 We'll begin in one minute. Okay, so it's one, we can start. If you're ready, Selina, with the last lecture on astroparticle physics. Thank you very much, okay, so can you see it? Yes, I can. Right, so I have to leave at seven for me, so we're just in one hour. I still have outstanding questions and I think, so I wanted to reply to them during the talk today, but I still have a lot of slides to cover, so I think I'll try to go first to the slides and then maybe we can answer the question. There's just a one on nucleosynthesis, because for me, it turns out that it's well, I mean, the helium-3 is very well measured, but in terms of finding the, I mean, constraining the parameter, which gives you the quantity of variance, it's actually too large, so as far as not written on the plot, I think. But I'll, if there are questions, I mean, I'll try to answer the questions in the slide before sending them. If there are questions or so maybe, which I didn't answer, people could perhaps send them to the team who can perhaps collect it for me and then I can answer it separately if that's okay. We'll put them in the slides. All right, so yesterday, we were talking about dark matter signatures and we spend a lot of time on indirect detection, which is one method where potentially we can access to the dark matter mass. And I hope you realize from this that it's a great avenue, but it's indirect. And so this means that the dark matter has to either decay or annihilate. And so the question you may ask is, what happened if it doesn't annihilate or what happened if it doesn't decay? And so we need something else and that's something else that's our technique. It's called direct detection. Now technically, there would be a third technique which is producing dark matter particles at LHG. Initially I wanted to speak about this, but I think I will not have time, so I will just make a tiny parenthesis at some stage. So direct detection is really a method which is dedicated to dark matter, but originally it's coming from an idea which was developed basically for detecting neutrinos. So the very first people to propose this technique were actually Drukie and Stodolski. And you can see they were thinking about MEV neutrinos. And the idea was basically the neutrinos can pass through a detector and then basically will date what is inside the detector and you can detect the interaction. So this is essentially the beginning of a neutrino experiments to some extent and also in fact, dark matter detection. So why is that? Because I mentioned before when Huit and Nier when they were considering the argument, even though they were thinking of, I mean, the conclusion of that paper was that the dark matter has to be heavier than a GEV. In reality, they were taking natural leptons, so basically heavy neutrinos. And so once you know about the fact that you can have heavy neutrinos explaining the dark matter, and then you know about this technique, then you can put them together and this gives a paper which Goodman and Witten, so it's Ed Witten for those who have done a lot of maths and I mean, mathematical physics, you would know him in that context, but actually Ed Witten has done a lot of physics paper and phenology paper in a sense. So this paper is very, I mean, this is basically the beginning of direct detection for dark matter. And you can see this was 1985 and I can maybe read it for just to make sure people see the benefit of it. We consider the possibility that the neutral current neutrino detector recently proposed by Drukie and Sholoski could be used to detect some possible candidates for the dark matter in galactic halos. This may be feasible if the galactic halos are made of particles with coherent weak interaction and masses between one and 10 to the six GV. So this is perfectly aligned with the Lee and Weinberg argument. And then there's something important, with spin-dependent interaction of typical strength and masses of one to 100 GV, or strongly interacting particles of mass, one to 10, 13 GV. The point here, they started to mention interaction and in reality from this paper, we know there are two types of interaction, spin-dependent and spin-independent. This has been the status for, I would say, up to the 2010 and then something happened and you will see and now we know that they are actually possible, I mean, many more interaction possible. But for a very long time, this was it. You look for dark matter, you can have this kind of direct detection experiment and you find candidates which are, if you have an improton, and they have interaction which are either dependent on the spin of the particle and also of the material if you want. So we'll be, sorry, the interaction rate would be proportional to the atomic number of the material in the detector or spin-independent. Sorry, I'm not sure which one I just said, but it could end together. So it's spin-dependent and spin-independent. Spin-dependent and spin-independent. Another hard day in the life of a head of school. So the principle is fairly simple. We live in a halo, we are on Earth. Earth is, I mean, we live in a solar system and the solar system is evolving through the halo. We know our velocity is actually about, whatever, I mean, between 200 and 300 kilometers per second. So we're going through the halo. And as we're going through the halo, well, at some stage, we should actually interact with dark matter particles. Our dark matter particles should interact with us, should I say. So in principle, there should be an interaction between the dark matter particle and the material if in a detector, if you have detector on Earth. So what do you expect to see? Well, if a dark matter is heavier than a proton, then it can give some kick to the material in the detector. And so you can, you should see a recall energy and you can see these through various means. So you can, for example, see unifications, ventilation, you can detect phonons or you can detect heat. So there are various ways to detect this interaction. But here this method assume in principle that the dark matter is heavier than a proton. And this makes sense because if dark matter is too light, it doesn't, it can interact with a proton but would not give any momentum to the material in the detector. And so it's undetectable. I mentioned before that I was interested in light dark matter when I was, a long time ago when I was a PhD student, my whole point was actually to demonstrate that we could avoid the Lier-Wenberg limit. And you can see now that when I was doing this, I didn't need to worry about this technique of direct