 This session we are going to have three talks of 25 minutes overall so you have 20 minutes for your talk and about five minutes for the question and at the end of this session we'll have a 30-minute discussion where all the panelists will be, we'll have the possibility to answer a question from the audience. Right now we will start with, you'll be five minutes late but we are going to start with the first talk from Obina Abak from Queen's University Belfast and he's going to talk about quantum thermodynamics and quantum control so Obina you can please share your screen and I will let you know five minutes before the end of your talk about the remaining time you are so whenever you are ready please start. Is the screen full now? Yeah you have to put it in full screen mode. Okay perfect. Just don't see you. Okay it's okay. We can see you. You should share only the application. Stop sharing and relaunch the application. Okay. Okay. Yes. Okay thank you the organizers and thank you for the invitation to present part of my current research in the subject of quantum thermodynamics and quantum control. So today I'll try to give you a brief overview of what I've been doing in recent past and then I will start first by trying to give a brief motivation why we are interested in these kind of subjects and then I'll make a brief reminder on false truth or to refrigerator you are trying to you are motivates what we already know from your first or your third year summer physics and then I'll try to now bring up the idea when we have control in the system can we generate a better performance or not and also I will give you a second application of this kind of quantum control called shortcut to adiabacity. So if we look over the years the technology have done gone quite remarkable in downscaling sizes of our electronic devices for instance we are all having smartphone or tablets and this in the seventies used to be huge mainframe devices but in principle all these are limited or the challenges being faced by every of these electronic devices and our laptop is mainly how best can we manage the heats and as well as increasing the processing speed and the same when people have been trying to minimize these engines up to down and it was only in 2016 with the group of Ferdinand Keller Schmid Keller in mines that were able to downscale a functional heat engine auto single atom regime so in the last year a group in brazil here in this here this is the NMR experiment where they show that you can actually realize a an auto in a quantum heat engine by having an RF here which pumps some heat into the code but by doing some work this is just a short proof of principle experiment and they are able to show that yeah they can get efficiency up to around 0.44 percent and if we look yeah no 44 percent and if we look here we see that they have to be some time limit which must be exceeded before you the engine can actually perform a useful work and the implication of what this can be used are quite enormous and one can think several of application for instance in quantum computers or sensors and also in a number of places like nanotechnological application so let's look what is actually a refrigerator so it's basically any device which performs some work which you extract some heats from a cold bath to a hot bath and then in the process of adding some work and they cannot efficiency with this with the ratio of the input heat up on the output is given by this is given by this temperature ratios but the one of the problem is that although is maximum performance it is actually slow but for a real engine things happen in finite time and this actually leads to dissipation and one way to characterize this is always what people regard as coefficients of performance at maximum figure of merit so depending on what you choose as your figure of merit so and then if we take that the performance and then as well as the cooling power at a finite time then we'll have a expression that depends on the ratio of the temperature and let's say look now on the if we look at four stroke engine we can just think about our thermodynamic subject you have a piston and a gas you compress it you do some work and then you shine some laser which exchange heats so and then add some energy and then you can push the piston back and then put it in contact with a cold bath and then this is a typical engine cycle and if we perform the analysis like this for instance using a quantum framework in a weak coupling limits we can actually do the same so like you have an ion trap then your modulates your frequency and then your couple shine some laser lights and your modulate backward and here during the change of the frequency or the size of the trap then there is a q star here which tells you how fast or how slow you do this adiabatic process so we can compute the performance exact in this scenario and that is given by this expression here where this q star tell you how fast or slow you drive the process and when the q star one and two is equal to one that means that the process is quite very slow that's the adiabatic regime and in this regime then the performance of the refrigerator is characterized by the ratio of the efficiency the ratio of the frequencies and then if it's greater than one the performance is always less than what you obtain for the adiabatic scenario and let's look at the plot here you can see the blue line is the efficiency the coefficient of performance at the maximum figure of merit and then the dotted dark line is the cannot the cannot coefficient