Excitons, neutral quasiparticles formed by electron-hole pairs, play a key role in the optoelectronic properties of semiconductors. Understanding their formation, transport, and dissociation is essential for interpreting experiments, predicting material behavior, and designing new materials for targeted applications. Low-dimensional halide perovskite semiconductors provide a versatile platform for studying excitons due to their structural tunability and facile fabrication. Quasi-two-dimensional (2D) halide perovskites, consisting of metal-halide octahedral layers separated by organic spacers, are particularly promising. Their unique structure, which disrupts octahedral connectivity in one direction, results in anisotropic charge-carrier masses and dielectric screening, promoting the formation of strongly bound excitons. First-principles calculations of excitonic properties in these materials have been limited by the large unit-cell sizes of most experimentally synthesized quasi-2D perovskites. However, recent advances in hardware and many-body perturbation theory methods, such as the GW and Bethe-Salpeter Equation approaches, now enable detailed insights into these systems. In this presentation, I will showcase how these methods allow for a microscopic understanding of the emergence of intra-, interlayer and charge-transfer excitons and their coupling to lattice degrees of freedom in low-dimensional halide perovskites. Our calculations provide predictive accuracy, explain experimental observations, and open pathways for tuning excitonic properties in these complex, heterogeneous materials.

Metal halide perovskites have emerged as highly promising materials for optoelectronics and spintronics over the past decade. However, a complete understanding of their fundamental physical characteristics, such as thermal bandgap evolution, spin relaxation mechanisms, and transport properties, remains elusive. This knowledge gap is currently hindering the development of advanced pervoskite-based devices.

In the first part of the talk, we systematically studied ultrafast spin coherence, spin relaxation and bandgap revolution with temperature in two hybrid organic-inorganic perovskites MA_0.3FA_0.7PbI₃ and MA_0.3FA_0.7Pb_0.5Sn_0.5I₃. We observed contrasting spin lifetimes between the two samples, suggesting that the spin relaxation is likely due to scattering with defects via the Elliot-Yafet mechanism at low temperatures and the spin decoherence suffers from g-factor inhomogeneity due to impurities and vacancies. By measuring carrier spin lifetimes at elevated temperatures, we specify possible roles of defects and phonons in the spin relaxation channels. Our temperature-dependent experiments revealed drastic changes in both electron and hole Landé g-factors. We propose that this effect is dominated by the enhancement of dynamic lattice distortions (lattice vibrations) with increasing temperature, resulting in strong modifications of not only the bandgap but also the interband transition matrix and the spin-orbit splitting gap. [1,2]

In the second part of the talk, we directly monitored exciton diffusive transport from low temperature to room temperature using high-purity CsPbBr₃ single crystals and contact-free transient grating spectroscopy. We then converted the diffusion into an effective exciton mobility (\mu) using the Einstein relation. As the temperature (T) increases, the mobility decreases rapidly below 100 K with a scaling \mu~T^-3.0, and then follows a more gradual \mu~T^-1.7 trend at higher temperatures. Our first-principles calculations perfectly reproduce this experimental trend and reveal that optical phonon scattering governs carrier mobility shifts over the entire temperature range, with a single longitudinal optical mode dominating room-temperature transport. [3]

Our findings unambiguously resolve previous theory-experiment discrepancies, providing benchmarks for the future design of optoelectronic and spintronic perovskite devices.

In recent years, chiral materials have garnered significant attention due to their promising applications in optoelectronics, chemical sensing as well as quantum computing [1-3]. The chiral characteristics of both soft materials and inorganic systems offer valuable insights for enhancing their functional integration. Notably, hybrid materials have emerged as a rapidly growing area in materials science, especially in optoelectronics, as they allow fine-tuning of the properties inherent to both soft and inorganic components. Among these, chiral hybrid perovskites have stood out as a particularly compelling class, exhibiting strong circularly polarized emission without the need for costly ferromagnetic materials or extremely low temperatures. Additionally, they demonstrate intriguing chirality-induced spin selectivity (CISS) effects [4]. The chiral source influences specific non-covalent interactions within the scaffold, which in turn modulate the efficiency and expression of chiral properties [5]. Owing to advances in multiscale modeling and simulation, it is now possible to design chiral systems with unprecedented accuracy. In this talk, I will present recent contributions in predicting chiral behavior in chiral hybrid perovskites [6-8] at high pressure conditions. I will introduce novel chiral design strategies that integrate enhanced sampling simulations with time-dependent density functional theory (TD-DFT) calculations derived from computed free-energy landscapes. This approach accounts for various contributions -such as molecular rotations within the chiral framework - that can critically impact the emergence and optimization of chiral properties.