detection at the time because it was lighter than a GV and therefore there was no signature in a direct detection experiment. That is nowadays, this is no longer true. Now you need to worry about it because people have found a way to use a detector to measure those interaction even with electrons. Sorry, well, even if, I mean, not just looking at the proton but with electrons. So I mentioned the phonons, the scintillation unization. Some experiments are actually using other technique like separated liquids. And they are, as you can see on these slides, which I borrowed from an experimentalist, you can see that they are plethora of experiments. So the question is, are they giving something consistent in their search for dark matter? And or are they in disagreement? Now what you can see here, so on the main diagram where they are the arrows, you can see a few examples in particular between phonon and unization. On the right, you have a few experiments which are very important, CDMS, Edelweiss, Germanium. So those are basically using Germanium but I don't need to explain too much at that stage. Then at the bottom of a diagram, you have Zippin, Xenon, Lux, and you have a few including RDM using argon. So Xenon is using, obviously Xenon. Argonne is obviously, I mean, RDM is using argon and a few others are using different gas. Now on the left at the bottom, so the scintillation part, you will see some very famous experiments. You have Dama Libra, which I'll explain, I mean, I go back there. You have Anais, Deep, Clean, Xmas, and so on. All of those are extremely important because they actually very sensitive to the low dark matter mass. In the same way, actually experiments like Crest and the Rosberg and so on are also sensitive to the low dark matter mass. So there are plenty of experiments and as I said, the question is which one? Well, are they compatible and which one has more potential actually to detect dark matter? So this is a map just to show you something which is going to be relevant eventually. You see that wherever you have a pin is wherever there are experiments. You see that it's not all of them but the pin shows basically where they're usually located. Usually you need to put them in, you need to shield them and the best shielding, sorry, you need to shield them because otherwise you can be contaminated by cosmic rays which come from the atmosphere. So the best way to shield them is actually to put them on the ground. And if possible underneath a mountain like this, you gain a lot of fuel. So this is a case in particular in Italy but you have other places where, for example, in England, there is a mine which is 1000 kilometers so really deep where the one experiment is located. And you have a ferraments in Canada so an issue in the United States. Now there was none really for a long time in the southern hemisphere. Now there is a proposal for one in Chile and there is also one which is being built. I mean it's actually built but it's actually going to be now installed from Italy to Australia in place near Melbourne if anyone knows Australia in Victoria. So it's also in a mine and it's also a very deep mine so it's going to be very interesting. For me it's interesting because it's a gold mine so I was wondering if one could have a benefit every time you go and visit but I doubt it. So here is one of the experiments. Sorry, quick question in the chat, a quick I guess, it is possible to know if that matter is charged through detection. Josh, to detection? Through if detection can tell us about the charge, okay. So dark matter has to be almost neutral and so probably if it has a charge it has to be tiny and there are experiments which can detect this and I will mention it very briefly later on. So the experiments which has really revolutionized the field is called Xenon. The one really which was important for the field was Xenon 100 so it's on the left but the precursor of Xenon 100 so the prototype if you want the baby Xenon is on the right, Xenon 10. So it's an experiment which contains 10 kilos of Xenon and the other one is 100 kilos. On the photo with the little prototypes of Xenon 10 you can see it's very tiny it's actually holding on the table and the person with the hand on it is the PI of this experiment, Elena Aprile. Maybe some of you know Elena. And so this was a groundbreaking experiment for the following. So it's, I think it's the first experiment which was using and I'm not an expert so I hope I'm not saying something incorrect but I believe it was the first really using a combination of liquid Xenon and gas Xenon, Xenon is form of gas and the principle is you have a dark matter particle going through the detector having an interaction with liquid Xenon and then you have some scintillation out of this so you have what we call the S1 signal is the primary scintillation signal but then you have a drift of the electron and then eventually they interact with the gas and that gives you a secondary signal. Now the fact that you had those two signals extremely important basically for making the measurements because you basically was providing a way to control the background and to determine what kind of interaction happened what kind of particle has interacted in the detector. So what I didn't say yet is that when you want to detect dark matter a dark matter is nothing else than just a neutron if you want, it's more or less the same for a detector and so it's a, a neutron is basically a wind if you want so you need to discriminate there are plenty of neutrons coming from radioactivity so you need to discriminate your background you need to know the background extremely well and it's a bit like what I said for with indirect detection with bananas in cluster with a potassium emission. Now if you want to discriminate the background extremely well you need to reach a level of sensitivity where you can tell if someone comes for example with a banana in the lab. So that's absolutely forbidden. So you need basically to protect the experiment so that only the particles which do not interact strongly first of all interact in the detector. So you need to get rid of all possible cosmic ray or possible radioactivity and then eventually you can't really discriminate which I mean there are some particles still from the background which are interacting you need to be able to tell whether it says background or this is a dark matter interaction. So it's a very hard exercise and here is the evolution of Xenon so I show you Xenon 10 and 100 and the next stage is Xenon one ton which is happening and then the next