Spin molecolari per l’Informazione Quantistica: Nuove Piattaforme per la Correzione di Errori, la Simulazione e l’Inizializzazione

Quantum information processing has emerged as a forefront area of research, holding the promise to address classical intractable problems. There are several areas in which quantum computers are expected to bring a disruptive contribution and one among all is the simulation of quantum systems. In the last decades, several quantum computing architectures have been proposed and experimentally tested: among the most developed are superconducting qubits, trapped ions, and photonic systems. However, despite the significant technological advancement, the path towards the realization of a scalable and fault-tolerant quantum computer remains full of challenges. This scenario underscores the need to explore alternative possibilities and an intriguing one are Molecular Nanomagnets (MNMs). Discovered in the early 90s, these systems exhibit classical macroscopic magnet behavior while also showing quantum properties. This combination of classical and quantum properties makes them excellent candidates as building blocks for quantum architectures. Leveraging on the spin degrees of freedom inherent in these systems, they can show at low energy a well resolved energy spectrum addressable by electromagnetic pulses. These systems have naturally long coherence times that reach in some cases up to times on the order of milliseconds allowing a coherent control and manipulation. Furthermore, MNMs can be chemical engineered allowing for a precise tuning of their properties and thus a great flexibility for different applications. This thesis collects the results of the research conducted by my group and me over the past three years. The thematic common at all works is the use of MNMs for the quantum architectures. Within this broad field, research can be divided into two primary strands: exploring the potential of a qudit-based computation and designing molecules that are intrinsically protected against decoherence. Regarding the former, a qudit approach can simplify quantum circuits by reducing the number of objects to be controlled and thus the number of two-body gates. This approach has been investigated both for quantum simulation and for quantum error correction applications. Regarding the latter, through the modelling of decoherence and the analysis of the mechanisms that generate it, it has been possible to identify a general property that systems must have in order to be intrinsically protected against decoherence. As in precedence, this property can be useful for both quantum computation and error correction. Besides these main topics, we have also worked on the design of time-resolved EPR experiments aimed at detecting chiral induced spin selectivity (CISS). Although this phenomenon is not fully understood, it is interesting for several applications, including quantum technologies. In particular, it could be exploited for the initialization and readout of MNMs. Finally, we have worked on state of art quantum computers exploring the variational quantum eigensolver algorithms. We have worked on the design of physically motivated ansatz, i.e., ansatz tailored to respect the properties of the target system such as conserved quantities or symmetries, showing that this approach can significantly extend the capabilities of these algorithms. As a study case, we look at small spin models.