Week 9: Lecture 42: Accelerator based neutrino experiments

In this lecture we will discuss uh the oscillation experiments but which are accelerator based. So I mean there perhaps other ways of grouping it but uh the way I have grouped it is there there are short baseline experiments in which you measure the nutrinos not too far away from the source uh and then there are so-called long baseline experiments. So these short baseline experiments are typically of the order of tens of meters or a few hundred meters and so on. Whereas the long baseline experiments could be anything between about a few hundred kilometers to uh 800 km or so. So the accelerator based nutrino experiments uh of course based on this uh fact that when a proton high energy proton interacts with the nucleus it produces pions uh slightly larger number of pi pluses as compared to pi minuses. uh and then these pions decay to muons and the muon neutrino that's a two-body decay later of course the muon also decays into electron or a posetron as the case may be and a muon type of neutrino and electron type of neutrino. So if you are able to look at the uh neutrinos when the pion is still in flight then there is a kinematic boost and so they are sort of focused in the forward direction and then after a certain path uh you could stop these pions so that then the uh the resulting decays either from the stopped pan or from the muons they are basically isotropic. Okay. So, uh the neutrino energy from p and dk in flight is given by this expression is the neutrino energy in the center of mass uh into m pi uh divided by gamma pi 1 - beta pi cos theta where e newino 0 is given by just mi^ 2 - mu^ 2 by mi whereas gamma pi is e pi by m pi So let me see I think there is a mistake here. I think this is probably gamma it should be boosted. So if gamma is large then this tells us that the so I think this is uh this needs to come here I think. Uh so uh anyway this will be I'll check that out and correct the slide which will be stored. So any case this uh so the neutrino energy coming out from a pion decay in flight is boosted as compared to its energy in the frame of reference of the pion. uh so as I already said the stopped uh inflight pion decay gives new mu only whereas the mu decay gives new mu as well as new e very high energy muons uh of course can also decay in flight and they can give rise to electron and muon nutrinos and this is of course the basis for the neutrino factory that was proposed uh maybe about 15 to 20 years ago. uh however the need for that factory uh the main need went away because theta 13 was measured in the reactor experiment and this is one of the things but of course there are other uses of the nutrino factory uh and so on. So I I just mention it because the muon can also decay in flight. Okay. So uh why do you need accelerator produced nutrinos? So accelerator produced nutrinos have the uh good thing that good characteristic that we know the composition of these nutrinos. You can be for instance inflight pond decay. So mainly it is new mu with a small new contamination. Uh it has a known energy spectrum and you can also look at off-axis pond decay. So uh you know the energy uh spectrum will get narrower and this arises of course because of the kinematics. For instance, if the uh neutrino decays in the same direction as the pion or in the opposite direction the energies are quite different and therefore you have a broader spectrum on on axis uh nutrinos as compared to the pan direction. uh whereas if you looked offaxis the number of uh nutrinos would decrease but then because you're let's say looking at 90° in the frame of reference of the pion with respect to its direction then these nutrinos get focused at a certain energy but then of course the number of nutrinos uh becomes smaller. Uh, of course, when you also have a new factory, then you know the energy spectrum because uh the the boost is much larger uh than the spread in the frame of reference of the muon. Of course we we people are still not talking about pion uh as a neuon nutrino factory uh because the pion decay is much faster than a muon and therefore people have only talked about the neutrino factories based on uh accelerating muons and storing them. [snorts] Uh okay so the other advantage of accelerator produced neutrinos is that you have a known time structure. There are narrow pulses and they have a micro structure and a macro structure and this can be used for background reduction. So you look for events of interest only during those narrow pulses uh and the rest of it is uh your uh time when you can measure the background. So the first uh generation experiments are these ones which use so-called neutrino super beams. So you have a driver proton beam of the order of a megawatt or so and then you have a pion beam and the pion decays in flight and you can choose pi plus or pi minus you can get either uh new mus or new mu bars uh using devices called horns magnetic horns and uh so these are the first generation experiments and as of now these are the experiments that have been uh that are ongoing or are being carried out. [snorts] The second generation experiments will of course use nutrino factories uh where you have a stored beam