stage Xenon 10 ton of Xenon so in a detector. And so this is the image that you see here of a control room that I took this from Twitter. So on the left this cylinder is basically where the experiment is. You don't put the experiment like this you need to shield it as I say. So to shield it you actually put it in a volume of water. And then you have a correspondence to the basically the control rooms you can see. I think it's very telling. You can see the level of experimental setup you need to succeed in such an experiment. So it's very impressive. But it's also as I will show you it's also the experiment which has revolutionized the field. So I think this is really instrumental that this experiment continues and be supported. So I show you the principle for Xenon. Xenon one ton is on the top of this slide and it's basically the same principle like Xenon 100 or Xenon 10. But below is the same with argon. Instead of having Xenon you have argon. There are few differences but it's mostly the same principle of having an interaction in the liquid and then you have drifting electron sorry which are going to give you another saturation senior. The difference here is that you have some light coming. So what is the UV light basically? And I don't think you have these things in Xenon but obviously this is giving you another source. On one hand I give you more power to discriminate on the other hand I give you some background. The other thing that you can consider is so maybe I'm going too fast but you have to see the jury as a question so I'm just going to say one thing and then I'll answer jury. You have to put these experiments on earth obviously so what happened? Well, earth as a side is moving. So as it moves it's not just moving. I mean, first of all you have solar system motion and then you have the earth motion with respect to the cell. So in reality you have to take into account the fact that you have two velocity as we circulate as we move in the galaxy. And so what you see in this diagram and take the diagram ways. So at the bottom of the slide there is the high and the lows. What you can see in the highs is when basically the cell and earth motion are basically in the same direction. And in this case you have a maximum basically I mean for the speed of earth. So you actually have more chance to encounter dark matter particles. So what happened then is that you have a high in the number of interaction. But then when the earth has rotated so for example in December for the northern hemisphere then you have the earth and the sun which are basically orbiting in the opposite way. And then in this case you actually have less when the interaction rise is basically lower than otherwise. So you have oscillations in the interaction rate. So the high happened basically in June and you can see it from this diagram near the halo. You see the oscillation. So if you pay attention to the x-axis you can see. So the maximum is expected in June, the minimum in December, maximum in June and so on. So this is basically what we call the annual modulation signal that is expected from the fact that the sun and the earth are moving in a dark matter halo. And so I'm just going to show you this and then I will take a question by Dewey. I'm going to show you if we can work. It's the same thing but you can just see, I mean I'm basically all the system involved in all these changes. The probability to detect a signal. So in principle, so this is northern hemisphere. Oops, sorry. And then you can imagine if you're in the southern hemisphere then you should see the opposite. And so that's why having an experiment in Australia for example is very important and you do want to have an experiment which is sensitive to this annual modulation in order to determine whether potential signal which would be seen in the northern hemisphere would be actually correlated with what you see in the southern hemisphere opposite. If you do then it's very likely that it's a dark matter of particles interacting with the detector. So Dewey maybe, while Dewey unmute maybe I can say. So the pioneer of this technique was actually a drukkie, frieze and sperger. So it's a paper that you see at the bottom of the slide. And the image is the one experiment, the first experiment which has done this which is located in Italy. And it's called Dhamma and the next version was Dhamma-nibra. Okay, yes. Can you hear me? Yeah. Yeah, so maybe this is a silly question but like when the direct detection is defined it's always the WIMP interaction with the nucleons. So in case this is true for Majorana cases but like the last time we had if there is a Dirac or a complex case in that case is the anti-WIMP interaction also. I mean, how does one distinguish then between the WIMP and the... And if it's taken into account for the Dirac detection cross-section as well. Yeah, yeah. I wish I will have something to show you to answer this question. So maybe we can wait and then I'll mention you again. But essentially for direct detection for the experiment it doesn't matter, right? I mean, they do the experiment if they have an interaction that's great. Then that would be the job of the theoretician and terminologies to understand what kind of particle interacted and then for that there is a framework which I will show you after. Okay, because for example, xenon plots they just plot the WIMP nucleon cross-section against the mass. So they don't... So I was just thinking whether it also includes the anti-WIMP somehow or fold it into it. You need to translate it indeed. So depending on... Yeah, yeah. For example, if you take a neutralino which is a Majorana. So it's a supersymmetric particle. It's a Majorana. You can make the prediction for interaction in the different detectors. And you will find basically a certain number of interact... I mean, you can make a prediction. Now, if you have an interaction in one of the detector, then you can see if it's compatible. Now it could be that several types of candidates is compatible with this number of interaction. And so I think your question is how do you discriminate them? And I suspect the answer is you can't just by using direct detection, you would have to see... I mean, if you have a detection, it's miraculous. It's already fantastic because you would have proved that that matter as particle exists. So then the question would be indeed, what is the kind of particles? I will show you there is a table which you can use nowadays to see if it's compatible. So you can see which model works. But then after that, you would need basically other experiments