References

[1] J. Crassous, M.J. Fuchter, D. E. Freedman, N. A. Kotov, J. Moon, M.C. Beard, Nat. Rev. Mat. 8(2023) 365.

[2] G. Albano, G. Pescitelli, L. Di Bari, Chem. Rev. 120 (2020) 10145

[3] S. Jiang, N. A. Kotov, Adv. Mater. 35 (2023), 2108431.

[4] H. Lu, C. Xiao, R. Song, T. Li, A. E. Maughan, A. Levin, R. Brunecky, J. J. Berry, D. B. Mitzi, V. Blum and M. C. Beard, J. Am. Chem. Soc. 142 (2020) 13030.

[5] A. Pietropaolo, A. Mattoni, G. Pica, M. Fortino, G.Schifino, G. Grancini Chem 8 (2022) 1231.

[6] M. Fortino, A. Mattoni, A. Pietropaolo, J. Mater. Chem. C 11 (2023) 9135.

[7] M. Fortino, A. Mattoni, A. Pietropaolo, J. Phys. Mater. 7 (2024) 045009.

[8] M. Fortino, G. Schifino, M. Salvalaglio, A. Pietropaolo, Nanoscale 17 (2025) 5823.

Lithium-rich manganese-based layered oxides are promising cathode materials for next-generation lithium-ion batteries, offering exceptionally high energy densities through combined transition metal and oxygen redox. However, this high energy density presents a critical limitation: these materials suffer progressive loss of energy density upon cycling, due to progressive decrease in average voltage; a phenomenon termed ‘voltage fade’ [1–4]. Understanding and controlling the underlying mechanisms of voltage fade are essential to realise the full potential of these high-capacity cathode materials.

Voltage fade has been linked to the formation and growth of nanoscale voids within the cathode bulk [1], but the atomic-scale mechanisms of this process are not well understood. The conventional approach for modelling battery cathode materials at the atomic scale is density functional theory (DFT). However, DFT cannot be used to directly investigate nanoscale void formation and growth, because the necessary system sizes are too large to be computed.

To investigate void formation over extended cycling, we have developed a novel computational approach combining DFT calculations, cluster expansion models, and Monte Carlo simulations. By applying this methodology to Li-rich Mn-based cathodes across the Li₂MnO₃–LiMnO₂ compositional space, we find that nanoscale voids form through two concurrent processes: formation of O₂ molecules within the bulk and extensive transition metal migration that forms transition-metal-deficient regions via phase segregation. Under extended cycling, these voids coalesce, driven by surface energy minimisation, in a process analogous to Ostwald ripening.

We further find that void coalescence—and thus voltage fade—depends strongly on the initial Mn/Li configuration in the Mn-rich layer, suggesting that targeting specific initial structures can inhibit deleterious structural evolution during cycling. By establishing the direct link between void growth and voltage loss, we show that preventing coalescence offers a route to maintaining electrochemical performance. Through systematic mapping of voltage fade across the Li₂MnO₃–LiMnO₂ compositional space, we identify optimal structures and compositions that minimise degradation whilst retaining high energy density. These findings establish clear structural and compositional design principles for developing Li-rich cathodes with sustained performance over extended cycling.

[1] McColl, K.; Coles, S. W.; Zarabadi-Poor, P.; Morgan, B. J.; Islam, M. S. Phase Segregation and Nanoconfined Fluid O₂ in a Lithium-Rich Oxide Cathode. Nat. Mater. 2024, 23, 826−833.