and a large flux in a particular direction. Okay. So these are typical layouts of nutrino beams from accelerators. So this is the sun uh CGS so-called beam line which sends nutrinos from sun in Geneva Switzerland to uh the grand saso laboratory. So the opera experiment for instance was carried out there. This starts out with a higher energy beam of 400 GV protons hitting a target and then there is a so-called horn which is basically a pulsed magnet which can focus pi plus and defocus pi minus or vice versa and then uh these pans decay in flight to produce the new mu beams uh that ultimately reach your target. So this distance is uh of the order of about I think 700 odd uh kilometers. Uh there is the so-called newumi beam line. This is in firmy lab. Uh uh this also starts out with a high energy proton beam. In fact in this case it is uh 120 GB beam. to the uh Minos experiment was done with about 300 kowatt beam power on the target. So again you have a magnetic horn and then there is a decay pipe where the pions decay in flight. Then there's a near detector and then there is a far detector 700 odd kilm away 7 to 800 km. So there are two experiments uh the Minos experiment at some distance of 700 odd kilometers and nova experiment which is of the order of 800 odd kilm. Then there is the shorter long baseline experiment. Uh this this shows the K2K but now of course there is the so-called T2K beam line. So this is the KK2 Kamocande beam line. So you have much lower energy proton sitting at target. uh and then of course a horn and then a decay pipe of about 250 mters. uh and uh you can see that you know because of the higher energy average energy of the ions this decay path is like about 675 m and uh let's see in this case if you can read it it's of the order of about uh thousand mters or so and uh any case the K2K then there are near detectors and then there is a far detector so this is the general approach that you have a near detector uh close to the source within a kilometer or a couple of kilometers and then you have a fire detector sitting at uh in this case of of the order of about 250 or 300 km and uh in these cases of the order of 7 to 800 km. uh the near detector. Uh the main idea is that then you sum up when you uh when you make this measurement here the spectral information to first order goes away when you are looking at the far detector because you have already measured this with very high statistics in the near detector. Of course there are issues about the angular dependence and so on but those are higher order things which uh which of course contribute to error but to a higher order second order rather than to the first order itself. Okay. So the short baseline experiment has come to where the distances are small and the first one which actually showed a very surprising result was done at the uh Loselamos ion factory which is called the lamp facility. So it basically had 800 me protons of the order of a milliampere in current hitting a target and then uh you had pions and then they came to rest and then the muon also decayed when it came to rest and so this experiment was actually looking for anti- muon type of neutrino which is coming from the pion decay at rest u and going over to a anti- new E uh this is searched by its charge current interaction with protons in a 167 ton liquid cintilator consisting of mineral mineral oil then some cintilator material that is uh put in there dissolved and then you looked for the interaction with a proton to measure these uh newi bar that were produced in this kind of oscillation. uh they found the surprising result that uh they actually could make a plot of the allowed values of these uh this mixing uh with sin^ square 2 theta versus delta m^ 2 and it turned out that the mixing parameters were of the order of a fraction of a eV squar okay and then this sin^ square theta of course there's a two-dimensional plot so depending on what delta m squ you chose you got the corresponding sin square theta but these sin square thetas were not very small in this region so if you took let's say e^ squ then this is of the order of 10 minus 2 so it's pretty uh large mixing now so there was a uh about a four sigma evidence for new mu bar to new e bar oscillation and uh uh so this is what is shown in the plot here L by E new in meters perme is plotted and uh if you want to fit this then you have to put in that the kind of mixing that was indicated here and this is again uh what is shown here but this this uh you know region which was uh what was favored by LSND was not compatible with atmospheric and solar neutrino measurements which are shown by these dots here. they are fairly precise measurements. New mu to new x and new e to new x. Uh so they they don't quite uh add up. So anyway, so another experiment was of course necessary and this was the miniboon experiment which is also a a cherenov based uh measurement uh based on 800 tons of mineral oil. It is not a cintilator but is a mineral oil based chenov detector and they actually confirmed uh the LSND result for anti-utrinos but in the nutrino sector uh this is not compatible with LSND okay [snorts] so they in fact had a exclusion plot so things which are uh to this this side are excluded