such as... We'll see if you can produce it in LHC, if it's heavy enough, or for indirect detection to see whether there is a signal. So I don't think that would be one experiment which tells you the nature of that matter. It would be really the combination. And in fact, I had many papers 10 years ago where we were playing this game and seeing, can you have just one signature in one experiment or would you necessarily have three signatures? And there's always a way to have maybe just one signature or three signatures, there's always room. But that's a ferretian playing when you have an experimental signal, that's the opposite. And the question would be really, there would be many candidates which can fit as you close. But then the question would be whether the rate of interaction, so the couplings that you would need and so on are compatible with all the other constraints. And I suspect when we do this, because now we have so many data which are excluding most documental candidates, I suspect when we do it, there won't be so many candidates. So it might not be so hard to discriminate. I mean, to find the type of particle which is basically giving you a signal. Or we will see, that would be a fascinating time. Okay, thank you so much. There's also another question if you have time. Yeah, I think you'll take it, yeah. Yeah, it's about the last argument you made about why having Australian or Southern Hemisphere experiment, it's important. Okay, Damian detected seasonal changes. Let's say the Southern Hemisphere also detects seasonal changes. We just proved that there is something that goes with season. So if we're missing something experimentalist, experimentally, that doesn't change, you know? There are different things though. First of all, agree in principle. However, first of all, you're not using the same experiment as Dahma. You have different crystals. It's not the same. Second, so it's not the same experiment. So it could be a long time ago when Dahma was making the first claim which was in 1998, a strong claim in 1998. A lot of people thought maybe this is just systematic electronics, whatever. Now this, you know, Dahma is Dahma and it's another experiment and in fact, there are three or four other experiments looking for this. So it would be, you know, if you have, if everyone sees a signal basically, it's not just in two atmospheres in all those experiments. If everyone sees a signal, it's not systematic. It's not just electronics. It's already stating that there is something. Now, you're right. It doesn't have to be a dark matter in principle. It could be, for example, you could say, oh, it's neutrinos. I mean, I had a student a few years ago who did this paper trying to say, well, it's neutrinos interacting in the rocks and making neutrons. And then neutrons, as I say, are fairly similar to a dark matter candidate. The thing is you need to really explain the interaction, right? You will really need to explain the oscillation pattern. And if it's in two hemispheres, I don't think that would be really obvious to explain, even if it's, for example, neutrino interacting with neutrons. The other thing is people were saying, well, forget about the neutrinos, but it could be just a problem of shielding with neutrons. Again, that's not seasonal. And with other experiments, you can test that. And regarding the seasonal aspect, it could be also, I mean, people mentioned muons, atmospheric muons, but that's not actually giving you the annual modulation. And also you always have to compare the, I mean, you always have to look at the magnitude of a signal. So in reality, I think none of them will be able to explain the annual modulation except dark matter. So I may, you know, once research, maybe one day I'll tell you, you know, you were right. But I think it's very likely that if you see the two signal, it's just dark matter for sure. All right, so it's more of the same. I just wanted to show you that actually a Chinese experiment. And then as I said, there is Anais, a scientist, Saber, and so on. But I just wanted to draw your attention. So we're moving in the dark matter halo and we are moving against the dark matter wind, wind, depending on what it is, which is coming from a region of the sky, which is called sinus. And I hope you can see where sinus is in that with respect to the Galactic Center. But I just wanted to say it's a funny region, actually. There's a lot of things happening there. And so I'm starting to think that maybe, you know, maybe a dark matter detector can do more than just detecting dark matter, in fact. But this is really new development. So we'll see. But it's interesting that we may be sensitive to physics in this region of the sky, which we can't access otherwise. I think I will skip that because I will be late, I think. Maybe I'll just quickly say, I mentioned the scintillation. The question is what is the efficiency of a scintillation? And that was a big problem for Xenon for many years. And in fact, I did a paper with a PhD student where we were showing that they were making hypothesis. They were pushing extrapolating, basically, their scintillation efficiency a bit too much. And they were making claims, which at the time, I thought was not necessarily correct. And he eventually very advised this, and now we trust the result. But the one thing I wanted to highlight is you can be a fair addition. And because you have access to the data and those experiments, actually, they usually give you a fair amount of data, you can actually test whether you agree with them. So it's very interesting because then you can really analyze if you have your favorite theoretical model, you can actually use the data and analyze them yourself. This is another one I wanted to show you, Cogent, because I just wanted to make a point. So Cogent is an experiment which is a little bit different, but it's using Germanium semiconductors. I mean, I'm not an experiment at this, so I don't want to comment too much on that. But the only thing I wanted to say is that I told you, you have to discriminate your background, you have to know your background very well. Xenon, actually, because they have the S1 and S2 signal, they actually know what is their background. They can actually say, oh, if it's an electron, it should be in the upper region. If it's a neutron, it would be somehow, and so on. Cogent couldn't, Cogent just takes all the background and then the problem is how do you extract it to find whether there is an interaction with dark matter? And the way they were doing it was actually fit, they were having an