the contact and passivation layers of the solar cell. We are also pursuing device architectures that go beyond dual junction tandems to realize triple junction solar cells utilizing silicon and perovskite layers, which offer an even higher efficiency potential. I will present the progress we have made in that endeavor.  

GoPV is an Italian-based national project focused on studying new generation materials (absorbers and selective contacts) for perovskite and silicon heterojunction solar cells with the target of designing and realizing innovative perovskite/silicon and perovskite/perovskite tandem devices. Weaknesses in the emerging technologies of perovskite and Si heterojunction solar cells are being tackled. As for perovskite, organic-inorganic hybrid films are in use in single-junction and tandem-junction solar cells with record efficiencies. In view of a large-scale application of this technology, doubts related to the use of lead and organic groups need to be dispelled due to the possible effects on human health and the device stability over time, respectively. Therefore, inorganic perovskites, perovskites without lead or with reduced lead content, and, more generally, perovskite films with different bandgaps are being developed in the project. Regarding Si heterojunction cells, highly transparent selective contact materials are being evaluated as an alternative to doped silicon films deposited via PECVD, with the aim of minimizing parasitic optical absorption in the cell and evaluating less costly deposition techniques with fewer process safety issues. Transparent and conductive materials for the front contact of the devices are also being evaluated, aiming at reducing the use of scarce elements, such as indium contained in the commonly used indium-tin oxide.

In this presentation, the results obtained in the first period of the project will be presented and discussed. Various materials have been used to make different single-junction solar cells, such as dopant-free silicon heterojunction cells, low and high bandgap perovskite cells, lead-free perovskite cells, inorganic perovskite cells, etc. Preliminary results on perovskite/Si and perovskite/perovskite tandem cells will also be discussed.

Perovskite solar cells (PSCs) have shown great potential as a next-generation photovoltaic technology due to their high power conversion efficiency (PCE) and low manufacturing costs. However, the instability of PSCs, particularly when exposed to moisture, heat and light, has hindered their commercialization. Traditional metal electrodes, such as gold or silver, are not only expensive, but also prone to corrosion issues, further compromising the device stability. The use of carbon electrodes, can significantly improve the device stability under various environmental conditions, including humidity and high temperatures [1]. Carbon electrodes offer several advantages, including low cost, high conductivity, good hydrophobicity and electrochemical corrosion resistance, and compatibility with scalable printing techniques [2], [3].
This work focused on optimizing the perovskite-carbon electrode interface to enhance both the stability and performance of perovskite solar cells. To address this challenge, we explored various strategies, including interfacial engineering, perovskite composition modification and passivation techniques, aiming to reduce the impact of factors such as ion migration and non-radiative recombination that are known to negatively affect the long-term performance and operational lifespan of devices [2]. Besides the adoption of an n-i-p stack fully printed in air with an optimized formulation of FAPbI3 perovskite [4], we investigated the combination of a passivating agent, hexadecyltrimethylammonium bromide (HTAB), and a proper hole transport layer (HTL), poly(3-hexylthiophene-2,5-diyl) (P3HT) (Figure 1). This combination demonstrated excellent thermal stability for more than 600 hours, contributing to the overall robustness of the device. Further stability tests, such as light soaking, are under progress.

