Searching for Dark Matter, the LUX and LZ Experiments - Dan Akerib

hello welcome to the SETI colloquium series I am a TA joke and have the honor to introduce than a crib who comes to give us a talk Hicks coming from slack as Sanford before being at Stanford dan was undergrad at University of Chicago then he got his PhD at Princeton he was a postdoc at Caltech and Berkeley and he spent a long time at Kate Case Western but we have better weather so he came back to the Bay Area and he is working on the experimental detection of dark matter which is great a great subject and he's gonna tell us all about it and thanks to you thank you so for those of you who are unfamiliar with Case Western it's in Cleveland Ohio current home of LeBron James I have to apologize when I'm going with the home team for for the NBA Finals so but once that's over I I look forward to Steph Curry being my guy too so anyway yeah right right right one apiece right now so anyway it's exciting and dark matter is also exciting and I hope to hope to convey my excitement to you for that subject as well and so I'll be talking to you about wimps weakly interacting massive particles which are one of the favored candidates for a subatomic particle that might have been produced in the Big Bang and make up the dark matter and I'm part of two scientific collaborations the lux experiment which stands for the large underground xenon experiment and its successor experiment which we're currently planning i'm prototyping and slack called LZ it's a merger of two collaborations the lux collaboration and the Zeppelin collaboration and we're searching for this cosmic dark matter there's sort of three three themes and if you will to the talk one is that you know we in searching for dark matter we posit that we that we understand gravity through you know the work of Newton Einstein but there's a missing piece there's missing stuff this dark matter and if we can't find it it's hard to really say that we understand gravity all right they're altered there's alternative work going on rather than looking for dark matter looking to see if the laws of gravity themselves can be modified to account for you know missing force rather than missing stuff right so it's a big mystery and you know we would like to solve it another aspect of the mystery which I'll tell you about is that if Newton and Einstein got gravity correct then the stuff that makes up this missing mass can't be made out of ordinary stuff it can't be made out of protons neutrons material from the periodic table or particle physicists call it baryons proton as a baryon assist a fancy name for you know sort of subatomic particle physics speak for what ordinary matter is made of we know that there can't be enough ordinary matter in the universe to make up the dark matter so if we see dark matter and the form of wimps that we're looking for it will also tell us something new about the fundamental interactions and the third thing is you know can we find it so I'm going to talk you through how we've built up this hypothesis of dark matter the wimp hypothesis in particular and the detector technology that we've built to go and look for it and I'll tell you about the lux experiment which already has scientific results we're currently the best in the world at not seeing dark matter and that's important and that we're following it up with another experiment called LZ which will you know increase the sensitivity by by a factor of 100 okay so that's kind of the overview there's competition out there too but okay if the next light comes up that would be great searching for the next slide yeah there we go okay okay so we'd start with the story of gravity and we understand you know how gravity works in the solar system the Sun makes up 99% of the mass in the solar system and the planet's orbit around it and they follow a nice Keplerian law where the gravitational pull of the Sun keeps the planets moving in their central orbit so if you measure the speed of the planets the orbital speed as a function of distance from the Sun you get this you know nice curve predicted by Newtonian gravity and everything works fine now if you apply the same principles to what's going on in a typical galaxy measuring the rotation speed of stars and other tracers of the gravitational strength as a function of the radius and you make a similar type of graph now the rotation speed of the various test particles as a function of the distance from the center Newtonian gravity would tell you once once you're beyond where the Bulge and the disk are most of the visible matter then you should also fall off you know with a 1 over R squared force law and the speed of the object should go down as you get further and further away just like we saw in the solar system but what's observed in virtually every galaxy that studied is that out to very large radii the speed of the objects is moving very fast so much faster than can be accounted for by Newtonian gravity and these these graphs go by the name of flat rotation curves because they don't they don't fall off but they stay flat out too out to a very large radii so that's that's evidence for some missing mass or a misunderstood force law on the on the scale of galaxies and then we can go to even larger scales clusters of galaxies where now each of the yellow blobs in this in this photograph a Hubble photograph are each of each of those things is a galaxy and they're gravitationally bound together by their mutual gravitational forces and one of the ways of studying the amount of mass in those systems