 Hello, everyone. I'm Shaina Dhameja from Arijit K.D. Group, Isar Mohali, India. I'll be talking about real-time tracking of coherent vibrational motion in ground and excited electronic state and its implications in biology. So the motivation behind doing this work is excited state charge transfer, fluorescent proteins, which are indispensable bio-marking tools nowadays. So first fluorescent protein was obtained from jellyfish aquaria Victoria. And then several mutants were engineered. It consists of a chromophoric unit, which resides inside the beta barrel, as shown here. And the chromophoric unit is formed by an autocatalytic reaction between three amino acids. So it fluoresces basically via excited state proton transfer, as shown here, from the chromophoric unit shown in blue to an accepter amino acid via the hydrogen bonding network. And once it undergoes excited state proton transfer, it forms the deprotonated form, which fluoresces. So this energy level diagram here shows the photo cycle in such proteins, which have excited state proton transfer process, followed by fluorescence, and then a ground state proton transfer back to the ground state. Upon photo excitation to a higher line electronic state, it can also undergo photoconversion. While it can be easily studied about the photo cycle and photo conversion time scales, using techniques like fluorescence lifetime and pump-rope spectroscopy, it is quite important to pinpoint how vibrations control such charge transfer processes. On the other hand, there are many small molecule systems where electron transfer processes happen. And while such electron transfer is happening, the molecule undergoes structural changes, like twisting or isomerization. So it is important to pinpoint specific vibrations, which lead to such a charge transfer and what causes a particular deactivation channel. So first of all, I'll be talking about the photophysics of fluorescent proteins, which exist predominantly in the anionic form. The first protein, which I'll be talking about, is a yellow fluorescent protein called venous, where we found that there is a dual emission kind of behavior in this protein, showing emission from both LE state and the relaxed state. While in case of wild type GFP, such an emission from LE state is quite low. It is feeble, which could be because of the unique nature of potential energy surface in this protein, which we started. And another thing which we started is a photo conversion in an enhanced GFP, where we saw that upon illumination with UV radiation, it shows a new fluorescence lifetime component, which keeps on the amplitude for which keeps on increasing upon UV illumination, while the amplitude for the already existing fluorescence lifetime keeps on decreasing. And then upon analyzing the absorption and fluorescence kinetics and fitting it with analytic models, we could find out the time scales for photo conversion. Then upon doing global fluorescence lifetime analysis, we could see that the photo converted EGFP form kept on increasing, while the amplitude for the EGFP, which was the originally existing form, kept on decreasing. We also found out that these two forms were spectrally inseparable, and the fluorescence spectrum was overlapping in the two forms. So we could distinguish them based upon their fluorescence lifetimes. While we could study the photophysics of such proteins, it is also important to ask certain questions. Those are the structural changes and the intermediates which are involved in such charge transfer processes and what is the effect of environment. And also, what specific vibrational modes lead to a particular deactivation channel. So there have been studies on a few of such proteins. And one of those is a photoactive yellow protein, which has been studied by a Tahit-Aharaz group. And they have employed a technique called time-resolved impulsive-stimulated Raman scattering. And this plot here shows the vibrations in the excited state. And these are changing as the system evolves on the excited state, while it undergoes conformational changes from trans to the cis form. So first of all, I would like to discuss how to track the origin of vibrations, meaning whether these are arising from the ground state or these are arising from the excited state. In order to understand this, first of all, we need to understand the photon picture of electric field of light matter interaction, where basically one photon is absorbed and another photon is emitted. In case of Raleigh scattering, these two are same. While in case of Raman scattering, which could be either stoke shifted or anti-stoke shifted, this could emerge stoke shifted as compared to the incoming radiation or anti-stoke shifted. While in case of fluorescence, it is a stoke shift, but the energy level is an electronic level and not a virtual level as there in case of Raman scattering. So while this photon picture is familiar to all of us, it does not give us any information about how the macroscopic polarization was created and what is the order of the process that we are looking at. And that is why it is quite important to understand the electric field picture. So these are the electric field pictures for all these processes which I have shown here, where