OPEN SESSION
ZOOM LINK TO JOIN IN: http://s.ic.fo/QTD_QuoVadisQThermo
Friday Oct 23, 2020 / 15:00-16:00 CEST
Quo Vadis Q Thermo?
On Friday, we will have the discussion session “Quo Vadis Q Thermo?” on open problems in the field.
Feel free to contribute before the meeting and propose topics for discussion.
Leave your comments below. We count on your participation to make a lively and fruitful event!
Are there any open questions that require experimental proof?
Quantum Thermodynamics: Science or Fiction?
Topic for discussion: the future of thermometry in small thermodynamic systems
The design of precise protocols for measuring temperature is an exciting and important problem. In this context, some interesting questions are:
1) Which are the fundamental limits for the precision associated with temperature measurements in small thermodynamic systems?
2) Can these be achieved in experiments?
3) How can estimation theory guide us to achieve these aims?
Topic: Role of thermodynamics in near-future quantum technologies.
This question is related to those raised by Matteo and Karen. It is known that in general quantum computers require a high degree of isolation from the environment to showcase an advantage with respect to classical machines. However, we are now entering the so called NISQ era, where quantum processors are still subject to external sources of noise but are sufficiently large to potentially exhibit some quantum advantage. This is specially important for applications related to the simulation of quantum open system dynamics, e.g. in Chemistry or Biology. In this scenario, that naturally includes the interaction with a thermal environment, what can thermodynamics tells us about the performance and resources required to optimize such processes? In particular, how can we apply thermodynamics to the analysis of speed limits in quantum simulators?
Topic: The cost and limitations of control
One question I keep asking myself concerns the fair evaluation of efficiencies in driven thermal processes. We all understand that if a quantum system is externally controlled by an agent (an in particular all quantum thermo experiments are in this regime), the work cost of control and observation is many orders of magnitude above the potential energy scale of the system. Why are we then so fixated on the quantum efficiency, COP, etc? Is this a remnant of classical thinking or is quantum work intrinsically more valuable, such that we can just safely disregard all the classical cost in the protocol. The same question arises for the evaluation of stochastic work in thermodynamics: any measurement of the work statistics will cost much more work than the work observed-so how can we get to self-contained notions of work fluctuations?
Topic: Does quantum technology in its near-term perspective (the ‘NISQ’ technology) really care about energetics?
It is easy to see how quantum thermodynamics will be vital when imagining a quickly-rechargeable, super powerful quantum battery for your quantum-computing laptop of the far future. But what about the near future? What realistic milestones can q-thermo pose to lead to something actually useful? This of course also relates to the question raised by Matteo about demonstrating a clear quantum advantage at a thermodynamic task.
The following topic is a long term and very broad question but it might be interesting to be considered.
What is the energetic/thermodynamic cost of information processing and classical/quantum computing? There are already several very interesting papers on the subject but it is still a widely open question. This is a relevant question due to the massive use of computation and information processing that our society does, as we all know. And apparently it is not going to slow down any time soon. Then, it might be that soon, the amount of resources required to complete a computational task becomes as crucial as the time it takes to complete it. It might be also that quantum computation is “better” than classical computation not because it allows for faster computation but because it requires less resource to complete the same computational task.
This kind of questions might become very important.
(By the way, this is probably this kind of ideas behind the special issues in Entropy and Journal of Statistical Physics that Sebastian Deffner is co-leading, and also the one by Gerard Milburn, also in Entropy).
Topic: Non-classical fluctuations as an interdisciplinary endeavor: What (more) can we learn from other fields and what can they learn from us?
As most of you know, defining work as a fluctuating quantity in the quantum regime poses a substantial challenge. This challenge is not unique to the quantum thermodynamics community. As an example, the transport community has faced similar issues when computing higher moments of the charge transferred through a conductor. In analogy to work being the time-integral of power, charge is the time-integral of current, and there is an ambiguity in how to time-order the operators that appear in the higher moments of work/charge.
Since quantum thermodynamics is highly interdisciplinary, I believe that we can (and already have), learn a lot from the approaches to fluctuations taken in different fields. In addition, creative novel approaches have been put forward to describe fluctuations in quantum thermodynamics. These may be beneficial for other fields.
I envision a brainstorming group that aims for obtaining an interdisciplinary understanding of non-classical fluctuations, identifying and solving outstanding problems by employing approaches across disciplines.
Open problem of Quantum Virology.
Quantum contribution in thermodynamic parameters for nucleic acids, obtained in water solution at moderate ionic strength ( 1.0 M NaCl) and physiological temperature?
Topic: Finite-time thermodynamics.
Optimising finite-time thermodynamic processes is known to be a hard task as one needs to optimise over “protocols” (i.e. continuous functions in the parameter space). Recent results show that general results are possible in limiting cases, namely the slow and fast driving regimes, and in the presence of Markovian dynamics (see e.g. arXiv:1907.02939, arXiv:2010.00586). This raises the question:
* Can we hope to find general solutions for the power-efficiency tradeoff of (quantum) heat engines?
Any progress in this topic (which is essentially 200 hundred years old) would be really nice!
Related interesting open questions are:
* How do fluctuations enter into play here? In other words, what is the regime of validity of thermodynamic uncertainty relations?
* How do such optimal protocols scale with the size of the engine? and what are the implications in many-body systems?
Two proposals for discussion:
1. Quantum thermodynamical advantage and roadmap: Can we agree on an operational task which, if performed by a collection of quantum machines, would showcase a clear-cut and relevant advantage over what can be done by classical machines? If so, can we define a minimal set of “thermodynamic DiVincenzo criteria” that would provide us with a roadmap to eventually build a device demonstrating the advantage (with fair resource bookkeeping)?
2. Quantum thermodynamics as a tool: What important problems from other fields can we solve using quantum thermodynamic tools? (e.g. diagnostic of quantum circuits, metrology, (actual) cooling problems, control, quantum simulation of open quantum system dynamics etc.)
Topic: Entropy production and irreversibility.
The non-negativity of entropy production is a very nice and concise mathematical formulation for quantifying the irreversibility of a process. Irreversibility is a concept that, in my opinion, only makes sense if we compare forward and backward processes in some way. Therefore, can we talk about entropy production at all without mentioning explicitly which forward and backward processes are being compared, as some approaches do by simply defining a non-negative quantity as the entropy production? Also, is there (can we find) an operational interpretation of why some processes have a larger entropy production than others? Are such processes more difficult in some sense of being reversed? I feel such questions would be nice to be addressed/discussed. ^^
Topics for discussion:
Topic 1: Is, or is not, the concept of work in the quantum regime now settled?
What is the answer? 😉
I also would like to know what people think on this. In particular there has been some recent results claiming that when not only the eigevalues but also the eigenstates of rho change in time the change in energy due to the eigeinstates (the coherences) is work (since it does not change the von Neumann entropy)