is to is to take advantage of gravitational lensing so occasionally you find a cluster of galaxies which is just along the line of sight with a very bright object in the background and because of gravitational lensing the the space being bent around the around the cluster the background object can be split into multiple images and that's these little blue blobs that you see here the the mass of this cluster is splitting the background object into into multiple images so by looking at the amount of that lensing one can work backwards to find the strength of the lens and compare how much mass is needed to do that lensing with the luminous mass that that you see directly in an optical telescope and a mathematical fit to the strength of that lens says that the mass profile can't be explained by the spikes due to the luminous mass but rather that there's some smooth hump of additional mass that's required to to cause all that bending of space-time now this was first noticed in the 1930s by the astronomer Fritz Zwicky was you know as you can see by that photo a bit of a character he said that you know how much of the universe is not understood and he called a dark matter he didn't use the technique of gravitational lensing that that came that came later but he studied the relative motion of how fast the different galaxies in that cluster we're moving around and applying applying a principle called the virial theorem and that showed that things were just whizzing around much too fast that could be explained by the amount of gravity associated with the luminous stuff okay so we have dark matter on the scale of galaxies we have dark matter on the scale of clusters and then on the scale of the full cosmos there's there's a set of measurements that have been made that give us an inventory of the universe on cosmological scales now each of these each of these measurements is a colloquium in its own right so I'm not going to try to do that justice but just to refer to the fact that our inventory of the universe is made up of two components the energy density of the universe and the matter density of the universe the energy density is the thing that's causing this expansion to excell accelerate it's dominated by this so-called dark energy and then the matter density is is the piece that we're interested in so we learned something about the difference between the energy and the matter from studying distant supernovae we learned something about the overall geometry the sum of these two quantities by studying the sum and that gives us this band here in orange and then we learn something about the matter density by studying the overall distribution of matter and it didn't have to be that these three measurements would all sort of intersect at a single unique point right this tiny little error ellipse from the combination of these three measurements tells us that the energy density in the universe in cosmological units which I'll explain in a moment is about 0.7 and the matter density in the universe is about 0.3 and they add up to 1 in units of what we call the critical density so if you had just a density that would only had matter in it you could ask and you know we have the Hubble expansion taking place you could ask what would the matter density have to be if you waited infinite time for that expansion to coast to a stop and the value we get for that is called the critical density and it's kind of the natural units of density for for cosmology so in those in those cosmological units we've got 0.7 energy density 0.3 matter density now we know from other studies that the total amount of ordinary matter that could be in the universe in cosmological units is shown by this red line here it's it's about only about 5% of in in units of critical density so now we have this huge huge mismatch observations tell us that the overall matter density is about 0.25 in these units and observations of ordinary matter say the most you can have is is 5% so there's a there's like a factor of 5 mismatch between the total matter density and the matter density and ordinary stuff so the dark matter problem is a problem about finding the missing stuff to account for all the gravitational forces also not being able to go to the periodic table and say oh it's it's made of ordinary stuff it's got to be some new stuff so it's kind of a two-fold mystery okay so the the windpipe ASSA hypothesis the idea that weakly interacting massive particles might have been created in the early universe and make up the dark matter is is the hypothesis that we're looking at basic idea is that in the early universe everything was very hot particles were colliding at very high energies just like the artificial collisions that we make in accelerators today and because we learned from Einstein that a equals MC squared you can have two particles collide together at very high energy and make more massive particles by converting that kinetic energy into mass so one example is that a quark and an antiquark at these high speeds in the early universe could collide and make a wimp and an anti wimp and you know maybe those wimps are still around today and we can find them so we get a clue from particle physics as to what might be possible in the early universe now the we're kind of in a sense imposing you know the answer by making this wimp hypothesis we're saying the property of the wimp has to be that it has mass so that it's a source of gravity it also has to be weakly interacting otherwise we would have seen it already and that sounds a little bit circular but