Raleigh scattering has one electric field interaction followed by the emission of the signal. While in case of Raman scattering, which is stokes here, there are two electric field interactions creating the vibrational wave packet in the ground state followed by a third electric field interaction, which basically interrogates it and then the signal is emitted. It can also be anti-stoke shifted as shown here. In case of fluorescence, two electric field interactions create a population in the excited state and then fluorescence is there. While this is the energy level picture, this can also be understood using double-sided Feynman diagrams where there could be an interaction from the brow or the catch side of the density matrix and showing the same processes here. So after having understood about this, all the schematics which I showed you involved three electric field interactions. So it is a third order non-linear spectroscopy. So suppose you have an experimental setup of this kind where all the interactions are coming from three different pulses. So the signal will be emitted in the fourth corner and hence it is called a box-cast geometry and the signal is emitted in minus k1 plus k2 plus k3 direction. So in cases where the first two interactions come from the same pump, the signal will be emitted in the direction of the probe which is basically called a pump probe geometry. It is a two beam geometry. So all the things that I showed you here were about Raman scattering but now I will tell you what is the difference between a spontaneous and stimulated Raman scattering. In case of spontaneous Raman scattering, the second electric field interaction comes from the vacuum field. While in case of stimulated Raman scattering, this second electric field interaction comes from the external field. So in case of SRS, which was developed by Botfeld and Geller, the two pulses are temporarily overlapped in time. While the Raman pump is a picosecond pump and it is stretched in time domain, it interacts twice with the system. So the interaction E1 and E3 come from the Raman pump and the second interaction comes from the white light probe and then the signal is emitted. While there is another technique called impulsive stimulated Raman scattering, where the first two interactions come from the Raman pump creating the vibrational coherence in the ground state and then the probe interrogates it. So the necessary condition here is that the temporal duration of the Raman pump should be much shorter as compared to the oscillations we want to look at. So there is a spectrally dispersed version of this technique which is ISRS. It is called SDISRS, which I'll be showing in the next slides and it has several advantages over these techniques. So as I told you, the SRS technique has contributions from any unwanted signals because the pulses are overlapping in time, but the signal is there over the entire spectrum. In case of ISRS, the signal is only there at the spectral wings because there is huge rally scattering while there is no background fluorescence and all the range of vibrations from zero to 4,000 wave number can be covered if a very short pulse is used less than seven femtoseconds. And in case of spectrally dispersed ISRS, it has all the advantages of these techniques. So this is the data for a solvent, which I'm showing here. It is for chloroform and after data analysis, we could get three chloroform modes which are shown here. These are Raman active vibrations of chloroform. Then while doing such an experiment, using a solute and a solvent, we could get many, many signals and there can be many, many pathways which will contribute, which I'm summarizing here. So the first two correspond to the non-resonant contribution of the solvent and the outgoing signal can emerge either red shifted or blue shifted as compared to the probe wavelength. While there can be many resonant contributions as well as shown in the blue box. So the coherence can be created either in the ground state as shown here or it can be created in the excited state. So, and we can also observe population dynamics and on top of that population dynamics, we will see coherent oscillations which we want to get. So in this case, a ground-shaped least stimulated emission and ESA and on top of that coherent vibrations will be there. So if we observe oscillations in the SE band or the ESA band, then these can be assigned to the excited state vibrations for that particular molecule. If we get it in the GSB band, these can be assigned to the ground state for that molecule which is under study. So the data which I'm showing here is for a simple molecule which is a diatomic molecule iodine dissolved in carbon tetrachloride and these are the time domain oscillations which we observed. And after Fourier transform, we got five Raman active modes out of which the first two correspond to iodine and the last three correspond to the solvent. So the first one is the excited state mode for iodine and the second one is the ground state mode for iodine and the rest three are solvent modes. We could also see that there is a node in this particular vibrational mode which is for iodine and there is a phase flip around this wavelength which is the absorption maxima