First principles evaluation of structure and properties of multifunctional materials

Abstract

[eng] Organic chemistry is recognized for its ability to arrange a limited set of atoms in numerous ways, a quality that is crucial in molecular magnetism. This branch benefits from the adaptability of organic chemistry to conceive a plethora of radicals, i.e., open-shell molecules with unpaired electrons. These organic radicals are sought for their potential applications in data storage and spintronics, paving the way towards miniaturized, lightweight and environmentally friendly solutions. However, challenges remain, particularly regarding the stability and reliability of these radicals together with their performance and resilience at room temperature. Accordingly, the successful integration of these radicals into technological innovations requires progress in their synthesis, characterization, and their property prediction to better suit real-world applications. Yet, predicting the capabilities of these radicals is far from straightforward. Even minor modifications in their molecular structure can have significant impact, affecting both their electronic configuration and their arrangement in crystals. Consequently, there is a high demand for the strategic design of organic radicals and a more rapid evaluation of their properties to fully harness their potential. Our research has encompassed three distinct but interrelated topics, all contributing to the understanding of open-shell molecules and materials. Central to this PhD thesis is the exploration of magnetic properties of these materials using computational chemistry techniques. A significant part of the research involves examining how triarylmethyl-based 2D Covalent Organic Radical Frameworks react to external mechanical stimuli. This investigation has demonstrated the efficiency of these frameworks in transitioning between various electronic states, each exhibiting distinct properties and characteristics. In the second study, we make use of quantum mechanical insights and qualitative models to design and characterize a new set of diradicals displaying a robust triplet ground state. Notably, we employ pentalene-based couplers as building-blocks, contributing to pivot the role of pentalene from academic curiosities to potentially viable candidates for high throughput applications as ferromagnets. Lastly, the third study consists of the development of a new quantum-informed descriptor for Machine Learning applications in chemistry, referred to as Molecular Orbital Decomposition and Aggregation (MODA). We prove that MODA excels in predicting magnetic exchange couplings with respect to other well-established descriptors such as Bag of Bonds and Smooth Overlap of Atomic Positions. Overall, the interplay among the three facets provides viable prospects, underscoring the potential of organic radicals in advancing molecular magnetism and paving the way for innovative applications in data storage and spintronics.

[spa] La química orgànica és reconeguda per la seva habilitat en organitzar un conjunt limitat d'àtoms de nombroses maneres, una qualitat crucial que aprofita el magnetisme molecular. Aquesta branca es beneficia d’aquesta versatilitat per concebre una gran quantitat de radicals, és a dir, molècules amb electrons desaparellats. Aquests radicals orgànics són buscats per les seves potencials aplicacions en emmagatzematge de dades i espintrònica, obrint el camí cap a solucions miniaturitzades, lleugeres i respectuoses amb el medi ambient. No obstant això, encara hi ha reptes per resoldre, particularment pel que fa a l'estabilitat i fiabilitat d'aquests radicals. La integració exitosa d'aquests radicals en innovacions tecnològiques requereix progressos en la seva síntesi, caracterització i la predicció de les seves propietats per adaptar-se millor a les condicions requerides en aplicacions reals. Tanmateix, predir les capacitats d'aquests radicals està lluny de ser senzill, ja que modificacions menors en la seva estructura molecular poden tenir un impacte significatiu, afectant tant la seva configuració electrònica com la seva disposició en cristalls. La nostra recerca ha abastat tres temes interrelacionats, tots contribuint al camp de les molècules i materials amb electró desaparellat. Central en aquesta tesi de doctorat és l'exploració de propietats magnètiques d'aquests materials utilitzant tècniques de química computacional. Una part significativa de la recerca implica examinar com els Covalent Organic Radical Frameworks reaccionen a estímuls mecànics externs. Aquesta investigació ha demostrat l'eficiència d'aquests materials en transicionar entre diversos estats electrònics. En el segon estudi, s'ha fet ús de conceptes de la mecànica quàntica per dissenyar un nou conjunt de diradicals que mostren un estat fonamental triplet. Notablement, emprem acoblaments basats en pentalè com a unitats fonamentals, contribuint a pivotar el paper del d'aquest des de curiositats acadèmiques cap a candidats viables per a aplicacions d'alt rendiment. Finalment, el tercer estudi consisteix en el desenvolupament d'un nou descriptor per a aplicacions d'Aprenentatge automàtic en química, anomenat Molecular Orbital Decomposition and Aggregation (MODA). Demostrem que MODA destaca en predir acoblaments d'intercanvi magnètic respecte a altres descriptors ben establerts en la literatura.