we're in the business of proposing new particles to solve open questions as long as they haven't already been ruled out by observation in the case of wimps there's also clues from particle physics studies of the standard model of particle physics that say you know there are questions in you know the various Maskell the quarks and other subatomic particles that we don't understand yet we've proposed new models one is called supersymmetry hasn't been found yet like dark matter people have been searching for it for decades but it has just the right particle properties to suggest that the same outstanding questions in particle physics can also address the dark matter problem so it's one thing to say these particles could have been produced in the Big Bang but this reaction can also can also go the other way oh just as a quark and an antiquark can find each other and annihilate and form a pair of wimps the opposite reaction can take place wimpin and anti-women could find each other and annihilate away and we wouldn't have anything left over and the the in particle physics to sort of quantify those reactions we use the idea of cross-section so let me just explain that a little bit imagine that you were playing a billiards game blindfolded tables about 1 by 2 meters the balls are about 3 inches on his side if you were blindfolded and you somebody set you up to hit the cue ball there's a reasonable chance that you could get a collision to take place because the cross-sectional area of the balls and the number density if you will is the right physical scale for a collision to take place right so we associate the the probability that two balls hit each other with their physical cross-section now if instead you were playing billiards on on a table the size of a football field but the balls were the same size now on average the particles would be much further apart and the chance of a random collision would be much lower or if you were playing on a normal sized billiard tables but now instead of three inch balls they were a millimeter in diameter the chances of a random shot causing a collision would be much lower okay so the whole question as to whether dark matter these wimps would have survived the early universe that that they would not have annihilated away depends on their cross sectional area if they have very small cross sections then they might still be around so what we can do is and say well we're looking for from gravity studies we're looking for this amount of total mass in the form of wimps how small would and how massive would the cross how small the cross section of the wimps have to be in the early universe and how large would their mass have to be so that they survived in sufficient numbers they didn't denial ate away to be able to make up the dark matter okay and so you can do a detailed calculation of the number density of wimps as a function of time as the universe was cooling and at very early times the wimps would have a high number density as the universe cools the typical kinetic energy and a collision starts to fall and we have a so-called Boltzmann suppression factor the number of wimps that you could make starts to fall because it takes more energy sorry it takes more energy than a typical thermal collision and eventually if wimps have a small enough cross-section then they stop finding each other to annihilate away and they freeze out at some some specific density that doesn't fall any further except for the expansion of the universe and we call that the freeze-out and by going through this calculation you can work backwards to say if I need about 0.25 units of critical density then what mass range and what cross-section would the wimp have to be to solve this Dark Matter problem and we end up with masses on the 10 to 1,000 GeV scale so proton weighs about 1 GeV so these particles would be you know 10 to a thousand proton masses and they would they would have cross sections that are typical of the so called electroweak scale in particle physics that's the weak you know the weak interaction so as I mentioned before one of the one of the proposals in particle physics is a new theory called supersymmetry it's actually an old theory now but it's new in the sense that it hasn't been discovered yet and that's one of the one of the things that's being searched for at the at the Large Hadron Collider in Geneva they sometimes talk about trying to make dark matter in the laboratory by smashing protons together at very high speed they're trying to reproduce the same sort of collisions that would have taken place in the early universe and see if they can make wimps or cousins of wimps in the laboratory and that would be great for us because then we would know we weren't just chasing our tails maybe there was actually something out there to detect okay so now we take this idea of the of the wimp freeze out to get some inference as to the cross-section scale of the of the wimps and their mass and then and then we construct a wimp of scattering experiment so now we say well let's let's go back to the rotation curve of the galaxy and let's say all of that missing stuff was made up of weakly interacting massive particles with the cross-sections inferred from this freeze-out argument okay so then the Milky Way is embedded in some halo of dark matter the wimps are orbiting around the galactic center just like the solar system is so we know roughly their speed they're moving at about a thousandth the speed of light we don't know their mass but if we if we pick a mass then and we say well we're gonna infer