for this particular molecule. So that is why we can easily assign it to the ground state mode and this one can be assigned to the excited state vibrations. We also observed that there was a recurrence feature for this data set which we didn't know whether we should assign to the vibrational wave package revival for iodine or but then we figured out that it can be assigned to the isotopal logs of the solvent that we were using which is a CCL4 and there are many isotopal logs and then basically they beat against each other giving rise to such a recurrence feature and upon summing up across the regions which are marked in these boxes we got an ISRS spectrum. As you can see here these two modes which are shown these are quite close in wave number so it is very difficult to distinguish these two modes if the probe is not spectrally dispersed which is an advantage of using this technique as you can see here these appear at very very different detection wavelength of the probe. As you can see here that there is a split line shape here and the difference is three wave numbers. So three wave numbers basically corresponds to 10 picoseconds in time domain and that is why we observed a recurrence feature which I showed you in the earlier slide. So I told you that in case of 30 generate kind of ISRS experiments there is a huge rally starting which is shown here and then I told you that there is an advantage of using a white light probe but using a chirped white light probe has an additional advantage which I will tell here. So these two schematics basically show a degenerate kind of ISRS experiment and the signal can emerge either red shifted or blue shifted and these two pathways contribute equally while in case of a white light probe which has a flat spectral profile so the red part of the probe and the blue part of the probe contribute red part and blue part of the probe they contribute equally giving rise to a signal which is emitted at same detection wavelength as you can see here but if the probe is not chirped then this blue part and red part these will arrive at the same time then these two pathways will destructively interfere and we won't observe anything so it is necessary to have an optimal chirping of the probe to get some vibrations or to enhance some specific vibrations so then we extended this work to polyatomic molecules where we could capture some of the vibrations this is a data for Nile Blue and Methanol where we could capture the ring breathing mode and compare it with DFT calculations as well so the same technique can be extended in the excited state so whatever process I described can be done in the excited state also there will be an additional actinic pump followed by Raman pump and white light probe in order to study the structural dynamics in the excited state and right now we have been doing this technique and we have been implementing it on these molecules so the first one is a push pull steel beam where there is an electron transfer from one end to the other end while such electron transfer is happening there is isomerization or twisting so we want to know which vibrations lead to isomerization or twisting in this molecule and then we will be extending it to photo switchable fluorescent proteins which photo switch upon irradiation with a light of particular wavelength and the chromophore also changes its conformation and we want to track the structural dynamics so I would like to thank my lab members Aizel Mohali and the organizers for giving me this opportunity to deliver this talk. Thank you. Thank you, Sharina, for this very dense talk with a lot of techniques, maybe a lot of work. The paper presentation is now open for questions from the audience if any. As usual you can use the chat or you can just switch on your microphone and ask. So I can break the ice again. When you go from iodine to nile blue or more complex structures is there a problem with having too many modes or so too dense spectra or is it rather easy for you to deal with three n minus six vibrational modes? Yeah, as we go on from diatomic to polyatomic molecules it is like difficult since there'll be many, many vibrational modes and then those have to be assigned. So we have to simultaneously do theoretical calculations and then assign which particular modes correspond to which vibration and then it becomes more complicated as the system undergoes some kind of structural changes as well. So yeah, it is complicated. Thanks. I see a question from Giovanni Bressana. Giovanni, can you switch on your microphone and ask the question or should I read it? Okay. So I will read it. Could you please expand on the white light cheer factor? Kukura shows how cheer-fed white light washes out of the coherent sun pretty good. Have you got the question? So apparently the white light cheer washes out of the coherent sun pretty good. Okay, okay. So yeah, it has been shown but also Sandy Ruhman's group and Shol Mukhamil's group also have shown that it can also help in enhancing some specific vibrational modes which cannot be otherwise detected at all. So there has been a lot of theory. So Sandy Ruhman's group has shown how only in the presence of chirp, we can observe it as I'm showing here that if this blue and the red part of