estimate of their background and they actually fit their background. So what they were doing at the end is taking the data, removing what they think belongs to the background and then eventually what was left was meant to be dark matter. So they claim, based on that, they claim that they had the evidence, to see my evidence for dark matter. And again, because the data in this case were fully available, which was really hot off for the experiments to do that, for a collaboration, we could actually redo their analysis. And what we've done is just realize that their distribution, the way they fit their background was had some assumption and we could use a different distribution. And by doing this, and we got a better fit to their background if you want, and by using this better fit, we realized that there was no signal for dark matter. So Kojian didn't really last, at least at first we published this paper saying, we don't think there is a dark matter signal. And then they've redone some analysis and so on and they basically concluded something similar. So just to show you now, I told you, well, we have all these experiments, are they compatible? Are they telling you the same story? So what they measure is a wimp interacting with a nucleon, so at least a proton as a cell, because it has to be of a few GV. So you can plot, you can plot basically the interaction rate or the cross section with respect to the mass of dark matter. You can see a sharp cutoff. The red, you can see a red curve, which is basically corresponding to Darwin. So Darwin is basically the Xenon-Anton, if you want. You can see the sharp cutoff because precisely dark matter has to be heavy in order to interact with, to leave a strong signature in your detector. So what you can see is mass experiments are sensitive to heavy dark matter with this technique. So you can see all the curves being basically more or less vertical at 4, 5, 6 UV. Then some of them basically go up. So everything which is up is excluded. Everything above was lined is excluded. And you can see you go down. So COOP was not extremely good. Zeppelin, CDMS, CDMS was actually better than COOP and Zeppelin, even though it was earlier. Super CDMS, Xenon-100, and you can see now why Xenon-100 was basically revolutionizing everybody in the field because that was 2012 and it was much deeper than everything else. I mean, the curve was much lower. So the interaction rate they were probing was much lower. And then you can see the rest now with Darwin. So the Xenon technique still beating every other technique. You can see that some experiments are trying to probe the light dark matter range with, again, interaction with nucleons and his Pico and Super CDMS in particular, which is really putting a dent in the cross-section for dark matter particle at low mass. Now, in 1998, as I said, Dama made a claim which was basically was probably dark matter has been found. They had at the time a signal which was around 100 GeV. And eventually they moved towards, so you can see the pinkish, it's not pinkish, but yellow pinkish region of Dama. There are two, in fact, so there was a first region and then eventually a smaller region or different region at lower mass, larger cross-section. And then you can see Cogent with the Woodley claim, which was so incompatible with Dama, but in the region of low dark matter mass. There was a strange claim, which was not bad by Christ, but some people using Christ data that they had a signal for dark matter signal. But the point is all those evidence, so Cogent, Christ, Dama, they actually excluded and they were actually excluded, right, by Xeno 100, but also in 2011, they were already excluded by Edelweiss. So maybe I'll skip the explanation, but Edelweiss was basically a group which was initially started in Dama. And then, so it's mostly a French group and German. And then they went on doing their own experiment. Now, the thing I want to draw your attention to is the, you see a very thick orange dashed line and it's called the neutrino, it's basically called the neutrino line. And it's the moment where the experiments can actually detect neutrinos. Now, in 1998, where Dama was, there was no way we could think that a dark matter detector would ever detect neutrinos. But then nowadays, we're not very far away from this limit. And you can see, so if you follow Xeno one ton line, which is a green, dark green dotted line, dashed line, you can see that it's touching the neutrino line. So the neutrino, what we call a neutrino flow. If you, so if you detect, if you just at the border, you can detect a few neutrinos. And then after when you're in the yellow region, you detect many neutrinos. The problem is you need, if you were trying to detect dark matter, you will need to remove this background, which is a neutrino background now. And it's very hard because basically, then it's an experimental problem basically, but it's also about the number of neutrinos that you can detect. And the errors on this number is actually larger than what you would expect from dark matter. So in a sense, the main problem for the field now is that if we reach, if Darwin does a good job and we continue, then we will detect neutrinos. So the dark matter detectors would become neutrino detector. And then the question is, how do you remove this background of neutrino in order to detect the dark matter? And a way to do that is maybe to use the directionality of dark matter. So the fact that there is a dark matter comes from, there is a dark matter wind from sinus, maybe there could be a way to remove a neutrino flow. So just to summarize, maybe in a less, I mean, I will clear away everything which is in pink is excluded. And so we can say we've done a good job, but then you can see the cross-section and it tells you the dark matter cannot interact, has to interact less than for a particle in between, let's say a particle of 30 GeV is 10 to minus 48 centimeter square, extremely small. So dark matter doesn't interact much with neutrinos. Maybe it interacts very strongly with itself. Maybe it interacts with neutrinos very much, but not, sorry, not very well otherwise. So here is a proper plot for Xenon, but it's tough. Basically they're really doing an amazing job and no one could expect that they would do so well basically in 20 years ago. So really it's, this is the experimental work which impressed me the most because they were doomed in a sense and they beat, I mean, they worked so well that they managed to beat a lot of background and have extremely sensitive, so sensitive that they can beat a neutrino