of performance that's the maximum you can achieve at zero power and the red and the green ones this is when you now start to yeah in drive this in a finite time so you have some non adiabatic effect and then you see that the efficiency or the performance reduces so then the question is is there any way we can actually improve this performance of this engine in a finite time so we need to take some control and why would do why do you need this is not only an application that is necessary for thermodynamic engine but it has been quite useful in other areas of yeah modern physics critical systems and then metrological protocol and then also circumventing the coherence in a system so let's say consider one of the known way of doing this which is called short cut to adiabaticity and this is basically when you just a kind of a slowed down a fast motion video so you start a process you start a process for instance you look at the plot then we want to end at the same point where it will end in adiabatic situation without minding what happens in between and then if we look at the first question here then what it tells you is that we need to construct additional Hamiltonian which will suppress the non adiabatic transition during the driving protocol and by giving some bounds that your frequency need to fulfill then we can actually construct yeah this kind of protocol for be it two level system harmonicus later for the case of harmonicus later then we can actually have expression that looks in this form as our additional Hamiltonian that if we put it with the original Hamiltonia it will ensure that we suppress the adiabatic adiabatic contributions however the question becomes what about the cost of this additional Hamiltonian and how do we see it so one way to look at it is like for instance if I now consider some instantaneous power analysis and do this then this tells me that maybe if I'm looking between around two and eight it will appear as if I'm not really yeah doing any work or maybe I look two point five and then eight points around nine point five or so then but if I want to know my actual power I have to compute what we call in electricity the average power and if I take this analogy and then apply it here I can actually compute what I have a definition for the cost of my driving and then this mean the cost is actually also in the same form which we used to characterize our be it refrigerator and other thermal devices and then if we compare this with what we call the non adiabatic work then we see that this in a very short time the blue the red line is greater than the gray which is the non adiabatic work that's the friction so in a very short time it is actually very huge cost to perform this kind of a share shortcut protocol but there is actually a good range of time where you see that it's quite lower than the work friction so if we now come back to the analysis of our refrigerators and then we put the cost in the definition of the performance so we can then compute yeah the cooling power yeah in this form and then the dotted black line is the adiabatic situation then the red is the shortcut protocol and the blue one is the non adiabatic situation and you can see actually that in a very short time that it doesn't you doesn't gain anything by doing the shortcut protocol but between the intermediates which intermediary job then you actually have a reasonable advantage and for the cooling power then this you also have advantage than the doing the non adiabatic protocol and therefore the analysis on this is looking at the figure of merit we choose and we see here that for that figure of merit that you can actually have an hierarchy which tells you that although the shortcut protocol is lower than the adiabatic situation it is always greater than the non adiabatic situation provided that you are above the region where this time is greater than the time where the costs of doing the shortcut is greater than the costs of the friction so that tells us that we can simultaneously enhance both coefficients of performance and the cooling power then we now look at the green plots here we look since we are talking about time and the quantum mechanics tells us that there is limit to the speed of evolution of a system which is usually called the quantum speed limit so we can use now this framework also to try to get a kind of a bar a tighter bound on the performance of our shortcut protocol and if we use this expression here due to an adan and the Aranov the boss angle the ratio of the boss angle between the density operators and then the with the mean energy of the associated with the shortcut so we can actually get a bound based on the quantum speed limit and that's the green plots here and we see that this is always higher than the sda and likewise also when you look for the costs no the cooling power so more information on this you can see on this article yeah recent article here with Mauro Paternostro and then very close then the next part of the talk or net upness application of the sti I want to present to you is like yeah we consider one of the system you have two system we know in physics are basically the harmonica oscillator and then two level system and for these two level system what happens when we bring two of them together and this is quite relevant for range of experimental platform be it's a cavity QED like found in the mass plank and the other places around the world and then also like the superconducting or sacred QED and then this yeah fine so and then as well as the linear ion trap