the matter density then we can pick a number density so now we start to have knowledge about the speed of the wimps the flux as a function of mass and we can start to put together an experiment now one problem is that the wimps are so weakly interacting that and they're similar in that sense two neutrinos a neutrino or a wimp could pass through a Lightyear of lead and maybe scatter once right so this is how you know this is how much they don't interact with with ordinary matter now on the plus side there are many many wimps if this hypothesis is correct that are traveling through us okay so we could say that you know a wimp needs needs a brick of lead a light year and lengths to travel through or we could make a detector that was ten to the minus sixteen light-years across that's about a meter so we can build detectors that are about a meter across there it turns out that given these numbers there are on the order of 10 to the 16 wimps per year passing through the detector so we're starting to be in the regime where if all this hangs together and we can build such a detector we could start to see you know maybe a few wimp interactions each year okay and that's that's the scale that we're at now just to put this in a little bit of historical context one of the first women experiment search experiments that ran was in a in a germanium detector in 1988 it was designed for a different type of study called double beta decay which I won't go into but it's it's it's very similar technology has required a detector with very good energy resolution very low radioactivity and so forth so this detector was converted over to look for energy depositions from a whimp scattering from a germanium nucleus in a solid-state detector and after something like a 100 day run of the experiment a plot of the energy deposition seen in the detector were were dominated by various residual radioactive species that couldn't be eliminated so it was basically the detector and the components of the detector were you know were the cause of the eventual background to it right so if if wimps existed at a particular scale then they should have shown up in this energy spectrum you would have seen a falling exponential if wimps were at a smaller cross-section than they might have been buried in the background so what you can say from this experiment of a particular duration of a particular mass is that any any with parameters that would have led say to this this recoil spectrum are ruled out and anything below the background we can't say anything about it yet we have to build a larger detector one with lower background okay now the other thing I mentioned is that we don't know the mass of the wimp we know roughly what the range might be so we have to interpret the data for all possible wimp masses so if we had a wimp that was was on the lighter scale it wouldn't deposit as much energy if we had a wimp on the heavier scale it would have a lower flux but a harder spectrum it would it would come out to higher energies so each of the spectra that I've sketched here are just on the edge of being detected so what we can say is that any any any light wimp so now this is whit mass versus basically the probability of interaction any wimp mass that we can rule out lead to a dot in this in this graph so by by considering continuously all of the different wimp spectra that could be ruled out by this detector we end up with with a curve a so-called upper limit on the WIMP cross section so wimp raid or wimp cross section on the vertical axis wimp mass or ignorance over here and what we're saying is that this experiment could rule out all all wimp possibilities with mass and cross section above this curve and below this curve we haven't run a sensitive enough experiment to say anything yet okay so we'll come back to that curve later how do we make so now we want to switch gears a little bit and talk about the the what goes into building a wimp experiment the the basic idea is that you you have to take advantage of the the way that particles deposit energy in the target of your detector right the three main modes where a particle will collide in your detector and deposit energy are two well eventually everything ends up as heat if you have a particular type of material that scintillates then you can get about percent of the energy appears in in the form of light and then if the if the collisions are energetic enough they can also ionize the medium and you can try to collect free electrons ionization signal so there's a very active field word worldwide that is trying to exploit different aspects of energy depositions there are experiments that run just with ionization like this germanium experiment that I showed you from 1988 and there are sort of modern versions of that which are continuing to make progress there are experiments that just look for the heat energy that's deposited experiments that just use the scintillation light and then there are classes of experiments like Lux which try to measure two quantities simultaneously so in the liquid xenon detector that I'll describe to you we measure both an ionization signal and a light signal and whenever you combine two of these different channels it gives you more information about what what type of particle might have caused the interaction okay now that's important because there's always radioactivity left over in your apparatus you do everything you can to minimize it so what goes into a good wimp experiment is trying to reduce the amount of radioactivity trying to have some immunity to backgrounds