and uh that was shown this is the mini boon the thick line shows was this. So it uh it uh it basically ruled out the area here. But it is of course possible to have uh you know parameters uh which are outside of this. So this dark line seems to have excluded uh in the neutrino sector but was compatible with the anti-utrino uh sector. So again this is another uh uh you know contradictory thing that they saw. So there is a third experiment carried out called the microboom detector and this was based on a very different uh concept of detector. This was based on liquid argon. As I have already said earlier in in some earlier lecture that the liquid argon you actually it's like a realtime uh cloud chamber but with 100 micron accuracy. So you can actually get the tracks with about 100 micron precision uh and uh you know you can measure all these tracks. It's a very precise instrument. Uh so they they had built a 170 ton liquid argon PPC detector and they used the NUMI beam at Fermy lab the new MU beam and uh they actually the the summary of that or the result which they got was that uh the uh microboom actually rules out the area that is favored by the LSND and mini boon. Okay. So uh the the area to the right of this uh red line here is ruled out by this experiment and uh similarly so this is the microbon result. Uh and similarly it it also rules out some things which are uh favored by I think uh the mini boboon. Okay. So microbon actually ruled out the parameter space that was favored by LSND and many boon experiments. Of course this is also in conflict with the gallium results which are the ones which are shown here the galax sage and BST. Uh this is this purple area and so that is also in conflict with the microbound results. So we have a of course a problem between the radiochemical experiments and the realtime experiments such as microboom based on accelerators nutrinos. Okay. Then there are the long baseline experiments where you produce nutrinos at some place and then look at them uh far away of the order of 7 to 800 km. So you produce them at firmy lab for instance. This was the Minos experiment. The main injector nutrino oscillation search minos. This was uh started data taking in 2005 ended in 2012. So it was a 7-year run. Uh and the this is a schematic of the detector. So you have a target then again a DK pipe big shield and then a near detector uh which is near detector is this. It's a 1 kiloton detector where size 3.8x 4.8x by 15 m thick. So this is the thickness and this is the width and the height. There are 282 steel plates with uh interleved with uh cintilator planes and then there is a tooidal magnetic field which you get by actually at the center of this there's a hole and you pass a very high current through it such that you get an average field of about 1.2 Tesla uh a toidal field. So the field goes like this and uh this is the near detector and 5 and a half times or so the larger detector is at the far end which is a 5.4 kilot 8 m by 8 m by 30 m thick and uh these steel plates of course are about an inch thick and there are 484 of them interled with cintilator planes and uh there's again a toidal field of about 1.2. So as you can see these nutrinos travel all the way here and you are looking at uh a neutrino uh a muon type of neutrino in these detector. So you are actually measuring the survival probability. Of course they also measured the uh the uh new ease that came about because of the oscillation. So we'll come to that in the next slide. Okay. So this is the so-called this is a simulation of course this is the unosculated spectrum and if uh by that time of course super kamio and other experiments had given some idea of what the parameters were so if it oscillates then of course the survival probability of these muon type of neutrinos comes down and this is the spectrum that you get as a function of the visible energy in gev. So you can see the spectrum goes from about uh roughly about a GV of that order to about uh 10 GV. Uh so the average energy is of the order of few GV 4 GV or so. Okay. So what are you looking for? You're looking for the ratio of oscillated to unosculated. Of course the unosculated is from some kind of theory or alternatively from the spectrum that you measure at uh the small distance in the near detector. So this is the kind of uh this reduction is uh proportional to sin square 2 theta whereas the place where you get the first uh minimum that gives you information about delta m². So the minos near detector here has much larger number of events. So for 10 to the 16 protons on target you get a peak of about 6 into six uh events uh for a reconstructed energy of uh about 3 GV or so. Okay, this is I think there is a events per GEV. I think there is a unit missing here. I hope I I I'll try to correct that. But in any case in arbitrary units this is the kind of spectrum that you get uh as a function of the reconstructed antiutrino energy uh and uh the fire detector measures the reconstructed antiutrino energy here uh as and that is what is plotted is events per gev and the reconstructed neutral energy. So you can get a delta m squar or a modulus of that for antiutrinos or nutrinos from this plot as a function of sin^ square 2 theta. So