experiment. Just wanted to make a point and I think that was truly, I will answer Julie's question. So at some stage you have to see which one of direct detection and indirect detection and energy, for example, mechanism of dark matter production in a collider or in a particle physics experiment is the most relevant. And before it was very hard because you needed to specify a fairy, then a model and then eventually make prediction. And for each you change a little parameter in your initial fairy and then everything would change. So for example, with supersymmetry you change one parameter and you had to redo everything. So that meant people busy but if you use simplified models then you can do this comparison extremely well and extremely fast without having to specify a fairy. And so you can compare the LHG with direct detection and so on. So this is basically the table I was referring to and Julie you can see now. So this is a list you have all sorts of dark matter candidates. So it can be Dirac Majorana. It could be a scalar, complex scalar, real scalar and so on, could be a vector and so on. And then for each of these possible dark matter candidates you define the mediator and once you have defined the mediator you have to specify whether you have a gamma five or not. So whether you violate parity at some stage. But then eventually you can compute the elastic scattering. So dark matter nucleon for example. And then you can, it's the same diagram basically if you have dark matter, dark matter, so dark matter nucleon, dark matter nucleon then you, this diagram basically reverted give you nuclear nucleon gives dark matter, dark matter. So you can basically translate this into the collider limit and you can see in the plot below at just at the bottom of the slide you can see the different limits which the LHG is setting on the independent the spin independent cross section for dark matter nucleon interaction versus a dark matter mass. So this is really important because this is where we didn't expect really a competition and now we can see very well which one is more powerful depending on the type of interaction you consider. The other thing I wanted to raise is that Whitten and Goodman had spin independent, spin dependent, that's it. And then we realized that the need that well actually there are more parameters for that. In fact, you see them also there is a one up to or I think it was 15 or 16. They're written on the right column. You see all sorts of operators which basically saying, well you can be sensitive, you can have an interaction and it's not necessarily just spin independent or spin dependent it can be much more subtle, it can be a combination of spin or not. And in the table you see the spin of a particle of a dark matter particle and the spin of a nucleon. So it's a very complex problem. And if there was a signal, we would have to go back to those tables and see which one is correct. Okay, so more of the same and I will skip this. The other thing I wanted to mention very briefly is this was also a question with a dark matter as a charge. And if a dark matter as a charge, it has to be very small and probably has to pass through something else which could be a dark. So either side a dark vector or a z prime. And so for example, if you have a dark photon, then in this case a dark photon could mix to the normal photon and then you can have an interaction with the solar model. So it could be a charge but it would be merely charged by the smaller and the type of experiment, there are plenty of experiments looking for this but one important type of experiment is called the beam dump experiment. So that's the way basically to detect the dark matter directly but you need basically to produce, you can't produce a dark matter. So you need to go through electrons and then electrons heat and target material. Then you're hoping to produce the new mediator. So dark photon. And then you hope to produce a dark matter this way. All right, so I forgot them to say that using this you can put a lot of limits on the dark photons and the answer is that, well, if they exist, the coupling is very, very small. You can do the same with vector formulas and now that I mean there it's really LHG but again, you can put some limits. All right. So last thing I wanted to say on direct detection is the fact that so far we're saying dark matter has to be heavy to be detected directly. And the question was how do you detect light dark matter? And some people found the way. So you can basically make the dark matter interact with electron and it's something I tried myself and I failed at the time because I thought the signal would be too small and most people actually managed to do it. So there's always personally, I was always impressed and jealous but this has started somehow a new field of detecting light dark matter. And this has I think is one of the papers which to me has really created a range of new experiments. And you can see an example with of a new type of detector. So super conductive detectors for super light dark matter. I think basically this has changed the mentality of a field and people are starting to think about even using CCDs which are usually used in astronomy on basically an experiment. I mean on the telescope. Telescope for detecting normal light. Now this is used basically to detect dark matter. So there is one experiment which is a small one which is called Sensei. And the next version which is a bit bigger and a bit more powerful is called Demich. And all of them are doing fantastically well. So it's very, very impressive. All right, so I'm sorry to rush but I just wanted to finish off with dark matter because I promised I'll go back to cosmology and I wanted to wrap up things. So dark matter on CMBN and large scale structure. So if you remember I spend a lot of time on showing you that all the information we have basically from dark matter comes from cosmology and from astronomy. And so therefore the question is can we extract more than just knowing that there is dark matter? Can we get more information than that? And when you look at the CMB plot I explained that while we measure all those peaks with amazing precision so it's incredible. But really the question is can we actually prove the nature of dark matter through this plot? Because apparently whatever the dark matter is it doesn't make a big change to the plot. And the thing I wanted to show you is that it can but as long as it has interaction. So here is a simple explanation and I'm not showing you the CMB angular power spectrum. I'm showing you