which is to say whatever radioactivity you have left can you tell the difference on an event by event basis whether whether a detected event was more likely to be due to radioactivity or whether it was more likely to be due to a wimp and by combining these signals we we get some additional immunity to backgrounds of course you want to be able to instrument a large mass because that's basically our collecting area if you will the larger the mass the more sensitivity per unit time because you've just put more atomic nuclei in the path of these wimps to give them a chance to interact because you can't get rid of all the radioactivity you also want to shield the experiment and this is led us to running these experiments deep underground and and then because the wimps are not moving very fast you want to have a very low energy threshold so okay okay so let's see so in terms of shielding radiant activity from natural materials tops out at about two and a half MeV so you know the Earth's crust is sort of for me polluted with uranium and thorium long-lived radioactivity and that radioactivity leads to gamma rays the gamma rays the most energetic ones that occur naturally or at about two and a half MeV so if you put your detector inside a water shield this graph shows that that a typical gamma ray I think this is actually an animation will travel about a meter before it runs out of steam in Zenon it will only travel about ten centimeters before it runs out of steam so this is just kind of sets the energy scale the properties that your detector and its surrounding shield should have if you if you put a meter or two of water you can block radioactivity if you put ten centimeters of liquid xenon that's enough to also block the radioactivity and this just shows in a little more detail that these sort of multi MeV gamma rays travel about a meter lower energy ones you know run out of steam after maybe twenty centimeters and they don't travel very far and xenon at all the reason this is important is because you can't get rid of all the radioactivity how are you going to deal with the remainder okay so this now takes me to the lux detector which is a bucket of liquid xenon about 50 centimeters on a side instrumented in what we call a time projection chamber okay so this is just a sort of artist illustration of a wimp coming in and scattering from one of many xenon nuclei there was a flash of light there scintillation which got detected mostly down here and then an ionization signal where electrons were liberated from the xenon atoms drifted up in an electric field and create a second flash of light okay so there's a lot of information gonna try to play that try to play that again just to talk it through yeah here we go okay so again this is a this is a bucket of liquid xenon which is our detector medium its instrumented with electric grids and high voltage to drift the electrons wimp comes in we get a flash of light these are detected with photomultiplier tubes the electrons are then drifting up through the liquid and they have a pretty well-defined velocity so that by measuring the time difference between the first signal and the second signal we can tell how deep in the detector volume the events took place we can also because we have an array of some 60 photomultiplier tubes up here we can tell the lateral position of where the event popped out we call that the the the XY position and the depth the Z position okay now by measuring the actual physical location of the event we can we can use that information in a number of ways so remember I was pointing out that the residual radioactive background is going to sort of peter out within the first 10 centimeters so the first thing that we can do is say well we're only going to accept wimp candidate events if they're kind of in the Indra in inner hundred kilograms anything that occurs in the outer ten centimeters we're gonna say that had too high a chance of being from radioactivity right now if you go back to that germanium detector spectrum that I showed earlier the only thing that was measured was the amount of energy that was deposited there was no information as to where in the detector the event took place so having knowledge of that of that 3d position in space means everything because the xenon itself is very pure but we can throw away so so it won't light itself up in the interior any background coming in from the outside we can throw a huge fraction of it away just by saying you know it was in the outer ten centimeters okay the question was is this heavy water a regular water all the fluid inside the central chamber is actually liquefied xenon so yeah and then it's going to be in a water shield and I'll show some photos of that okay okay so just some family photos of what the of what the detector looks like again it's about 50 centimeters in diameter it's made up of copper polyethylene and it's assembled by people in clean suits because dust is radioactive and so we try to keep dust away from the detector and actually humans are also radioactive you know the potassium and your you know finger oil in your fingertips is like really nasty stuff at the scale of a dark matter experiment the the scintillation signals are all collected with photomultiplier tubes copper turns out to be a very good material in terms of having low radioactivity so the arrays of photomultiplier tubes are held above and below by by these slabs of copper where it's all been carved out to hold the photomultiplier tubes and then the the apparatus itself is fully assembled and then put inside