the theta bar refers to the antiutrino and the uh there should have been a delta ah this there is a delta m² if I if you look closely at it and so uh that is for the antiutrino and this is for the nutrino. So you can see that uh these uh areas are the one which are favored by the experiment. this dark area here and with uh of course with uh for different scenarios or for different uh you know parts of the uh data 2009 to 2011 for new muubar uh new bar 2009 to 2010 and anyway this is the final one the dark one 2005 to 2010 not the final one sorry this is still a 5year data set they actually took data up to 2012 12 uh or 2011 I think. So anyway so this is the so you you can see that the uh error on the uh sin squared uh the 2 theta that is of the order of about plus - 5%. And on the delta m squ for at least for the nutrino is somewhat smaller uh that is of the order of about uh you know about 4 and 2.5 or so. So that's about 15% or so. Okay. Now there is another long baseline experiment with nutrinos starting from firm lab. This is the so-called off-axis uh newi appearance experiment. So that was the survival experiment. This is an appearance experiment. So you you start out with a new mu beam and then you look for new ease. Okay. the different flavor of nutrinos. And this is the near detector at about a kilometer. There's a 220 tons or so liquid cintilator plus uh wavelength shifter fibers through the liquid cintilator. And then uh at the end of those fibers, you put so-alled avalanche photo diodes to measure these cintillations. The fire detector is about 810 km. It's 14 kiloton, but it is on the surface. This is the first uh I mean detector that was put on the surface. Uh I mean at such a large distance 810 km they chose to put it on the surface and use parts of the detector as cosmic ray vetos. But the uh their main plus point was that the beam was pulsed and so you could use the nancond structure substructure of the beam to time it uh such that you reduce the cosmic ray background by several orders of magnitude. uh so in any case so the off-axis part leads to a narrow energy spectrum which is peaked at around 2 GV whereas if you took the whole spectrum of nutrinos then this is peaked at a higher energy of the order of 7 MV or so but then you can see that that is broad and as I said this is broad because uh all angles in the frame of the pion uh they contribute to this uh and especially the forward and backward uh moving nutrinos, they contribute largely to this width of the neutrino spectrum. But if you go off off axis, then you're looking at only those uh decaying at about 90° in the center of mass frame and uh that's how you get a narrow but of course you're looking at only a small part of the spectrum. So the total neutrino flux is much smaller but you have a very good energy definition. So the NOVA preliminary uh data u is shown here. I somehow missed out on the reference here but the I think this was sometime in 2010 or so. The year is not given here. Anyway the reconstructed electron neutrino energy is shown here for uh the fire detector and this is here. So there are a handful of events about I think 150 200 events uh and similarly uh for the nutrino beam and for the anti-utrino beam for the anti-utino beam you have lesser events partly because the flux is smaller and partly because the interaction cross-section of course of these antiutinos smaller about a factor of two or so. So from this experiment uh this is just one uh so this started in 2014 okay so this is from some reference in 2024 I think it's from some uh uh review article or uh a uh conference talk as you can see that this is what is plotted here is delta m squ versus again sin^ square I don't there should be a sin square 2 theta I think in any case this is broader than the delta m^ squ and so this is the limits from this experiment are rather narrow. So it measures this delta m32 fairly accurately but also it measures the other mixing angle uh sin^ square 2 theta 2 3. Okay. So uh let me go back. I think I should mark this here. The I should put that in the final PowerPoint slide. Okay. And the reference. Okay. So the uh T2K experiment is uh in Japan. Uh you produce these neutrinos at uh the 12GV accelerator facility in J Park. as that's that's at Tokcoy Tokai on the you know coast on the eastern coast of Japan slightly north northern parts above Tokyo and then these nutrinos travel about uh 300 kilometers uh to uh the uh super kameoc detector. So you also have again uh near detectors and this is a onaxis near detector here called ingrid. Uh there is also an off-axis detector uh here and they are fairly close to the uh you know source at about 300 m or so. Uh again the idea of the off-axis experiment is that you have a narrower energy spectrum of the nutrinos that uh basically are detected in this far detector. And so the idea of this experiment is also similar to the ones that I talked about. So from this experiment again you can get delta m² 32 for the no oscillation to uh you know oscillation. Uh so sorry this is nor sorry this is not no oscillation this is normal ordering uh or uh inverted ordering. Okay, so these this change slightly and so this is the uh the blue uh exclusion plot is for the