One of the simplest yet often overlooked methods for improving light management in Perovskite Solar Cells (PvSCs) is the incorporation of an Anti-Reflection Coating (ARC). In this work, we combine both optical modeling and experimental approaches to design and fabricate a hybrid bilayer ARC structure, consisting of a bottom planar MgF2 layer and a top tilted MgF2 layer. The concept behind this bilayer ARC design is to create a gradient in the refractive index between the two MgF2 layers. This gradient is achieved by tilting the deposition angle of the top MgF2 layer over the planar layer. Ellipsometric studies were conducted to determine the optical constants of both the planar and the tilted MgF2 layers. This ellipsometry data is then used to optimize the layer thickness and achieve a close match between modeling and experimental results, with an error of less than 1%. The bilayer ARC fabricated through a low-energy adatom-based physical vapor deposition technique exhibits more than a twofold increase in device current density compared to the single planar MgF2 layer. Specifically, the bilayer ARC structure with top MgF2 layers deposited at three different angles increases short-circuit current densities (Jsc) that are 2.75, 2, and 1.75 times the increment with planar MgF2 ARC device. With the optimized bilayer ARC, the fabricated PvSC achieved a Voc of 1.08V, Jsc of 24.6 mA/cm², fill factor (FF) of 83.4%, and a power conversion efficiency (PCE) of 22.3%. The improved performance of the bilayer ARC can be explained through various characterization techniques where ellipsometry measurements show that the tilted deposition of MgF2 results in a lower refractive index for the top MgF2 layer, forming a gradient structure. Additionally, AFM topography and cross-sectional SEM images confirm the presence of nanostructures, which facilitate light trapping and thus enhancing light transmission. Furthermore, contact angle measurements reveal that the optimized ARC exhibits a high surface contact angle of approximately 98°, which enhances its anti-soiling properties. This work presents an innovative approach to significantly improving the efficacy of the ARCs in PvSCs, potentially doubling their performance.

The power conversion efficiency (PCE) of single junction Pb-Sn perovskite solar cells has overcome the threshold of 24% [1]. Their use is especially promising for the fabrication of tandem solar cells to surpass the efficiency limits. However, rapid oxidation of Sn²⁺ cations during fabrication process and operation of solar cells affects the V_OC. In addition, the solution-based process of fabrication is complicated by non-uniform crystallization of the perovskite phase and very sensitive to changes in conditions, in particular temperature. In this work, to improve the morphology of Pb-Sn perovskite films, different methods of quenching (vacuum, N₂ gas flow, and antisolvent) during the spin-coating process were applied.  Furthermore, we present a comparative study of the influence of various additives (metal halides, thiocyanate salts, phenylethylammonium (PEA) halides and their fluorine-containing derivatives) on the performance of Pb-Sn perovskite solar cells with 1.26 eV bandgap. The surface treatment of perovskite films with ethylenediamine (EDA), EDAI₂, butane-1,4-diamine iodide (BDAI₂), PEAI, and 4F-PEAI solutions was also applied in comparison. The best combination of additives and surface passivators significantly enhanced V_OC and FF of the APb_0.5Sn_0.5I₃ solar cells (A=Cs, MA, and FA). The stabilized PCE of 18.6 % was achieved in the encapsulated p-i-n single junction solar cells with the structure of ITO/PEDOT:PSS/perovskite/PCBM/BCP/Cu (V_OC =0.87 V, J_SC =27.5 mA/cm², and FF=77.7%). However, the previously reported combination of additives (SnF₂, guanidine thiocyanate and PEAI) taken as a reference [2] provided only 16.45% (V_OC =0.80 V, J_SC = 27.4 mA/cm², and FF=75.1%). The obtained solar cells were applied as bottom subcells in the 4-terminal all-perovskite tandem cells, providing the PCE of >23 % in preliminary measurements.

Expeditious record-breaking power conversion efficiency of perovskite solar cells (PSCs) reaching 27% owes to the scientific mainstream devoted in understanding its intricate scaffold of designing the architecture.[1] The ease of processing and modifying the staked layers paves a pathway in realizing the targets of low-cost production, high efficiency, and better stability.  Most fabrication processes include the use of inert environments and high temperature.[2,3] Such conditions are majorly suitable for hard substrates curbing the possibility to transfer the technology on a flexible base which uses polymeric substrates or thin metal foils. Flexible PSCs has its potential in the integration on curved surfaces and light weight enabling applications in wearable electronics, portable power sources, and building-integrated photovoltaics (BIPV).[4] In addition, change of hazardous and toxic solvents used in the fabrication processes is the need of the hour and address alternative solvents while reporting the device performance. This will positively impact in large scale industrial production and commercialization in agreement with government norms.[5,6] Augmenting the panoramic research of PSCs, this work emphasizes 4 key areas in fabricating a flexible PSC. Starting from fabricating out of the inert environment, reducing high temperature processing requirement, scaling up from small area device to larger areas, and filling in a more sustainable and eco-friendly use of green solvents, this work demonstrates power conversion efficiencies of more than 21% on glass and 18.5% on flexible PET substrates. The results are creating a pathway in adding more understanding to the technology transfer from rigid to flexible substrates clearly marking the state-of-the-art in PSCs research.