so and this is well described by a well-known Jens Cummings model so where this is the describe your harmonica oscillator then this is the two level system and the coupling between the two and this can be have a transition yeah between the state different levels and the resonance you can actually complete complete a perfect transfer from your ground states n plus one to the end of the excited state at some time but this is not possible for all n at the same time so what can we do with this then what we set out to do was to generate a non-classical state using this kind of framework combining it with this shortcut protocol and one way to go about this is like first we know that we can map this to a Landau-Zener problem form where you have something that look like your sigma x z and then sigma x where these are the spin operators and they would with this in mind construct the shortcut to a debacteristic protocol and that you can do population transfer in a finite time and if we now use this framework we can actually construct different kind of non-classical states let me just give you example of one of them here so which is like one known important one called the Schrodinger cast states so the idea is that we start with a fork state then we make a pi over half pi over two posts and then you perform the STA protocol and we can concatenate this into many times and after that then it makes a measurement so if we do this then you measure the system let's say we start yeah for instance here we want to go from this to this yeah zero plus two states then we'll say that we actually have the state at a very high fidelity why if we do this with time independent protocol we get 0.08 so this scheme offer large improvement on the fidelity so in conclusion I have a try to show you here the refrigerators efficiency can be better if we do shortcut protocol and then I show you that quantum speed limit also impose some bounds on performance of quantum machines and cooling is not always equal to power so it's always good to look at the overall performance because of whenever you do this kind of a shortcut protocol and then the last I also show you that we can combine the shortcut protocol and James Cummings model to generate non-classical states and this is quite robust against noise and many imperfect pulses as we study so with this yeah you can see more on the reference yeah about this to work me one is this on the refrigerators yeah publishing the fiscal review research and for the parts on the state engineering that's this year PRL here with Ricardo Puebla and Mauro Patanastro so with that thank you for your attention thank you Obina for this nice talk anyway within the time limit thank you for that now if there any question I've seen on the chat that Ali has a question if he cannot mute yeah thank you Steve thank you Obina for this excellent talk I have two questions the first one is in one of the slides you showed a theory for the auto cycle for a quantum harmonic oscillator is what changes in how it is number two is way to extract the amount of dissipation for a thermodynamic system but how much do these results change when you go down to you know smaller scales okay if I understand yeah the first just I cannot yeah really go back okay yeah the first one is about the cycle of the auto this one yeah so here what's here is the system is closed yeah in the first part of the protocol so yeah and then for this particular situation we are considering yeah so this is just a kind of unitary evolution so when you couple with the lights so this is the quantum harmonic oscillator yeah the quantum harmonic oscillator yeah in this case but I guess my question is if you had and harmonies yes let's say if it was not harmonic how would this change you if for what would change basically yeah you have a higher energy in the system yeah probably and then yeah what's another thing is that when you you do the the open side when you couple the reservoir so depending if you are in weak coupling limit so far the study so far what has been shown is that you have basically what still looks more or less like the linear one except that now you have modification in your performance both at maximum power so because you have putting more power and then these changes yeah them you have your performance at maximum figure of merit looks like I say one okay then the second one is a yeah how the energy changes yeah for this small system yeah in the experiment we are with the 2006 yeah with the color what we find is that when you scale actually the energy so the amount of energy being produced yeah by this and then a kind of a try to scale in the sense that by dividing by considering the per kilogram so like you check the performance of a car and then the performance of this kind of engine by the weights of the medium that's performing the cycle then this is just by a difference of around 10 between the two so if you check it just by the yeah okay the weights yeah efficiency per weight and there's no much difference yeah between the two the only difference yeah we start we can play around here is that maybe yeah we can start to think about doing some yeah like using some quantum approach like using like squeeze light can we get more by doing this instead of just coupling yeah ordinary thermal reservoir okay thank you thank you okay thank you Abba there's actually another question but I will ask Mathura to hold that question and prepare it for the discussion session slightly later you want to try to be on schedule for the next talk of esterina