 Okay, cool. Yeah, I really appreciate the organizers for this workshop wonderful place for all the non equilibrium and stochastic thermodynamic people get gathering together. And today I'm trying to pack actually lots of different interesting toys that my research group has been playing in the past a few years, and mostly toward the design principles of interesting non equilibrium kind of machines or molecules. So without further delay, let me show you. Okay, you see my mouse, hopefully. Yes. Yeah, thank you. Um, so before we start I want to really say that well thermodynamic laws are really beautiful and general. But if you're only looking at the thermodynamic laws, no matter how general they are, how pretty they are. Actually, there is a gap before you can really design real life machines. So that was actually resolved by lots of engineers they're using laws of thermodynamics. They're collecting more and more specific data for certain kind of substances and then the, the, the, the basically converted those very universal and dry principles into design principles that is kind of more practical. So I don't allow that line. I want to basically kind of lead us to a set of questions here that is in stochastic thermodynamics it has been really successful in the past few decades, explaining systems that are at the nano scale system that are arbitrarily far from equilibrium. In fact, their first law, second law are all kind of brought into non equilibrium and nano scale by the advancement in the stochastic thermodynamics. Furthermore, there are so many interesting directions emerging in the field of stochastic thermodynamics it has information thermodynamics that talk about what is the thermodynamic cost of information processing, talking about sensing talking about computation, and also there are nice, nice results of uncertainty relation, which relates entropy production rate and the fluctuation of current. And also there are there are studies of strongly interacting systems, and also a study of geometry in thermodynamics, and those are all nice and beautiful theories. And I apologize that this is such a small space I cannot really pack all the, all the interesting references here, but I want to give us a pause and ask one question. Before we declare victory in understanding how we can design things out of stochastic thermodynamics, we have to mind the gap. There is a gap between the general and universal thermodynamic, thermodynamic laws and theories, and the practical design principle of living matter. Let me show you what do I mean here on the left I'm showing you well we can probably call that a living matter it's definitely alive. There is an immune cell that is a hold at this location, and on the left, an experimentalist a whole the pathogen that the immune cell would like to eat. And you will see that this little bag of molecules are able to detect the environment and sense that there is something you want to eat and generate motion and deform itself trying to go to the to the thing that they want to eat. It's kind of amazing that this tiny bag of chemicals and molecules can kind of work together to recognize temporal patterns of external stimuli, and also they can respond to this stimuli and carry out certain functions. And also very interestingly that they are able to autonomously extract ambient non equilibrium energy. So please notice that I'm trying. I'm putting a non equilibrium here, because if someone says we're able to extract energy from a equilibrium without other, without other costs and then we know we should, we should not trust that person. But here, if in realistic environment, when the environment itself is all of the equilibrium, maybe some smartly designed or nicely evolved things, living matter or artificial they can, they can probably extract the non equilibrium energy from the already out of equilibrium environment. But really, can we design all of this using one set of universal theory, I guess the, the answer is no. But theories can still take the challenges. We can take the challenges stochastic thermodynamics we have tools to study thermal fluctuations. In thermodynamics we have also the ability to describe non equilibrium systems, but there are still lots of challenges that I want to highlight here that if you want to design a living matter that is so smart, able to extract, extract information from a time changing environment, recognize the patterns and respond accordingly. You have to admit that we need a theoretical tool that can deal with things that are not just at non equilibrium steady state, but kind of driven by a time dependent protocol. And also, depending on what kind of task you want your little living matter can achieve the system might have a different kind of complexity. And also, at the end I want to emphasize that while we're kind of getting our hand dirty, going into the specific applications or specific performances of interest, it is very likely that we cannot find so universal laws, like, like the laws of thermodynamics as our design principles. But still, we seek to this, we seek, we want to seek the design principle of non equilibrium kind of from non equilibrium thermodynamics and stochastic thermodynamics and apply that or convert that into a language that is kind of getting a little bit closer to the practical design of certain certain certain systems. If I were to just put a little bit tiny advertisement of what we have been doing in the past a few years in my research group, we have been studying kind of different little lovely design principles of different interesting behaviors. The first one that we'll talk about is how can we design the energy landscape of catalyst or enzymes that can really harness energy from oscillating environment. And also there are things when we are interested in driving system out of equilibrium. We also would like to ask are there shortcuts that can allow us to drive a system from one state to another state without delaying in any of the slow dynamics. And also in our group we have worked on a bunch of interesting information theory and biological sensing kind of works. And one that is we actually designed a toy model approval principle demonstration that a single molecule can perform information computation just like a finite determinist automaton or it can be used to recognize temporal patterns of signal and store it in its transient metastable configuration. And there are also other information kind of sensing and information kind of works we worked on. But today we will focus on the first one. That is how can we design or how can we come up with a design principle of good catalysts or good enzymes that can harness energy from a oscillatory environment. So imagine your environment is changing in time. The fact that it is changing in time itself is actually can be considered as a thermodynamic driving force. Can we use that to drive a catalytic network to achieve something that equilibrium or stationary state thermodynamics that can never allow us to do so that is the that is the public that I want to discuss today. And there are other other finds on thermodynamics I will just skip through this slide. So first of all, I want to say that, in fact, it is not new a concept that enzyme or catalyst or molecular machine can harness energy from a periodically changing environment. In fact, in the most interesting, I guess, paradox, in that sense is the flashing ratchet by Perondo, and of course later studied by many, many more. And also, there are stochastic pumps studied by many people I also see people from our, our workshop here. And also there are studies