another thing which is really related to CMB but this is basically at lower scale. This is called the matter power spectrum. So it's exactly the same information if you want but at lower scales. And what you see here it's called a P of k. So it's a Fourier analysis of a distribution of matter across the sky. That's just the Fourier mode. But what you see is that you have a data and you see that they oscillate. And what they oscillate is somehow the same physics as you've seen with the CMB, with the oscillation that they self-function, the same physics. And this physics is basically the fact that the baryon dissipates. It's the fact that the baryon interact with photon. And when they interact there is pressure basically from, so if you want to have gravity which tries to collapse the the pocket of matter where the baryons are. And then you have the interaction which put pressure and try to expand. So you have a fight between gravity which is trying to pull down I mean basically collapse the fluctuations of dark matter. And then you have basically the pressure between the baryon and photon which is repealing everything and trying to expand. So you see that eventually in what we call the baryonic acoustic oscillations. Those were proven not so long ago and the price was given I think in this decade, in the past decade. So what happens when you have interacting dark matter? Well if you have interacting dark matter you should see those oscillations again. Not because of a baryon photon interaction but now because of dark matter interaction with either the baryons or the photons or the neutrinos, whatever, or the self-interactions. So the typical effect of those dark matter interactions would be like the baryonic acoustic oscillation would be basically the oscillation and you see the power spectrum now which is oscillating. So you can see in the bottom panel you can see the power spectrum that you expect. What happens if you add the axioms? And I told you so this is just to remind you that we are speaking about particles but axioms behave more like a wave. There are plenty of limits on the action mass but there are still a range of parameters where they could be. And so the question is would axioms have a similar power spectrum or not? And the answer, and there are plenty of words by David March actually and so I gave one reference but I would encourage you to read this paper or I found them extremely well written. And you can see the power spectrum is also oscillating but less than with a normal particle but you also expect some oscillations. So I will skip all these but I will let you in the slide so that you can go back to this if you want. But the main thing is what is the universe looking like if you have something like an axiom or if you have something like an interacting dark matter particle. And we know dark matter interacts. The question is, well we don't know but we think dark matter interacts. The question is what is the strength of interaction? So you have two extreme case here. On the left side you have a case where dark matter doesn't interact at all. And you can see it's a rich universe there are plenty of structures. You see the filaments which I spoke before you see everywhere you have a point it's a galaxy and so on. It's very rich. The big yellow red is a cluster. It's a very rich environment. The one on the right has numerical issues but the one on the right you can see is where dark matter has too many interactions. And then you can see the universe is the big structure can still form but the universe is pretty empty at that stage. And so what is interesting here is that you can see filaments still forming. There are lots of artifacts here because we've been pushing really the limits but what is interesting is that you can see it's not such a dense universe. So the nature of dark matter whether the microscopic nature whether it interacts a lot and it's mass ready those are the properties which can change the way the universe looks like. And I'm just going to show you this to finish off. This is a simulation where you see basically the whole universe in the case where dark matter doesn't have interaction. We're zooming in. We are now in Milky Way Hello something which looks like the Milky Way Hello there are plenty of structures. So all the points you see is galaxies. You see many, many galaxies. And then we're going to switch the dark matter interactions. So the same simulation but at the very beginning we would have considered dark matter interactions. And you can see that now the Milky Way doesn't have so many satellites. So there are galaxies in the Milky Way but not so many we've compared to the case where dark matter doesn't interact. And now if you had done the same with huge dark matter interactions basically the Milky Way would be sterilized it would be totally empty and that we know is not actually the case of the Milky Way it doesn't correspond to reality. So we can put a constraint on the dark matter interaction by looking at the number of structures which form in the galaxy. So in other words all those lecture we're trying to show you the complementarity between from one hand the cosmology to the particle physics aspect. And essentially we're now entering an area where you can't just be if you're working on dark matter you can't just be a particle physicist or you can't just be a cosmologist. You have to be both because basically the nature we're able to prove the nature of dark matter by putting everything together and having this consistent story. So if you consider dark matter interaction if you're large enough to change your cosmology and depending on the cosmology you're thinking well that may change with dark matter. So you really have to consider both together nowadays. We far from putting interesting limits though but I just wanted to show you one case which is for me very interesting. This is a summary plot and this is basically you see cold dark matter on the top left warm dark matter on the right and then you go down and this is a case where dark matter interacts with a photon but it will be the same with a neutrino. So this is a bit of an extreme case. Now if it does interact with a neutrino or even a photon but that wouldn't be realistic you could put a constraint which is as good as 10 to minus 36 cm2 for a particle of 1 MeV and that is the weak strength I mean that's the strength for the weak interaction in the Stonar model. So somehow in cosmology using cosmological data that you have and we are going to have plenty very soon with LST Euclidean and so on. So we have even