First, I will show that thick methylammonium lead iodide single-crystal films operated in X-ray photovoltaic mode can overcome one of the central limitations of perovskite detectors: instability under high external bias.[1] At zero applied voltage, these devices achieve single-photon-sensitive X-ray detection, a detection efficiency of 88% at 18 keV, a noise-equivalent dose down to 90 pGy_air, spatial resolution up to 11 lp mm^-1, and operational stability exceeding one year. I will then discuss the extension of this concept to mixed-cation MAFA lead iodide single crystals, where improved charge transport enables millimetre-thick absorbers suitable for the full medical diagnostic energy range. These devices reach mobility-lifetime products up to 0.2 cm² V^-1, diffusion lengths of about 3.5 mm, detective quantum efficiency above 90% at 60 keV and 50% at 120 keV, and record perovskite-imager spatial resolution of 19 lp mm^-1.

Finally, I will connect these detector advances to radiovoltaic energy conversion. Using solution-grown MAFAPbI₃ single crystals, X-ray/gamma-voltaic cells reach a power conversion efficiency of 10.1% under X-ray irradiation, approaching the calculated Shockley-Queisser limit for this radiation regime and demonstrating the feasibility of low-cost perovskite-based nuclear batteries. Together, these results show that perovskite single crystals are not only promising detector materials, but a broader platform for converting hard radiation into useful electronic signals and electrical power.

Single-crystal halide perovskites provide a powerful platform for probing intrinsic structure–property relationships and for translating controlled crystal growth into high-performance optoelectronic and emerging devices. In this invited talk, I will use our work on in situ organic cation chemistry to illustrate how precursor reactivity can be used to grow and engineer hybrid halide perovskite single crystals. In this approach, amide solvents such as N-methylformamide and N-dimethylformamide act not only as solvents, but also as sources of organic ammonium cations generated under acidic conditions. This chemistry first enabled the room-temperature growth of bulk methylammonium lead halide crystals without relying on pre-synthesized organic ammonium halide salts [1].

The talk will focus on how this approach links precursor chemistry to crystal growth, compositional control, defect formation, lattice strain, surface quality, and device response. I will discuss mixed dimethylammonium/methylammonium lead bromide crystals as a case study in A-site organic-cation engineering, where cation mixing suppresses structural phase transitions, modifies organic-cation dynamics, and enhances photodetection [2]. Related studies further reveal that precursor composition and concentration can promote the formation of metallic lead during crystallization, highlighting the need to understand solution chemistry as an active contributor to defect formation in halide perovskite single crystals [3].

Building on this link between crystallization chemistry and optoelectronic response, I will also discuss methylammonium lead bromide single crystals with compressed emissive edges, produced by manipulating in situ cation generation and the crystallization process. This example shows how spatially non-uniform crystallization can create persistent local lattice compression within a single crystal, leading to edge-selective emission and improved photodetector response [4]. Finally, I will discuss surface-engineered MAPbBr₃ crystals in which an in situ-grown polycrystalline passivating layer reduces surface defects, improves X-ray detection, and enhances photodetector performance [5,6].

As a brief outlook, I will also show how related in situ organic cation chemistry can be extended to lead-free layered copper halides, where local structure, halide vacancies, and ion migration govern resistive switching behavior [7]. Together, these studies show that in situ organic cation chemistry offers more than an alternative synthetic route. It provides a chemical framework for designing halide perovskite single crystals and related crystalline materials with controlled defects, engineered surfaces, and device-relevant functionality.

Single-crystal halide perovskites provide an excellent platform for investigating the intrinsic electronic properties of metal halide semiconductors while eliminating many of the complexities associated with polycrystalline films. However, translating these materials into high-performance devices requires a detailed understanding of their surfaces, buried interfaces, and their response to intense X-ray irradiation during advanced characterization.[1]