from chemistry, from the field of chemistry people have studied a catalyst that pumped by oscillation. The oscillation can be a temperature oscillation in the environment that pumps your catalyst, or it could be a oscillation of electrical field. There are many, many references here that I cannot fit in here. And also, there are interesting general theories of how does periodic oscillation can give you non equilibrium studies, give you states that mimic non equilibrium steady state that was done in this in this reference. But just to move on, I want to kind of bring bring to your attention for audience in our workshop. I'm pretty sure everyone knows the flashing ratchet, but just for the students in the audience. I would like to just show you that the flashing ratchet is actually quite a simple idea. That is the following. If you have a system on a one dimensional degree of freedom that is under diffusion. You flash between two kind of energy landscapes one is a sawtooth shaped that is asymmetric, and another is flat. If you keep flashing back and forth. At the end of the day, the probability distribution of your your particle or the system will actually have a net drift to one side of the energy landscape. And imagine if you introduce a tiny tout that is a that is upheld to the right, then this kind of this kind of drifted motion is actually harnessing energy from the non equilibrium work that you have performed by switching the energy landscape and store that energy into the potential energy of the particle. But with that many theories and examples and toy model system and more realistic systems, still missing is a kind of a general design principle that is, can we have something to say about what is a good shape energy landscape to actually facilitate this kind of ability of extracting energy from an oscillatory environment, and that is the task that we're going to address today. So, on the left, I'm showing you a very simple model but the theory itself is general it can be applied to arbitrary number of states of catalysts. And also be applied to catalyst of multiple cycles on the graph, but here I'm just showing you a three state graph of catalyst that catalyst can be in the state one, two or three, but by completing a cycle, it will convert one high free energy reactant into a low free energy reaction. And we know that, according to thermodynamics, the affinity of the affinity or the free energy change of the reaction is basically set here and that tells us the reaction, the reaction should always go in the clockwise direction on average. But the question is the following. Now imagine if I am oscillating the environmental condition, maybe that will cause a change of shape of energy landscape between two shapes, or maybe I can fix the energy landscape and just also leave the temperature. Can I achieve, can I achieve one performance, that is, even though at each stationary condition A and stationary condition B, we always want to go from R to P, but after the oscillation, can we have this reaction inverted. So you can see this two energy landscape on the left are just the stationary energy landscape and they are tilted to the right. I just want to find out that you're into the question time. Yeah, I'll be very quick here. So, at the end of the day we basically want to ask how can we design this. And now let me just tell you something really simple that is for a Markov model of catalyst, we can write down the master equation and master equation is basically giving you a rate matrix that is controlled by external control parameter. Here it can be inverse temperature beta. And at steady state you can always offer the steady state probability distribution, and you can basically predict the steady state performance of a catalyst. However, if we want to go to the oscillation limit, let us consider one catalyst defined by one energy landscape but controlled by different temperature in the rate space. Here we define the rate space that is a Cartesian space where each axis is just the one of the reactions reaction rate. And for in step reaction you have a two in dimensional space, and the catalyst energy landscape will basically give you one curve that on each point of the curve. That is basically one temperature and maybe this is another temperature. But if you start to oscillate very quickly between the two conditions. The system will actually achieve an effective non equilibrium steady state that is dictated by a rate vector that is here. So you're basically achieving a new rate matrix. Now imagine if this new green dot is achieved and giving you a different thermodynamic direction here on the left in the art space. I have an ISO surface of affinity or so called free energy and everything above you have positive affinity and everything below you have negative affinity. So imagine if I have this black curve that is on this shape. But if I start oscillating, I will basically achieve an effective rate matrix, or the catalyst will behave as if it is below the surface, that means the reaction direction is is changed. So we have this design principle here using the geometry of basically skip through the details but basically we want to ask how, what is the direction of the banded curve, and it can be described by the second order derivative. And what is the direction of the increasing of the thermodynamic affinity by making a dot product, you can find the general design principle that is only related to activation energy. I will skip through the more general case but basically you can design, use the same design rule for arbitrary thermodynamic performances and arbitrary environmental parameters and their applications we have shown but I probably have no time to discuss and I would like to just end here and if there's any question. Thank you. Thank you I think we have time for one question. So if anyone want to ask a question. Okay, Peter. Yeah well you mentioned thanks for the talk. Right I like this idea very much and you started by motivating that you want to extract sort of generic principles. Can you give an example of a generic principle you now have uncovered here. For example here we basically have introduced well maybe start from the simple one. It turns out if you want to design a catalyst or catalytic energy landscape of the reaction, such that it can harness energy from a temperature oscillation environment. The principle is in a very clean form that is independent of the temperature that you choose independent of the temperature protocol you choose, and it's only having this very simple dependence on activation energies of each reaction and its inverse. So until the day the design principle is the following if you want to have the reaction, invert a spontaneous thermodynamic direction after temperature oscillation, you might want to have the square sum of all the positive reactions activation energy to be very small, but the sum of the other negative reactions activation energy to be very large. And this actually is a pretty counterintuitive thing because you want to want the reaction to go backward but you're making the backward activation energy, more kind of rough. So, I don't know if answers the question. I'm glad that the last thing is not so clear but but but but but but yeah yeah I think you you address it. Well, you know, yeah, I hope to. Well, I'll be able to share this if you want to send me an email but thank you for the question. This derivation is actually just three lines so it's actually quite simple. Okay, if you have a time let's stop it thanks. Yeah. Okay. So I think we're kind of running late so we'll have to move on.