Perovskite solar cells (PSCs) which are progressing at a rapid pace are still not suitable for commercialization due to the costly and time-consuming electrode deposition process. Even the most of high efficiency PSCs utilize costly electrode material such as gold (Au), silver (Ag), aluminum (Al) or copper (Cu). It is necessary to employ and develop new electrode material and deposition process. The printed molten metal electrodes (PME) were realised additively utilizing StarJet technology. Bulk metal alloy was printed with jet more directly on the substrates. With this new cheap and effective electrode material and deposition process, the PSCs achieved power conversion efficiency (PCE) ~ 11% under STC (1 sun, 1000 W/m² AM 1.5) and 16.72 % under indoor conditions (1000 lux, white LED). The PME based PSC had T₈₀ 13 hrs as compared to T₈₀ of 33 hrs for reference device (with Cu electrode) under continuous light soaking stability test under ISOS L-1 (1 sun, 1000 W/m² AM 1.5). Despite having lower PCE and stability of PME based PSC compared to reference PSC, the PME material and deposition technique altogether pave the path for large scale production and commercialization.

Perovskite solar cells(PSCs) have remarkably advanced, reaching power conversion efficiencies (PCE) over 26%, making them a promising candidate for next-generation photovoltaics ^[1]. However, most high-efficiency records have been achieved through spin-coating, a technique suitable for lab-scale but not for commercialization. Although cost-effective industrial methods such as coating and printing exist, these methods face challenges when scaled up to larger areas. This is mainly due to the increase of solvent consumption and the numerous hazards associated with highly toxic solvents like N,N-dimethylformamide(DMF), N-methyl-2-pyrrolidone(NMP), which raise significant safety and environmental concerns.

Recent advancements in green solvents within the perovskite research community have achieved PCEs comparable to DMF-based PSCs ^[2,3]. However, this research primarily focused on narrow bandgap perovskite formulations (<1.6 eV). This study addresses these concerns by employing inkjet printing as a scalable fabrication technique coupled with green solvents to process wide-bandgap (WBG) perovskites suitable for tandem applications. The environmental, health, safety, and toxicity indices of solvents are evaluated to determine the greenness of the solvent. A biomass-derived green solvent is utilized to formulate WBG perovskite ink (~1.68 eV). Due to the solvent's low donor number (DN), precipitation in the ink is observed when dissolving lead halides. Typically, WBG perovskites contain significant amounts of bromide and cesium ions, which are challenging to dissolve. Therefore, a solvent with higher DN must be used as a co-solvent to delay perovskite crystallization. The concentration of ink is carefully engineered to ensure its solubility and printability. Wetting and drying on the substrate are optimized by understanding the interaction between the surface and the ink. An additive in the bulk and a surface passivation at the perovskite and ETL interface enhanced the crystallinity and morphology of the perovskite and reduced surface recombination.

With these strategies, the green solvent-based inkjet-printed PSCs achieved efficiencies above 17%. To the best of our knowledge, is significantly higher than any previously reported PCEs for inkjet-printed wide bandgap PSCs fabricated with green solvents, indicating the potential of transferring this technology to silicon-based tandem solar cells. Our research offers practical strategies for a safer and more environmentally friendly scalable production of efficient perovskite photovoltaics.