In this contribution, we present a methodology based on synchrotron soft- and hard X-ray photoelectron spectroscopy to investigate the electronic structure of clean single-crystal halide perovskites.[2] By combining in situ crystal cleavage with the sequential evaporation of charge-transport materials and metallic contacts under ultra-high vacuum, we have developed a platform to fabricate clean interfaces and monitor their formation in real time. This approach enables direct observation of energy-level alignment, band bending, electronic structure evolution, and interfacial chemical reactions during interface formation, including buried interfaces relevant to photovoltaic and optoelectronic devices.[3,4] We also discuss the influence of X-ray exposure on the electronic structure and chemical stability of single crystals, highlighting practical considerations for obtaining reliable spectroscopic data.[5] Together, these studies demonstrate how advanced synchrotron characterization can provide unique insight into the fundamental mechanisms governing charge extraction, interface formation, and device performance in single-crystal halide perovskites, while establishing experimental guidelines for the characterization of these beam-sensitive materials.

The development of new technologies has always been accompanied by access to functional materials with targeted, exceptional properties. Looking towards the future, layered hybrid organic-inorganic metal halide perovskites (HOIPs) are poised to play a significant role in optoelectronic, spintronic, and quantum technologies due to their tunable bandgap, high carrier mobility, strong spin-orbit coupling and magnetic ordering.^[1-4] However, for their successful integration into devices and development of new applications, understanding the relationship between composition, crystal structure, and optical and magnetic properties, as well as how to control them, is key. Indeed, HOIPs incorporating transition metals are an ideal platform for tuning magnetic (spins) and even optical (photons) due to their chemical and structural versatility.^[1,5–8] In this work, we explore the effects of magnetism in the photoluminescence (PL) emission originating from spin-forbidden d-d transitions^[7,8] of antiferromagnetic Mn²⁺ HOIPs with different crystal lattice parameters. Using temperature-dependent PL and PL excitation (PLE) spectroscopy, we demonstrate the complexity of the magneto-optical coupling in these materials and provide new insights into this field. We reveal significant changes in the PL/PLE trends prior to reach long-range magnetic order. Our results highlight the interplay among the magnetic polarons, intrinsic magnetism and crystal structure as the origin of these changes, providing new insights into spin-optoelectronics.^[9]

Halide perovskites are remarkable optoelectronic semiconductors because their optical response is strongly linked to crystal structure, dimensionality, defects, and dynamic disorder. While bulk single crystals provide access to intrinsic material properties, low-dimensional single-crystalline perovskite nanocrystals offer a complementary platform in which size, shape, and surface chemistry can be precisely varied and correlated with optical performance. In this talk, I will discuss how advanced synthesis, in situ characterization, and data-driven analysis can reveal structure–property relationships in halide perovskites from the single-particle to the ensemble level.

I will first present our recent work on the formation of anisotropic CsPbBr₃ nanocrystals, where simultaneous in situ X-ray scattering and photoluminescence spectroscopy revealed how emissive crystalline intermediates evolve into either one-dimensional nanorods or two-dimensional nanoplatelets [1]. These results show how antisolvent properties and ligand-mediated assembly pathways determine crystal dimensionality, structural order, and emission energy.

Building on this mechanistic understanding, I will discuss a chemistry-aware machine-learning framework for precision control of perovskite nanocrystal growth [2]. By combining automated synthesis, high-throughput optical characterization, Gaussian-process regression, and Bayesian optimization, this approach enables data-efficient navigation of complex synthesis spaces and yields nanocrystal ensembles with targeted emission properties and improved homogeneity.

The second part of the talk will focus on how optical spectroscopy can be used to extract intrinsic information from perovskite nanocrystals. I will introduce a statistical framework that links asymmetric photoluminescence lineshapes to exciton thermodynamics, energetic disorder, and size dispersion [3]. Finally, I will show how combined interferometric scattering and photoluminescence microscopy enables high-throughput determination of size and quantum yield for thousands of individual CsPbBr₃ nanocrystals in situ [4]. This single-particle approach uncovers heterogeneity, size-dependent defect passivation, and light-induced degradation pathways that are hidden in ensemble-averaged measurements.

Together, these studies establish a quantitative framework for connecting crystal growth, structural heterogeneity, defects, and optical functionality in halide perovskites. They highlight how low-dimensional perovskite single crystals can serve as model systems for understanding the microscopic mechanisms that ultimately limit performance, stability, and reproducibility in perovskite optoelectronics.