[2] Csernica, P. M.; McColl, K.; Busse, G. M.; Lim, K.; Rivera, D. F.; Shapiro, D. A.; Islam, M. S.; Chueh, W. C. Substantial Oxygen Loss and Chemical Expansion in Lithium-Rich Layered Oxides at Moderate Delithiation. Nat. Mater. 2025, 24, 92−100.

[3] House, R. A.; Rees, G. J.; McColl, K.; Marie, J. J.; Garcia-Fernandez, M.; Nag, A.; Zhou, K.-J.; Cassidy, S.; Morgan, B. J.; Islam, M. S.; Bruce, P. G. Delocalized Electron Holes on Oxygen in a Battery Cathode. Nat. Energy 2023, 8, 351−360.

[4] McColl, K.; House, R. A.; Rees, G. J.; Squires, A. G.; Coles, S. W.; Bruce, P. G.; Morgan, B. J.; Islam, M. S. Transition Metal Migration and O₂ Formation Underpin Voltage Hysteresis in Oxygen-Redox Disordered Rocksalt Cathodes. Nat. Commun. 2022, 13, 5275.

Raman spectroscopy is a non-invasive and broadly accessible technique for probing atomic vibrations in solid-state materials. However, its interpretation often depends on comparisons with preselected reference systems. First-principles calculations offer a powerful alternative for interpreting experimental Raman spectra, but they become computationally demanding for systems with large unit cells, defects, or mobile ions. To address these challenges, we developed fast computational frameworks that integrate machine-learning force fields (MLFFs) [1] and machine-learned polarizability tensors to predict Raman signatures associated with mobile ions and point defects in solid-state ion conductors.

Using this ML-Raman framework, we identify low-energy Raman modes in superionic conductors and broadened peaks in disordered systems, shedding light on the conductivity mechanisms of mobile cations. Furthermore, we introduce a novel method that combines MLFFs with atomic Raman tensors to predict the vibrational signatures of ionic point defects [2]. This approach has been successfully applied to capture temperature-dependent changes in experimentally measured Raman spectra of Ni-doped SrTiO₃, which were attributed to local variations in the dominant ionic defects. Our framework establishes new synergies between theory and experiment, enhancing the understanding of dynamical properties in energy materials.

The combination of machine learning (ML) with density functional theory (DFT) accelerates material simulations, expanding both spatial and temporal scales. However, current ML methods struggle to address polaron trapping. Polarons are quasi-particles arising from electron-phonon coupling in a wide range range of materials and shape the properties of the hosting systems. Therefore, understanding polaron effects is key for technonlogical applications. We present a novel machine learning force field (MLFF) approach that incorporates polaron trapping descriptors, enabling large-scale studies of polaronic materials.
Using TiO$_2$(110) as a case study, we reveal how dopants and atomic vacancies affect polaron configurations and drive catalytic CO adsorption. Additionally, our method captures the dynamic evolution of polarons with unprecedented statistical robustness.
This work advances fundamental understanding of defect-polaron interactions while offering a fully automated and efficient computational suite for the study of polaronic materials, facilitating characterization and design of metal oxide catalysts.
 

RuO₂ is widely regarded as one of the most efficient catalysts for water splitting, particularly due to its ability to enhance the oxygen evolution reaction (OER)¹. Gaining a deeper understanding of the underlying factors that contribute to its catalytic superiority is essential for advancing water-splitting technologies. This study explores the role of magnetic interactions in the OER at the RuO₂(110) surface using density functional theory (DFT). By modeling an antiferromagnetic RuO₂(110) surface, we examine how magnetism influences the electronic and adsorption behavior of both singlet and triplet O₂. Our results indicate that adsorbed oxygen adopts superoxo characteristics, with one unpaired electron, and transitions directly into a triplet state upon desorption. This ability of RuO₂ to facilitate direct triplet O₂ formation, bypassing the energy-intensive singlet-to-triplet transition², likely underpins its superior catalytic performance in OER. These findings underscore the significance of magnetic effects in RuO₂’s catalytic efficiency and provide valuable insights for the design of more effective catalysts for water splitting applications.