better limits you can already touch upon the case of a light dark matter interacting with neutrinos which is something you cannot do with a normal particle physics experiment because you can't produce a dark matter and then how do you measure a dark matter neutrino interaction that's actually very difficult. So here basically you can do this almost for free as long as you have you have access to cosmology data then you can do this exercise and I just wanted to for concluding really I just wanted to show you that as we're starting to understand better also astronomy in the shape of our own Milky Way we're starting to make measurements such but now we also start to understand better the velocity at which a dark matter is going through the Milky Way and that is helping us with direct detection so it's helping us with indirect detection with direct detection and eventually everything comes together it's not just cosmology and particle physics it's actually also now with astronomy so you have to have a multiple hat if you're working in this field this is at the moment this is a bit extreme but I think in probably next decade and maybe 20 years from now everybody would be able to pass from one field to another so this concludes basically the set of lecture and there were many many other things to say obviously but I hope I kept as many students as possible and I really hope that you enjoy them and thought that it was a fascinating subject so thank you very much maybe I can take a few questions thank you very much Selina there's a question yes please go yeah going back to the simulation you just showed where you had cold dark matter warm dark matter and and so on so from what I understood the data we have right now it's basically consistent with cold dark matter that doesn't interact as far as we know not really okay yeah yeah okay then that's already maybe you can already so yeah yeah because there is something called the missing satellite problem and there's a number of satellite so satellite is the small galaxies that you see now and they're very difficult to find because they don't have many stars in them so this is something I mentioned yesterday and so it could be that there are plenty but we don't we don't see them because they don't shine enough however when the one we know and we're discovering more and more but the one we know is a tiny number some such it was 15 maybe now we reach 100 but it's very tiny number while cold dark matter predicts so collisionless dark matter actually predicts billions so we're very far from finding them yet so it's in between warm dark matter and cold dark matter you wanted to yeah yeah the next part of my question was could this be the interplay of having warm dark matter and you know a bit of like collisionless but warm so you got okay you could think about that I don't think this one in particular has been studied but we were looking at a mixed of things for example and that's that's also interesting the effect are less pronounced so that might be better and yeah that's a possibility so you indeed you could think that it's a light particle which may be warm enough but then there's some interactions and there could be a combination of two yeah that's possible that's another question by Thong thank you professor and also thank you Selin for one of the picture can you go back to the slide of the non-relativistic operator this one yeah yes this one so I want to ask that if we want to calculate the scattering between dark matter with neutron or proton then we build all of these all of the non-relativistic operator when we consider it's been spin-independent or spin-dependent then should we multiply with the form factor from going from the quadruple level to the nuclear level right? yeah yeah so this is just to say the type of interaction but then the calculation I didn't I don't think I express the interaction right but the interaction right indeed you have to sum of all the nucleons that you have and then you have the form factors and so on and and if and some kind of non-relativistic reduction operator will vanish because you know like yeah that's right so some of them depend on the momentum and some of them will vanish indeed some of them are very surprised and you can see it almost if you take the one in the middle you can see there is the longest expression for example involved two square or ten plus okay that's maybe not a good example but you can see the first two terms depends on the momentum and the momentum is very small so that would disappear so because the momentum is very small so if we multiply it with the form factor F1 and F2 just like Qcd then we just consider the F1 the F2 we don't care because well you can put them but in principle when you integrate eventually you would pick up depending on the momentum that you initially generate it may disappear so I think I forgot and there was a case where there is only one of his operators surviving I think it was a 15 I forgot there are tricks and you can play a game you can invalidate every single previous search because there is only one operator which would work and it's the one nobody thought about I've forgotten exactly but I think it was a one with a 15 I've looked at this a long time ago but it may be actually the one I took in the middle the longest time where you can kill Q square but then you still have your 15 operator okay and my second question like you said if we consider the dark matter the light dark matter it will interact with the electron the system of the the lattice and it will make some collection can it make like magnum or a system just like yeah so basically in Xenon basically generate a signature because you so in Xenon you've seen that the electrons are produced and drifting so if you have an interaction of dark matter with the electron then in principle you change the signature and you can detect it so that was actually the revolution is they showed that you could detect it you could have this interaction in Xenon if I'm not mistaken the first point was with Xenon but since then people have generalised it and then people are proposing new techniques as I said so for example having the CCD so it's a technique which could work to detect like dark matter but then it's more of a light coming from I mean it's really detecting the light from the interaction okay okay so I guess our time is over and there are no more questions so let me join the many messages in the chat that warmly thank you for the very nice lectures so thank you once again Selina thank you very much many thanks Ben I hope you enjoyed