the white light appear at the same time. So there is no chirp in the white light probe. In that case, we won't detect these vibrations. So for some particular vibration mode, it is quite necessary to have optimal chirping of the probe. So it has been shown. Okay. I think thank you, Shalina and I think it is time to go to the next speaker. Who is Uriel Murzan, sorry. We talk about X-ray absorption signature of short hydrogen bonds. Please Uriel, you have seen your screen and everything is fine, go. Thank you very much. Let me move this, it's blocking my screen. Well, anyway. So I'm very happy to be here. Thank you very much to the organizers for the invitation. I mean, it's a contributed talk, but yeah, I mean, for the accepting my presentation. So today I will talk about the topic that I'm very, very interested in, which is the ultrafast dynamics of electrons and ions around conical intersections. And in particular, I will talk about the role of the strong hydrogen bonds in determining these dynamics. So, okay. So conical intersections are extremely important for photochemistry and photo physics. They are very, very fundamental to determine what happens in photochemistry and photo physics. So basically these are just points in the energy landscape in which we have a degeneracy between two or more electronic levels. So in these points, we can have a non-radiative decay, non-radiative relaxation from an excited state to a lower lying excited state, or to the ground state, okay? So of course, the presence of these conical intersections in the energy landscape, determine a lot of what's going on, what will happen in the photochemical process. So one of the things I am very, very fascinated about is with the possibility of controlling the shape and the position of conical intersection, and hence controlling photochemistry somehow. So today I will talk about one specific way of doing that of approaching this idea, which is by adding strong hydrogen bonds. So I will show how the presence of strong hydrogen bonds can modify the shape of a conical intersection and the position of a conical intersection. And then I will show some optical properties that are derived from this phenomenon, okay? But first, okay, this is more or less an overview of the talk, okay? So first let's discuss what is a strong hydrogen bond. So normally we think about hydrogen bonds like electrostatic interactions, intermolecular interactions. So for example, in the example in the screen, we have some donor that is attached to an hydrogen and this hydrogen has some electrostatic interaction with certain acceptor, okay? And the potential energy surface projected in the hydrogen bond coordinate looks more or less like this. We have a deep well and another well that is not so deep, okay? But in nature, we can find hydrogen bonds with a wide range of strengths and in particular the strongest hydrogen bonds, they look much more like a three-center covalent bond than an intermolecular interaction. So they look more like just one single molecule than the interaction between two molecules, okay? So at the end, we have a broadened single well potential. Like in practice, this is more like a broadened single well potential, okay? So very recently, in the group of Gabriel Kaminsky-Schriele, they observed that in some protein aggregates that were also discussed in the talk of Chakramu, these protein aggregates, which have no aromatic residues or external groups, they show certain fluorescence, some fluorescence. And this is somehow surprising because we normally expect that in biological matter, the fluorescence arises either from these molecules called aromatic amino acids or from some external group turned here and compound is still fluorescent. So this was kind of surprising. And together with Ali Hassanali, they studied the molecular properties of these compounds and they observed that in their particular case, these compound contain a very strong hydrogen bond network, okay? These strong hydrogen bonds that I showed you before, well, they are very abundant in this combo, okay? So in order to see if there is a connection between the non-aromatic fluorescence and the presence of strong hydrogen bonds, the group at Cambridge synthesize a crystal of elglutamine. Elglutamine is just a non-aromatic amino acid, a simple amino acid. And this crystal contains an hydrogen bond network, a network of hydrogen bonds, which are of standard strength, okay? But upon heating, they convert the elglutamine system into something that we call elpyroglutamine ammonium. And this elpyroglutamine ammonium contains a very strong hydrogen bond net, okay? So both of the systems are non-aromatic. The main difference is that in the case of elglutamine, we have a normal hydrogen bond network. In the case of elpyroglutamine ammonium, we have a strong hydrogen bond net. As a consequence, and at variance to what happens in the case of elglutamine, in the case of elpyroglutamine ammonium, the system is fluorescent in the visible range, okay? So in principle, this tells us that there should be a connection between the non-aromatic fluorescence and the presence of strong hydrogen bonds, but which is this connection? So together with Gonzalo Diaz-Miron, who is a PhD student