The progress in first-principles codes and supercomputing capabilities have given birth to the so-called high-throughput (HT) ab initio approach, thus allowing for the identification of many new compounds for a variety of applications. A number of databases have thus become available online, providing access to properties of materials, mainly ground‑state though. Indeed, for more complex properties (e.g., linear responses), the HT approach is still problematic because of the required CPU time. To overcome this limitation, machine learning approaches have recently attracted much attention.

In this talk, I will review recent progress in materials informatics focusing on the response properties of inorganic materials which play of key role in various physical phenomena such as linear and non-linear optics, thermal conductivity, superconductivity, or ferroelectricity. I will first present our HT calculations of the response properties based on density functional perturbation theory. I will briefly introduce the OPTIMADE API [1,2] that was developed for searching the leading materials databases using the same queries. I will review the MODNet framework [3,4] for predicting materials properties and which is particularly well suited for limited datasets through the selection of physically meaningful features. Finally, I will show how these tools can be combined in an active learning loop to discover materials with specific properties.

Hybrid organic–inorganic perovskites (HOIPs) have rapidly become very promising materials in optoelectronics, especially for use in photovoltaic devices such as solar cells. These materials typically consist of an inorganic framework—often made of metal halides—combined with an organic cation, which introduces structural flexibility and plays a key role in shaping the material’s overall properties. Within this broader class, chiral halide perovskites have recently attracted growing interest due to their unique optical and electronic characteristics. Chirality in these systems originates from the incorporation of chiral organic molecules into the perovskite lattice, which induces asymmetry in the crystal structure, particularly affecting the metal-halide coordination environment.[1,2] In this study, we present a computational workflow based on Density Functional Theory (DFT) and Time-Dependent DFT (TD-DFT) to explore the origin of chirality transfer and its impact on the structural and optical properties of two-dimensional chiral perovskites.[3] Specifically, we investigate the chiroptical response of lead- and tin-based systems: (R-/S-MBA⁺)₂PbI₄ and (R-/S-MBA⁺)₂SnI₄.[4,5] Circular Dichroism (CD) spectra are analyzed in conjunction with ab initio molecular dynamics and electronic density of states (DOS) calculations to identify the key factors influencing their chiroptical features. Our results show that these features are linked to a chirality transfer mechanism driven by the electronic level overlap between metal centers and ligands. This effect is particularly pronounced in tin-based chiral perovskites, which exhibit stronger excitonic coupling. The role played by asymmetric non-covalent interactions in inducing distortions within the metal–halide bonds will be discussed, highlighting their influence on the material’s chiroptical activity.[6] Furthermore, the thermodynamic and kinetic aspects of the early stages of chiral formation will also be presented, offering insights into the nucleation pathways and structural evolution of these complex systems.[7]

Metal halide perovskites have attracted significant attention over the past decade due to their exceptional properties for optoelectronic applications. Their soft, mixed covalent–ionic lattice presents fundamental challenges for understanding and controlling their dynamic behavior across a wide range of length and time scales. At the same time, this soft lattice enables the emergence of novel functionalities, such as chirality, in this class of materials. In this talk, we present key insights into the dynamical behavior of halide perovskites, obtained through force field-based modeling approaches.

We begin by focusing on several processes that critically influence the stability of halide perovskites. These include phase transitions driven by lattice anharmonicity [1], defect-assisted ion migration [2], and material degradation at extended defects [3]. By uncovering the underlying atomistic mechanisms, our findings contribute to a deeper understanding of instability in these materials and point toward strategies for improving their long-term stability.

In the second part of the talk, we turn to the emergence of static and dynamic chirality in halide perovskites. Through an analysis of both chiral and achiral compositions across a range of temperatures, we elucidate the mechanism of chirality transfer [4], and attribute the loss of structural chirality at finite temperatures to lattice vibrations, some of which are intrinsically chiral themselves [5]. We further demonstrate that the chirality of these materials can be compositionally tuned by ion mixing, offering new opportunities for engineering chiroptical functionalities.

Finally, we briefly highlight how these modeling approaches and physical concepts extend to oxide perovskites, where anharmonicity plays a role in dynamic symmetry breaking and coupling phenomena. In these systems, we investigate how lattice dynamics give rise to emergent properties such as ferroelectric and magnetic ordering, providing insight into the microscopic mechanisms that govern multifunctionality in complex oxides.