With tandem solar cells (TSC) gaining increased attention in the solar community, the development of wide bandgap perovskites (WBPs) is becoming a key aspect of research activities. In particular, perovskites with a bandgap of 1.7eV have shown to be optimal candidates to be integrated with silicon cells and achieve the highest efficiency in TSCs. However, WBP still face critical challenges, primarily related to their poor thermal and light stability. Increasing the bandgap typically requires higher bromine (Br) content relative to iodine (I), which under illumination induces halide segregation into Br-rich and I-rich domains, causing efficiency losses and severely limiting device lifetime [1]. To mitigate halide segregation, studies have explored various approaches, including additive, composition and interface engineering [2]. However, the operational stability of WBP remains largely underexplored. Moreover, even though studies have reported some stability characteristics for WBP, from 1.63eV to 1.90eV [2], the majority emphasize performance metrics rather than identifying the fundamental degradation mechanisms. Furthermore, most of these studies have been conducted on perovskite films fabricated using lab-scale, non-scalable processes, limiting their relevance for real-world applications. Addressing these gaps requires a more systematic and comprehensive investigation into the stability of WBPs fabricated using scalable processes and device architectures.

In this context, the present work aims to systematically assess the stability of 1.68 eV WBPs fabricated using scalable deposition methods. Two wide-bandgap compositions are considered, one deposited by blade-coating (Cs_0.15(MA_0.2FA_0.8)_.85Pb(I_0.77Br_0.20)₃) and the second via a two-step hybrid process (CsFAPb(I_1-xBr_x)₃) and compared with a reference 1.6eV device processed by blade coating. Stability testing is conducted under two stress conditions as defined by ISOS protocols[3]. Maximum power point tracking (MPPT), following the ISOS-L2 protocol, is performed under operational conditions at 60°C. The time required for a device efficiency to reduce to 80% of its initial value (T80) is used as a key metric for comparing stability across different samples. Additionally, thermal stability is evaluated using the ISOS-D3 protocol, where devices are aged at 85°C in the dark under a nitrogen atmosphere. To monitor degradation, current-voltage (I-V) measurements are systematically performed in both forward and reverse scan directions, before and after each stress condition. Moreover, to gain deeper insights into degradation mechanisms, a comprehensive characterization toolbox is employed. Capacitance-frequency (C-F) measurements are carried out under applied bias voltage in both dark and illuminated conditions, enabling to correlate the performance degradation with intrinsic device parameters, such as interface trap densities and ion diffusion processes. Additionally, photoluminescence (PL) spectroscopy, including both steady-state and time-resolved photoluminescence (TRPL), is employed to evaluate changes in bandgap energy and other optoelectronic properties.

In a preliminary study, MPPT at 60°C was applied on these 3 different architectures. The best WBP sample maintains more than 95% of its initial efficiency after 12h, a remarkable result for such devices. However, this perovskite shows to be relatively sensitive to process variation, with another sample dropping to 75% of its initial performance in that period, with a T80 of 254min, lower than the 720min obtained for the hybrid WBP device. As expected, the reference 1.6eV device shows very good stability, reaching only 90% of its initial efficiency after 30h. PL steady-state measurements show an increase in the bandgap and FWHM enlargements in some regions of the cell for both low stability WBP devices after MPPT. Additionally, C-F measurements show an increase of capacitance after degradation, highlighting the increase in mobile ions concentration. Collectively, these characterization techniques provide a comprehensive assessment of the stability of wide-bandgap perovskite solar cells under thermal and operational stress conditions.

Amplified spontaneous emission (ASE) has been reported in different kinds of MHPs, including MAPbBr₃, a bright emitter in the visible-green range, which is the focus of this study. When the material is optically pumped, when the excitation fluence exceeds a certain threshold, ASE appears as a narrow peak on the low-energy side of the spontaneous PL spectrum that, when threshold is reached, grows that grows superlinearly with excitation fluence. Despite being widely reported, the origin of ASE is still debated, being unclear whether ASE arises from direct stimulated photon emission or stimulated emission of exciton-polaritons.