Mechanical strain offers a powerful, non-chemical route to tune the structural and optoelectronic properties of halide perovskites, but capturing strain-induced phase transitions with first-principles accuracy at experimentally relevant length and time scales remains challenging. Here, we use molecular dynamics driven by machine-learned interatomic potentials (MLIPs) trained on-the-fly to investigate the response of MAPbI₃ to biaxial strain. By modelling thin films as finite-thickness surface slabs, we resolve how strain is accommodated differently in the bulk-like interior versus the surface layers. We find that tensile biaxial strain drives a tetragonal-to-cubic phase transition, consistent with recent experimental reports of epitaxially stabilized cubic MAPbI₃ thin films grown under tensile strain [1]. We link this structural transition to systematic changes in the electronic band gap, and show that the surface layers exhibit a distinct strain and phase response from the bulk. Our results demonstrate the power of MLIP-driven molecular dynamics for resolving strain-induced phase transitions in halide perovskites beyond the reach of conventional ab initio approaches, and offer microscopic insight relevant to strain engineering in single-crystal and thin-film perovskite devices.

Hybrid lead halide perovskites (HLPs) have emerged as one of the most important classes of materials in the field of photovoltaics due to their exceptional optoelectronic properties and compositional flexibility. Yet, the thermal and chemical material instability, along with the toxicity of lead cations, remain open problems that require a precise control and understanding of the crystal growth and microstructure at the atomic scale. In this talk, we will discuss recent progress in physics-based models [1,2] as well as advanced machine learning approaches [3] for the large-scale molecular dynamics simulations of bulk, surfaces and interfaces. Specifically, we will report on the development of a new MYP2[4] model that, by the introduction of many-body interatomic terms, enables the simulation of crystal growth. This provides information on kinetics, activation energies, morphology and defects formation[5,6] while also allowing the study of complex 2D/3D interfaces and heterostructures in solution. We will conclude with a perspective on the evolving role of molecular simulation in the era of machine learning.

Layered heterostructures, formed one monolayer at a time, have been shown to host emergent properties at the interfaces where the monolayers meet. Although impressive advances have been made in controlling these exquisite architectures, these approaches do not readily lead to inexpensive, scalable, and reproducible syntheses of well-defined interfaces. This motivates the development of one-pot syntheses of heterostructures as crystalline solids, where the precise 3D atomic coordinates of the interfaces can be resolved through single-crystal X-ray diffraction. We recently reported a synthetic strategy for the self-assembly of layered perovskite–non-perovskite heterostructures into single crystals in solution, using bifunctional organic molecules as structure-directing agents [1]. I will present various layered heterostructures synthesized in our labs and discuss how the interleaving of two different inorganic layers can markedly transform the electronic band structure [2], optical properties [3], and low-dimensional magnetism [4]. 

Two-dimensional (2D) metal halide perovskites are emerging materials with bright and tunable emission that are highly attractive for photonics. Their hybrid organic-inorganic structure provides ample opportunities to tailor their optical properties and radiative recombination dynamics.[1, 2] Strong dielectric confinement results in high exciton binding energies, and their high refractive index favors strong-light matter interaction.[3, 4] In particular, highly homogeneous slabs of 2D perovskites can act themselves as optical cavities and lead to self-hybridization of the photonic cavity modes with the band edge exciton levels. The strong non-linearity of the complex refractive index inside the self-sustained cavity leads to low group velocity of the light and an abnormal dispersion near the band edge that results in high damping and leaky optical modes.

We investigate the reflectance and photoluminescence of single 2D lead iodide microcrystals that act as highly efficient optical cavities. We observe an extremely large number of cavity modes in microcrystals with micrometer thickness. This allowed us to determine the complex refractive index using the interference fringes method combined with a Cauchy dispersion for the transparent low energy range of the cavity mode spectrum. We then use the obtained dispersion of the refractive index in the calculation of the exciton-polariton anticrossing based on a Lorentz oscillator model, and by fitting the experimentally obtained reflectance minima near the band gap we obtain the salient optical parameters of the emitter material. We further observe additional mode splitting that will be discussed in terms of exciton-phonon coupling.