Deep-learning ab initio calculation is an emerging interdisciplinary field, which aims to greatly enhance the capability of ab initio methods by using state-of-theart neural-network approaches. Among these developments, deep-learning density functional theory (DFT) stands out as a particularly transformative direction, showing great promise to significantly accelerate material discovery and potentially revolutionize materials research. However, current research in this field primarily relies on data-driven supervised learning, which isolates the development of neural networks and DFT from each other, hindering their further advancement. In this work, we present a theoretical framework of neural-network DFT, which unifies the optimization of neural networks with the variational computation of DFT, enabling physics-informed unsupervised learning. Moreover, we develop a differential DFT code incorporated with deep-learning DFT Hamiltonian, and introduce algorithms of automatic differentiation and backpropagation into DFT, demonstrating the capability of neuralnetwork DFT. The physics-informed neural-network architecture not only surpasses conventional approaches in accuracy and efficiency, but also offers a new paradigm for developing deep-learning DFT methods.

Topological materials have emerged as promising candidates for future spintronics, quantum computing, low-power electronics, and optoelectronic applications due to their robust, symmetry-protected edge or surface states that are immune to backscattering and disorder. Extending this concept, higher-order topological insulators (HOTIs) host quantum states that are localized on hinges or corners, offering new functionalities for nanoscale device applications.

In this work, we present first-principles predictions of HOTI phases across two families of quantum materials. Firstly, we identify Li-intercalated graphene compounds that exhibit coexisting electronic and phononic topological features. The HOTI phases in these materials are protected by C₆ rotation and inversion symmetries, as confirmed by their calculated topological invariants (χ6 Z4). Topologically protected corner modes are noticed within the gapped bulk and edge states (in the armchair edge) in the nanoflake geometry of the Li-C₆ compound. Notably, the Weyl phonon arises at the valley-high symmetry point in the phonon spectra, accompanied by non-trivial Berry curvature and robust phonon edge modes.

Secondly, we examine a family of antiperovskite compounds such as Y₃InC, which host symmetry-protected triple-point phase without SOC and a twin Dirac node phase when SOC is included. These systems exhibit higher-order topological hinge states coexisting with gapless bulk Dirac phases, making them rare and valuable candidates for multifunctional device applications. Together, these findings unveil a rich landscape for higher-order topological phases in structurally and chemically diverse material systems, opening new avenues for their integration in future spintronics, quantum computing, low-power electronics, and optoelectronic applications.

Solid-liquid interfaces appear in many relevant processes such as heterogeneous catalysis and electrocatalysis and understanding them is key to upscaling energy applications to the industrial level. While the kinetics of desorption are well described by the Eyring equation, no ab initio equivalent exists for adsorption from condensed phases (e,g., from the aqueous phase). Common workarounds include: (i) using an Eyring-like equation, which actually applies only within homogeneous phases and therefore introduces dimensional artifacts when extended to heterogeneous surfaces; (ii) using the Hertz-Knudsen equation, which was originally derived for gas-to-solid adsorption and is inadequate for aqueous-to-gas adsorption; (iii) simplifying kinetic models based on assumptions about the rate-determining step, considering them to be either a diffusion or a reaction; and (iv) to empirically adjusting the kinetic parameters to fit a particular experiment. Because these approximations rely on assumptions that are not generalisable or do not hold for adsorption from the liquid phase, they induce discrepancies of up to seven orders of magnitude in the adsorption prefactor. Consequently, they fail to describe the reaction rates of even the simplest electrochemical processes, such as the relative rates of Hydrogen Evolution/Oxidation Reactions (HER/HOR) as a function of pH and electric potential, and their equilibrium lines. Here I will present the necessary conditions for a fully ab initio description of adsorption from the aqueous phase, using the HER/HOR and their equilibrium during electrolysis as a case study. The complete kinetic description combines energy profiles derived from Density Functional Theory data with microkinetic models, enabling a critical evaluation of all assumptions concerning the kinetic constants of adsorption. This work bridges a fundamental gap in interfacial science, significantly enhancing our understanding of solid-liquid interfaces relevant to heterogeneous catalysis and energy storage systems.