Here we report the results of a new experimental approach, based on tandem ultrafast spectroscopy, to investigate the microscopic origin of ASE in MAPbBr₃ planar waveguides consisting of a 100-nm-thick perovskite layer on top of a glass substrate and covered with an antireflective PMMA coating. To this aim, we configured a femtosecond spectroscopy station to measure both transient transmission and reflection spectra, from which transient absorption coefficient (𝛼) spectrograms were determined, allowing to detect photon gain with unprecedented sensitivity, down to a few tens of cm⁻¹. This result was complemented by time resolved photoluminescence (TRPL) measurements monitoring the coherent light emission from the same excitation spot as the transient pump-probe setup, with comparable time resolution.

ASE was found to occur without population inversion and photon gaina and, even at high excitation densities above threshold, negative 𝛼(𝜔) coefficient was never observed. Just above the threshold, a clear exciton peak was present, with a cross-section only slightly reduced compared to the unpumped waveguide. These results indicate that ASE mainly involves the stimulated emission of exciton-polaritons rather than photons. To confirm our hypothesis, we also solved Maxwell’s equations for the waveguide structure using the measured optical permittivity, finding that light-matter interaction in the highest excitation regime leads to hybrid modes with an energy-momentum dispersion below the bandgap, like that of the lower exciton-polariton band.

In summary, our results show that ASE happens without optical gain, it involves exciton-polaritons at low excitations and new hybrid states at high excitations, that we call band-edge polaritons. Our findings represent an important step toward understanding the fundamental processes governing light amplification in hybrid perovskites that, owing their compatibility with nanopatterning techniques, makes extremely promising candidates for integrated energy-efficient light sources.

As perovskite solar cell (PSC) stability expectancy surpass 1000s of hours, the cost and effort to detect instability in PSC becomes increasingly higher. Accelerated testing could offer a solution by reducing resources needed to advance in preventing the initial stages of degradation, especially if properly related to long-term studies and by use of semi-automated analysis. By suitable experiments, it becomes feasible to track physical and chemical phenomena occurring during the application of stress, and consequences for stability of reversible processes such as ion migration. [1] We have developed in-situ characterization methods to be applied in line with or parallel to ISOS-grade indoor and outdoor operational testing. These include X-ray diffraction (XRD), electrochemical impedance spectroscopy (EIS) and photoluminescence (PL). This presentation includes implementations of accelerated stress tests on PSCs under operando conditions, among others applying continuous light irradiation and bias-voltage. The device response with time under bias via in-situ XRD analyses in conjunction with quasi in-situ EIS, demonstrated a clear relation between halide perovskite lattice expansion/constriction, increased ionic motion, current decay with time and device stability. [2] EIS analyses revealed a threshold where some of these bias-induced degradation mechanisms become irreversible. By a combination of detailed characterization, we could elucidate mechanisms of actuation when using additive engineering to passivate shallow defects. [2-4]

Perovskite solar cells (PSCs) combine high efficiency, low-cost processing, and lightweight design, making them promising for both terrestrial and space applications; however, their environmental instability remains a key limitation. This work outlines an integrated approach to assess and improve PSC outdoor stability through materials screening, machine learning, field testing, and stratospheric trials. Nine commercial epoxy resins were systematically evaluated to identify formulations that minimize chemical interactions with halide perovskite layers. Two optimal formulations were used for ambient-temperature encapsulation, enabling encapsulated PSCs to retain over 95% of their initial efficiency after 1,500 hours in diverse outdoor conditions across Spain, Israel, and Germany¹. To accelerate outdoor stability assessment, we developed a machine-learning framework trained on accelerated indoor ageing data (temperature, light soaking, air exposure) that accurately predicts outdoor performance. By training these algorithms with various sets of indoor stability data, we can identify the most significant stress factors, thus providing insights into outdoor degradation pathways². Finally, to explore PSC resilience under extreme environments, we performed in-situ stratospheric stability tests³. Triple-cation devices endured the extreme conditions (–32 °C to +32 °C, high UV, low pressure), retaining 68 %–87 % of their initial power performance over 10 hours of flight time. The experiment allowed to test our encapsulation method under these conditions and demonstrated that it is able to survive the 10 h that the experiment lasted.