in Argentina, so we did some ab initio non-aromatic molecular dynamic simulations in which we computed the non-radiative decay probability as a function of time evolution in the first excited state. And we did it for the fluorescent system, but tuning the hydrogen bond strength to different strengths. Basically, for example, in the blue curve that you see in the screen, we have a very weak hydrogen bond, and you see that the non-radiative decay probability increases very fast during the first femtoseconds of the simulation, okay? But when we gradually increase the hydrogen bond strength, the non-radiative decay probability decreases, and when we reach the level of a strong hydrogen bond, the non-radiative decay probability is very low, and hence the lifetime in the excited state is higher, and hence the probability of fluorescence is high. So this shows the link between non-aromatic fluorescence and the presence of strong hydrogen bonds. So I will connect this with a spectroscopy that I'm extremely fascinated about. So in the first talk of the day, Professor Julia Weinstein, she mentioned that ultra-fast X-ray spectroscopy is marking the down of a new era for photophysics, and I really agree with that. So I think that there's many, many opportunities, new opportunities. If we apply X-ray absorption spectroscopy, which is really, I mean, it's really a challenge, but there's many, many physical insight that can be gained from this spectroscopy that cannot be done with classical techniques. So we recently developed a very, very simple approach to simulate this kind of spectrum, okay? And we are also currently, we are speaking with different groups that work in free electron lasers to be able to perform the experiments associated with the simulations that I will show you in a couple of slides. But basically, these simulations are basically some sort of a motivation for the things that could be able to, that we could be able to observe with an experimental spectrum like this. So basically, this is a pump pro-technic of two poses where we have at first UV poles that excites the system to one of the excited states, and then we probe the system with X-ray, okay? The main advantage, I mean, at least what I think is the main advantage is that the high energy of the X-ray absorption spectroscopy allows us to have a very, very high time resolution and allows us to probe really ultrafast processes. So we can probe not only the ion dynamics, the nuclear dynamics, but also at the same time we can probe the electron dynamics. So what we want to do with that is to probe the wave packet dynamics of the system when it is crossing a conical intersection, okay? Basically, the simulations I will show you are, you will see X-rays in the K-edge region, which is basically all of the transitions that come from the 1S core orbital and go to the balance orbit, okay? So basically the idea would be to probe the system while it is decaying through a conical intersection, okay? Well, I'm going the other way. Yeah, so this is an example of an ultrafast X-ray absorption spectroscopy spectrum, okay? This is for the case of L-bluetamine. This is the system that is non-phorescent, the system that decays to the ground state. So let's try to understand what we see in this spectrum. So first we have two bands, two big bands, okay? One is associated to the excited state. So this is the UV excited state, okay? And this is, so this is the X-ray response of the UV excited state. And this is the X-ray response of the ground state, okay? So we see that at the beginning, we excite the system to the first excited state. So all of the population is in the first excited state. And as time goes by, the system decays to the ground state. And the reason is that, once we excite the system to the first excited state, the system moves forward to a very close by conical intersection that allows the system to relax to the ground state, okay? So in this first region, in this first region, we see the crossing between the two bands, okay? And we see the decay of the excited state into the ground state, okay? And what happens in the case of the L-bluetamine ammonium, the system that contains a strong hydrogen bond? So as I told you at the beginning, because the ground state energy is broken, okay? In the case of the strong hydrogen bond, the conical intersection lies further away and higher up in energy. So it's more difficult for the system to reach this point, okay? And hence the lifetime of the system in the first excited state is much higher. So at least in this simulation time, we don't see a decay of the bond to the ground state. And maybe I skip the part of describing the spectrum. So this line, the horizontal axis is time and the vertical axis is energy, okay? So with this technique, we can differentiate the presence of a strong hydrogen bond with respect to a standard hydrogen bond, okay? So I think I don't have enough time to describe this part in detail, but I just wanted to mention that we developed a very simple and supervised machine learning approach in order to study both the electron and the nuclear dynamics along the conical intersection. So what is the system doing in order to transfer to pass the conical intersection, okay? So from a