normal ordering and the orange one or brown one as you might see is the for the uh inverted ordering and these are of course various fits at the 68% confidence level which is a one sigma or 90% or 99.7% confidence limit. So this is the outermost one which is of course larger. Uh so the again uh I might I might be repeating it but the plot here shows that area which is allowed by the experiment and all other uh things outside of this uh you know closed uh curve is not allowed. uh but another interesting uh I mean uh analysis of this data was carried out uh to look for CP violation indeed at a very low confidence level they have some evidence that CP is violated which means the neutrinos and anti- neon neutrinos behave differently and so this is uh depicted in this plot so what is allowed is uh this region so this is uh uh at the one sigma level and this is at a higher confidence level and so you know again the delta CP extracted depends on whether it's normal ordering or whether it is inverted ordering this is the reference here and so you get some uh you know parts of this uh two-dimensional plot which are allowed and this is what is excluded this blue green region is what is excluded by this experiment of course at the confidence level that they have here. Okay. So this is the 68% this is the one sigma confidence level and this is the 99.7 uh confidence level. So if you go by the one sigma then you have excluded quite a large part of this parameter space. So the conclusion of course from this uh experiment is that uh uh we have a measure of the uh you know the the standard parameters uh delta m² and sin squar theta. Uh I should mention here this is I think of the order of about 200 kilowatt uh or so 200 and at various phases of the experiment uh I think they went through about 200 kilowatt to 700 kilowatt beam on target power. So have we actually seen CP violation? I mean if you if you ask for the gold standard in particle physics that's a five sigma uh confidence level. So we haven't probably seen CP violation but this is an indication which uh other experiments need to confirm or uh uh deny this kind of conclusion and so of course there are other experiments which are trying to do that. Okay, I should at the end also measure an or talk about an experiment called opera where the nutrinos were sent from CERN to Gransaso again about 700 kilometers away. Uh and the detector used was very different because they were actually trying to look for new mu to new toao oscillation and uh how do you detect the new toao? through the charge current interaction where you produce a tow particle. Uh you produce this toao in a target material and then you try to detect it in your detector. Since the tow lives for a very short time uh I think it's of the order of few hundred microns. Uh so C tow I should have mentioned it here. So C tow for the uh tow electron that is produced as a result of the uh charge current interaction. It's pretty small. So normal detectors will not be able to measure uh you won't be able to tag it. Whereas uh they had a detector which consisted of uh lead. So it had a lead target and the identification was done within about 100 microns. So CTO uh is in this was in this case of the order of a 100 microns by the end of so this is the uh picture of the detector. So you have lead brakes and you have plastic uh you know track detectors followed by imulsion track detectors. Emulsions have a better resolution. So you can see typical events in this. So you you come in from somewhere here you produce a toao and then that decays in various hedrons and leptons and then you measure the muons in this muon counter and uh I think this is a muon counter or this is a there of course this consists of a sandwich of uh plastic track detectors and emulsion detectors and then backed by muon counters and these muon counters are actually resistive plate chambers. So they are operating in real time whereas these have to be actually the imulsion or the track detector has to be taken out and then correlated with the event uh occurring in a certain time uh window uh and then backtracked that okay this is the this is a muon this is a muon and so on. So anyway uh this process also was automated. So the scanning of the uh track detector and the imulsion detector was automated. So uh they removed it after every whatever some fixed amount of time that it takes to actually measure these things and then put it back uh into the you know put a fresh set of emulsions and track detectors into the uh opera detector. By the end of the run which was for four years between 2008 and 2012 they had actually measured 10 events and this is the paper this is the reference for that. So they actually were definitively measured new mu to new toao oscillation. Uh and that was possible of course because they have a high energy neutrino beam and so they could actually uh look at those new towels which are also high energy which can then produce the tow electron. So in summary we have discussed some of the short baseline experiments such as the LSND uh and microbon. We've also discussed some of the major long baseline experiments like Minos, Nova, Opera and the T2K experiment. Okay, thank you. [music] >> [music] [music]