We explore a general feature of the interaction mediated by the gravitational fields of spatially superposed masses. For this purpose, based on quantum information theory, we characterize the evolution of two particles each in a superposition state of paths. The evolution is assumed to be given by a completely positive trace-preserving (CPTP) map. We further assume that the probability of a particle being on each path is unchanged during the evolution. This property is called population preserving. We examine when a population-preserving CPTP map can create entanglement in terms of separable operations, which form a large class of local operations and classical communication (LOCC). In general, entanglement is not always generated by inseparable or non-LOCC operations, and one can consider a model of gravity described by an inseparable operation which does not create entanglement. However, we find that a population-preserving CPTP map is inseparable if and only if it can create entanglement. This means that the above model of gravity is incompatible with the possible evolution of spatially superposed masses.
Even though coherent control of quantum operations appears to be achievable in practice, it is still not yet well understood. Among theoretical challenges, standard completely positive trace preserving (CPTP) maps are known not to be appropriate to represent coherently controlled quantum channels. We introduce here a graphical language for coherent control of general quantum channels inspired by practical quantum optical setups involving polarising beam splitters (PBS). We consider different situations of coherent control and disambiguate CPTP maps by considering purified channels, an extension of Stinespring’s dilation. First, we show that in classical control settings, the observational equivalence classes of purified channels correspond to the standard definition of quantum channels (CPTP maps). Then, we propose a refinement of this equivalence class generalising the "half quantum switch" situation, where one is allowed to coherently control which quantum channel is applied; in this case, quantum channel implementations can be distinguished using a so-called transformation matrix. A further refinement characterising observational equivalence with general extended PBS-diagrams as contexts is also obtained. Finally, we propose a refinement that could be used for more general coherent control settings.
We ask whether there exists a relation between controllability of the fluctuations in extractable energy of a quantum battery and (a) how open an arbitrary but fixed battery system is and (b) how large the battery is. We examine three classes of quantum processes for the energy extraction: unitary operations, completely positive trace-preserving (CPTP) maps, and arbitrary quantum maps, including physically realizable non-CPTP maps. We show that all three process classes yield the same average extractable energy from a fixed quantum battery. Moreover, open systems are better at controlling fluctuations in fixed quantum batteries: while random unitary operations result in nonzero fluctuation in the extractable energy, the remaining two classes lead to vanishing fluctuations in extractable energy. Furthermore, when the auxiliary system used to implement the non-unitary physically realizable maps is restricted up to a dimension
n, fluctuation in extractable energy scales as
1/n for CPTP maps, outperforming the
lnn/n scaling observed for general quantum maps. Even within open dynamics, therefore, energy extraction via random CPTP maps exhibits greater resilience to fluctuation compared to processes based on arbitrary quantum maps. We subsequently obtain that fluctuations in extractable energy scale as the inverse of the battery's dimension for all three process classes. Unitary maps, therefore, perform - in the sense of as low fluctuation as possible - equally well as more resource-intensive open maps, provided we have access to large quantum batteries. The results underscore a fundamental trade-off between performance of a battery and the resource cost of implementing the extraction processes.