Minisymposium

Spin-crossover (SCO) molecules exhibit bistability between two electronic states, switching in response to an external stimuli that alters the material properties. This makes SCO systems promising candidates for molecular-level based applications. The transition temperature (T1/2) marks the point where spin populations are equal, and is a key parameter in SCO systems. While Fe(II) compounds dominate the field, Fe(III)-based SCO systems offer advantages for technological applications. However, experimental data on Fe(III) systems is limited. Electronic structure methods, particularly density functional theory (DFT) calculations, help in the design of new Fe(III)-based SCO systems with targeted T1/2 values. Our results demonstrates that DFT calculations accurately reproduce experimental T1/2 values and enables broad ligand functionalization screening. Moreover, all observed trends can be explained through the underlying electronic structure of the system. These calculations provide with valuable guidelines for chemists when developing new SCO compounds with specific properties. Additionally, this method allow us to generate data that can be used to train machine learning (ML) models employing SOAPs descriptors for the automatic classification of Fe(III) based molecules into high-spin, low-spin, or spin-crossover categories. This approach enhances the predictive capabilities for new SCO materials, accelerating their design and application in technology.

Amorphous materials, ubiquitous among natural and manufactured solids, exhibit topological disorder and increased surface entropy. Atomic vacancies, in the form of undercoordinated defects, profoundly alter fluid/surface interactions and can be exploited to design sorbents for direct air capture (DAC) of carbon dioxide and sustainable catalysts.Standard atomistic modeling and analysis techniques, however, are not tailored to amorphous solids, and new tools and approaches need to be devised to characterize surface heterogeneity. Machine learning (ML) provides powerful methods to address the complexity of amorphous structures, enabling the identification of intricate surface regions and relevant adsorption sites.We apply segmentation techniques to identify adsorptive structures with complex geometries that cannot be effectively analyzed using classical radial distribution functions. Additionally, we employ local atomic-environment descriptors to classify surface structures and develop proxies and surrogate models to accelerate the assessment and design of materials for energy and environmental applications.

Magnetoelectric materials in which applied electric fields (E-fields) can control magnetic properties are inherently compatible with existing E-field gated electronics and hold huge promise for new technological applications (e.g. low power spintronic devices).[1] At present, transition metal-based inorganic systems dominate this field. However, magnetoelectric systems that leverage the unique benefits of organic materials could emerge as a new platform for spintronic technologies. Based on DFT calculations on organic diradicals consisting of two spin centers connected by dipolar aryl linkers, it will be shown that E-fields can induce significant twisting of the linkers. This twisting alters π-conjugation and, consequently, the magnetic coupling between spin centers. The magnetoelectric response is influenced by steric hindrance, π-conjugation and polarization. This approach is also applicable to 2D covalent organic frameworks. Additionally, we will present our recent efforts to develop new quantum-inspired representations for accurate and efficient Machine Learning protocols to predict magnetic couplings between organic radicals[2]. [1] X. Liang, et al. IEEE Trans. Magn., 2021, 57, 400157. [2] R. Santiago, S. Vela, M. Deumal, J. Ribas-Arino. Digital Discovery, 2024, 3, 99.

Data-Driven Nanomaterials Design