nuclear point of view, we observe that most of the motion is located around the hydrogen bond, okay? Again, this is the glutamine, the system that contains the standard hydrogen bond. And the main motion that the system does when it crosses the conical intersection can be separated in three parts. So one part is an intermolecular separation between the two hydrogen bond molecules. Then there is a reduction of the strength of the hydrogen bond coupled to a CO stretching in the acceptor part, okay? And finally, there is a planarization of the amide carbon, okay? That is in the acceptor part. So all of these motions need to happen in order for the system to decay to the ground state. So if we hinder this motion, we would basically re-retard or delay the decay, okay? So the most important part of this motion is the motion of the hydrogen bond between the hydrogen bond molecules. And this is what I'm showing you here. So basically in the graph of your left, you see three histograms. So these are the histograms showing you the hydrogen bond strength for the ground state in black the excited state in orange and all of the points that are close to the conical intersection in blue. So there's something very interesting here. So when the system goes from the excited state to the ground state, the hydrogen bond strength decreases. So the hydrogen bond distance increases, but first the hydrogen bond strength, before it goes to the ground state, it first increases. In the conical intersection, the hydrogen bond strength is the highest and it is entering the region of the strong hydrogen bond. So here in pink, we see the region of strong hydrogen bonds. So we see that in the conical intersection, the system transiently behaves like a strong hydrogen bonding system, okay? And in the graph of your right, you see the same histogram, but evolving time. So you see how the system smoothly first go to the strong hydrogen bond region and then it goes to the ground state hydrogen bond, okay? So this is very interesting because this shows that in the spectrum, we can not only capture the presence of stable hydrogen bonds in the ground state, we can also capture the presence of transient hydrogen bonds that basically occur in the crossing region, okay? So when we see this crossing region, we are seeing the presence of transient strong hydrogen bonds, okay? So I think I don't have enough time to talk about this in detail, this is basically the electron dynamics that also is observed in the spectrum, but I think I will skip this part and I will go to a conclusion. So strong hydrogen bonds can delay the passage through conical intersections between the ground and the first excited state and this retardation increases the lifetime, of course, in the first excited state and this increments the probability of fluorescence. And also the PAMPRO UVX ray spectroscopy reveals the electronic and nuclear, well, I didn't talk about the electronic part, but just believe me, it does. The electronic and nuclear ultrafast dynamics along conical intersection showing a unique signature of the presence of strong hydrogen bonds, both stable and transient. So these are the people that participate in this work, mainly I wanted to thank Gonzalo who did most of the calculations and Ali and this all was done with the LIO code. LIO is a code that we develop in the University of Buenos Aires, it's a DFT code and it's free and open source and if you want to download it, this is the webpage. So thank you. Thank you for this very interesting and fascinating presentation, which is now open for questions. Again, you can write in the chat or simply switch on your microphone. I have a question for Kurian. I have a curiosity or possibly a nasty question again. You convinced me that when you go from a conventional to a strong hydrogen bond, you increase the lifetime of the excited state and this is good because they suggest that maybe the state can fluoresce. But now the question is the state that you calculate, the fluorescent state in which spectral region will fluoresce because for me the most strange part of this result on anomalous fluorescence of this system is that again you observe fluorescence in the visible region and not in the far ultraviolet. So can you clarify this point? No, I mean it's just so the fact is that in the experiments we see that this fluorescence is visible. I mean it shouldn't necessarily be the case but in principle this is what they observe. This is not, so we didn't focus on this part on the part of where the fluorescence comes but what I can tell you is that this, so the part of the spectrum where the fluorescence the fluorescence arise is very consistent with also with the calculation. So it's something that we are capturing in the calculation. So the explanation why it goes, why it falls in this part of the spectrum and not in another is something that I could not answer and I think it's a very interesting thing to try to disentangle to understand. But you say that you calculate not a less in the proper spectral window. Yeah. This is already very interesting. Yeah. Do you see any more questions? Any more please and it seems not. So if there are no more questions we can thank Uriel and all the speaker of this last session.