This project is both scientifically and technically challenging, such that the first half of this period was dedicated to installing and commissioning a new scanning tunneling microscope that is optimized for high magnetic fields to create the Zeeman splitting in the single spin system as well as radiating microwaves into the tunnel junction in order to excite the spin system. A new concept for introducing microwaves with frequencies up to 100GHz was developed including a focused antenna that radiates the microwaves towards the tunnel junction for most efficient coupling. The introduction of microwaves into the tunnel junction has turned out to be very efficient such that we have, in addition to the project's goals, explored the potential of microwave assisted tunneling in the STM. The microwave source extends the capabilities of the experiment and gives access to sample and junction properties, which are unavailable in conventional STM setups, in particular concerning coherent processes addressing the behavior of the phase during the tunneling process. The tunable frequencies of the microwaves also provide a new energy axis for spectroscopic measurements with the potential to reach a much higher energy resolution than along the conventional bias voltage axis. We have investigated the intricate interplay of different coherent processes during quasiparticle tunneling between superconductors. We found strong interference effects between coherent processes such as multiple Andreev reflections and microwave assisted tunneling, which cannot be modeled as a superposition between different independent processes, but have to be modeled in a one-step model combining both processes at the same time. Going beyond the ESR-STM concept and the outlines of this project, we devised a spin-off STM experiment to directly manipulate different atomic scale systems with microwaves. More specifically, we are manipulating Yu-Shiba-Rusinov states with microwaves to test their suitability as a few-level system for direct manipulation. After having demonstrated the principal capabilities of our ESR-STM in the so far unique frequency range between 60GHz and 100GHz, in the second half of the project, we have entered the scientific part of the project. We have found that the applied bias voltage in the STM induces an electric field that directly impacts the properties of the spin system. The dipole of a TiH molecule feels an electric force, which changes the molecule-substrate coupling, which in turn changes the g-factor of the spin system. In this way, we have identified a new mechanism of spin-electric coupling at the atomic scale, which provides the future prospective for direct electrical manipulation of spin transitions, one of the chief goals in spintronics. In general, electric fields are much easier to handle (fast switching, spatial containment) than magnetic fields, so that we see this as the achievement of this project with the highest potential impact in the near future. Furthermore, we have explored the suitability of Yu-Shiba-Rusinov (YSR) states for spin-based few level systems, where we can report two major achievements. The first one is concerned with transport between two YSR states, where we have shown that a very efficient tunneling current (YSR-YSR tunneling) can be passed through two YSR states even at extremely high tunnel junction resistances. As the YSR states present single energy levels, this YSR-YSR tunneling constitutes a minimal tunnel junction. The second achievement is concerned with the identification of the YSR ground state as it is a priori not easily possible to see if the system is in the free spin regime or the screened spin regime. We have exploited the Josephson effect to create essentially the smallest SQUID and use that to identify the YSR ground state through the presence or absence of a supercurrent reversal (π-junction).