Our experiments show that 2D perovskite microcrystals with highly regular shape and homogeneous thickness constitute a versatile platform for integrated and non-linear photonics

References

[1] R. Krahne, et al., Nano Lett., 24, 11124-11131 (2024).

[2] R. Krahne, et al., Acc. Chem. Res., 57, 2476-2489 (2024).

[3] M. Borreani, et al., Adv. Opt. Mater., 13, e01276 (2025).

[4] A. Schleusener, et al., Adv. Mater., 36, 2402924 (2024).

Metal halide perovskite single crystals have emerged as a promising platform for integrated nonlinear photonic devices. Their large exciton binding energies, strong oscillator strengths, and compatibility with micro- and nanofabrication have enabled the realization of room-temperature polaritonic waveguides, microcavities, and interferometers. 

The fabrication of such photonic elements requires the development of geometrically confined crystallization methods capable of providing precise control over crystal morphology. In this context, this talk explores different approaches, including microfluidic growth and dewetting-assisted growth, which combine precursor-solution confinement with fine control over nucleation and crystallization pathways. By optimizing the growth conditions and tailoring interactions within the precursor solution, these methods enable the realization of a wide variety of geometries, including wires, square pixels, and interferometric structures.[1-3] In addition, these approaches allow accurate control over crystal placement on the substrate, which is essential for device integration.

While these approaches address the fabrication challenges associated with integrated photonic architectures, the practical implementation of metal halide perovskites is still limited by their intrinsic instability. Continuous or high-power optical excitation can induce photo-induced degradation and chemical instability, compromising device reproducibility and long-term operation. Developing effective defect-passivation strategies is therefore essential to move from proof-of-concept demonstrations to reliable integrated photonic devices. 

The second part of this talk focuses on the use of ionic liquids as an effective passivation strategy to enhance both optical quality and material stability. In particular, the incorporation of the 1-butyl-3-methylimidazolium (BMIM)-based ionic liquid during crystal growth leads to higher photoluminescence intensity, longer excited-state lifetimes, and reduced optical losses, providing a promising route toward more efficient and robust integrated photonic circuits.

Lead halide perovskites (LHP) are promising materials for direct hard radiation detection, a field that requires high-quality crystals. Among them CsPbBr₃, a stable material containing heavy Cs⁺ cation,  shows the best electronic performance when grown from the melt by the Bridgman-Stockbarger technique. While phase purity of the crystalline CsPbBr₃ ingots is often confirmed by X-ray diffraction, quality assessment relies on optical microscopy and eventually device performance – a costly, iterative and irreversible process that requires full device fabrication. Furthermore, the orthorhombic crystal structure of CsPbBr₃ at room temperature, with distinct lattice parameters, makes reliable determination of crystallographic alignment in melt-grown crystals particularly important, yet remains challenging. To address these limitations, we establish nuclear quadrupole resonance (NQR) spectroscopy as a versatile, non-invasive technique for evaluating the quality of melt grown CsPbBr₃ ingots prior to the first device fabrication step. We demonstrate that, beyond probing the local environment around a quadrupolar nucleus, NQR spectroscopy is inherently sensitive to crystallinity and crystal orientation. We correlate key spectroscopic descriptors, such as linewidth and integral, with macroscopic and microscopic structural features, to establish a robust evaluation of the crystal orientation and quality for a guided sample selection. Decreasing crystallite-domain size leads to the pronounced broadening of the observed NQR signal, with the full width half maximum differing by up to 70 kHz between pristine melt-grown crystals and polycrystalline powder. We further support the experimental results by ab initio calculations of bromine NQR frequency distributions, which capture the effect of structural disorder on the NQR signal. Customized resonators allowed us to accommodate large ingots directly within the quartz ampules used for growth and enabled semiautomated spatial mapping of the spectroscopic descriptors across the ingots. We show that the removal of the impurities collected near the top of the ingot and subsequent recrystallization improve the homogeneity and overall crystallinity of the samples, highlighting the need for multiple purification steps. We also observed reorientation in crystal-domain alignment along the ingot which is preserved after cutting and polishing. Our findings deepen the understanding of both applied NQR spectroscopy and melt-growth processes of lead halide perovskites, paving the way for integrated quality control and innovative in situ applications.