a double wall tinium vessel we we do that because the the xenon itself is liquid at about 170 Kelvin so it has to be cooled and condensed into into liquid form so the outer the outer jacket forms like a thermos bottle and then that whole thing is assembled inside of a very large water tank about eight meters in diameter and there's one of our graduate students just for scale this whole thing takes place a mile underground in in South Dakota at the Sanford underground research facility so there's a cavern here called the Davis cavern this is the former home state goldmine I don't know if any of you saw this there was this HBO series some years ago called Deadwood this was about the the gold strike in South Dakota and eighteen I think it was 1867 this is what led to not a great period in American history where you know that land was supposed to go to Native Americans and then the gold is and Custer and the Black Hills anyway so gold was mined there up until about 2000 and the mining company you know gold mining being a bit of a dirty business they were done mining gold they don't think they anticipated current gold prices so it was Indian or they had mined down to 8,000 feet they didn't want a mined gold there anymore they wanted to walk away and not be a Superfund site and so they they said well we'll donate the facility to the federal or state government if we get to walk away and so eventually the state of South Dakota put together a Authority called the South Dakota Science and Technology Authority they took over the facility and started hosting science experiments which is why why were there in South Dakota okay why are we underground at the Earth's surface we're constantly bombarded by cosmic rays and so these would directly light up the detector and also they they produce a particularly pesky background neutrons so cosmic ray can collide with say a nucleus in the wall of the cavern or in the apparatus produce a neutron in that Neutron being neutral like a wimp sort of penetrate into the detector scatter and leave and it would be difficult to tell the difference between a wimp and and a neutron so we go deep underground first to cut the flux of cosmic rays so it's much less likely that that the detector is exposed to them then we also put the apparatus in this water shield so that neutrons that are created in the cavern have to migrate through the water and then there's only a small chance that they'll make it also as I mentioned earlier gamma rays of the rock walls of the cavern are somewhat radioactive gamma rays that are coming out of the walls will also be stopped in the in the water shield so the theme here is that there's kind of a multi-layered defense right there's that outer layer of xenon to throw away background there's the water shield there's being a mile underground you have to take all these steps kind of together to get to the point where you have a low enough background in the experiment where you can hope to see wimps that haven't yet been haven't yet been studied the site itself in terms of its physics has has a pretty glorious past the cavern that we're working in this is shown in in in 1964 was actually hollowed out for a solar neutrino experiment done by Ray Davis so he built a large detector based on cleaning fluid with chlorine and looked for neutrinos from the Sun and that was a very successful experiment which led to him being awarded the Nobel Prize and around 2002 I think by 2009 the cavern had been enlarged to two how's our project and here's a photo of the of the top deck of the experiment where the various umbilicals that connect to the detector and then in the lower level this is the water tank that houses that water shield so you know we we ride the cage down the same cage that the miners used to use it's about a 10-minute ride we're 40 850 feet below the surface and once you're down there you know you put your lunch in the microwave you can make a cup of espresso you know very very civilized conditions okay so I said a few minutes ago that the xenon itself is very pure the detector has to be made of low radioactivity materials there's one hitch though which is that we get our xenon from the atmosphere xenon is about a tenth of a part per million in the atmosphere and it also turns out that there's radioactive Krypton in the atmosphere it actually comes from it's a man-made it comes from reprocessing fuel rods why would you want to reprocess a fuel rod let's not get into that it's a lot plutonium nasty stuff but what that does is it releases radioactive Krypton 85 into the end of the atmosphere so when the when the xenon is collected from the atmosphere you inadvertently also pick up this Krypton it's also a noble gas so it's as you know it doesn't it's a it's a noble element it does it's hard to it's hard to remove chemically it's not it's nonreactive right so because this Krypton would be dissolved throughout the detector volume it's not one of those backgrounds that you could shield by throwing away the outer 10 centimeters it could be in the center of the detector it has a half-life of 10 years and so since we don't want to wait 50 years to collect the xenon and do the experiment our graduate students would have problem with that we have to find a way to remove it okay so in the xenon that we purchase it's got about a hundred and thirty parts per billion of Krypton and we'd like to get it down to be less radioactive than the most radioactive component in the experiment which are the photomultiplier tubes so the photomultiplier tubes is the arrays that