Thermal machines, i.e. engines and refrigerators, represent a pillar of all modern everyday technology. Generally, they can be grouped into two classes: (a) autonomous steady-state thermal machines, characterized by the simultaneous presence of multiple baths at different temperature and/or chemical potentials; (b) periodically driven thermal machines, which are operated by periodically changing in time both the mechanical parameters of the working system, as well as the temperature of its surrounding reservoir, thus generating power by external driving.
Classically, their performance is quantified in terms of their power yield and their efficiency, the latter quantifying the ratio of useful work to the input energy (typically heat) provided, and, on average, achieving maximum efficiency always comes at the cost of vanishing power output, and vice versa. At the macroscopic level, fluctuations of thermodynamic quantities, such as power, are essentially negligible thanks to the law of large numbers.
The ever-forward miniaturization has nowadays allowed to explore and control systems at the microscopic level, encroaching upon length scales where quantum effects become predominant. Genuine quantum properties such as coherent superposition and entanglement can now be achieved in a diverse set of experimental platforms, thus paving the way for next generation quantum technologies, such as quantum computers or nano-scale quantum thermal machines, whose promise is to outperform any classical counterpart.
Whenever such nano-scale devices are considered, however, fluctuations of all thermodynamic quantities, such as heat and work, become extremely significant, since the regime of validity of the law of large numbers ceases to be valid. Crucially quantum thermal machines must operate reliably, i.e. their output should ideally exhibit small fluctuations over many runs. Achieving a determinate precision however inevitably comes at a cost in terms of thermodynamic resources, such as dissipated heat or excess work, thus massively impacting the machines’ performances. While the presence of genuine quantum features may lead to advantages in terms of average quantities, it raises the important question whether this comes at the cost of less precision/reliability and/or higher thermodynamic cost.
The overarching goal of this proposal is to address the following question: what is the most general and fundamental thermodynamic cost of precision in genuine quantum thermal machines?
This is articulated in two main Objectives:
[O1]: Characterise the thermodynamic cost of precision for genuinely quantum periodically-driven thermal machines and quantum-measurement thermal machines.
[O2]: Determine the thermodynamic cost of information processing and of precision in the Quantum-Field Machine.
The results will be applicable, both in the near future and long terms, to second generation quantum technologies, ranging from quantum thermal machines to high-precision quantum sensors for metrology and finally to quantum computers.