collect the the scintillation light top and bottom are equivalent to about 20 parts per trillion of of Krypton so we start with 130 parts per billion and we want to get well below 20 parts per trillion so what we've developed is a system that uses charcoal to do chromatography to separate out the Krypton and the xenon so the idea of chromatography is that is that you have a medium here a filter paper with different colored pigments of different size and if you wet if you wet the filter paper the capillary action causes the pigments of different sizes to spread out so that's call it chromatography we do the same thing by dragging our xenon with this trace radioactive Krypton through a charcoal column and the idea is that the in in one loop if you if you feed in a slug of xenon the Krypton will travel through the medium a little bit faster and you can trap it and throw it away and then you can run a second loop to recover the xenon so we you know you you decide I'm interested in astrophysics I'd like to I'd like to study dark matter so what do I have to I have to build a xenon detect now to do this crazy thing about becoming a chemical engineer for three years so I can get rid of the crew I mean you just you do what you have to write because you can't buy a Dark Matter detector in a catalog you know if mcmaster-carr sold one I would buy it instead of building my own so anyway I just kind of wanted to give you a little bit of the flavor behind the scenes you know this this stuff doesn't make it into the Dark Matter plot in the end but boy we sweated for two years to do to do this thing so I have a little bit of PTSD so I have to have to kind of kind of tell you about it so the way that it works is that you know that so this graph shows you the the Krypton signal as a function of time and a xenon signal after it Krypton moves more rapidly and then the xenon comes out so we built this you know basically this high purity chemical plant you know consisting of charcoal column and a condenser and gas handling system and the all-important graduate student who really sweated the details to make all this work and you know I'm here because we were successful so in a matter of months once we built this thing we we managed to take this 130 parts per billion and reduce it down to four parts per trillion and we delivered the xenon to South Dakota just in time to begin the experiment so that was that was a nice success that that we that we had so okay now I want to shift gears a little bit and tell you about what the signals in the experiment looked like and that will kind of lead up to science result so if you think back to that animation there was the initial flash of light when when the Zeno were first struck by the WIMP and that light we call s1 it's the first of two scintillation signals and it's primarily seen in the bottom array of photomultiplier tubes so now there's 120 or so PMT 60 on the bottom 60 on the top and an event takes place and we mainly see a bunch of pulses in the bottom PMT's a little bit of the light from that first interaction reflects out and gets to the top array so we see smaller signals in the in the top PMT's then nothing happens for a couple hundred microseconds the electrons are drifting up through the liquid then they pop out into the liquid and that's the part that's got a snapshot here and they make a second flash of light in the in the gas phase and then are primarily detected by the top array of PMT's and so you see a bright a bright set of pulses here that we call s to the second scintillation signal and by measuring that drift time we get the depth of the event and we also learn based on you know where where most of the light showed up in the upper array we get to measure that lateral position so again that that 3d reconstruction in addition there's there's other information that we learn about the type of event by looking at the ratio of the s-1 light to the s2 light so our primary background as I was saying before comes from gamma rays from radioactivity and the main way that a gamma ray will interact and xenon is undergoing a Compton scatter so you have an electron on a xenon atom sitting there a gamma ray comes in scatters from the electron deposits energy through the electron okay a wimp on the other hand comes in it's a neutral particle the main thing it interacts with is the Vezina nucleus okay so what we're trying to discriminate once we've done everything we can to get rid of radioactivity it's can I tell the difference between an electron recoil which would signal a gamma ray or a nuclear recoil which would signal a wimp okay and so conveniently the the s-1 and s2 signals look different for electron recoils and and wimp a nuclear recoils so I'm not really going to go through this at all in detail there's a complicated sort of atomic physics pathway of excitation z' and recombination and so on that make up some number of free electrons and some number of photons from the first interaction and the ratio of those two signals is different whether it's an electron recoil from a gamma ray or a nuclear recoil from from from a wimp okay and just to show you that I'm not making all that up we can mimic these interactions in the detector using radioactive calibration sources okay so so now what this is is a graph it's actually two graphs of on the x axis is the energy deposited in the detector so the amount of s1 light that's produced so again it's just like that original germanium detector the amount of light that you collect tells you how much energy was deposited and on the vertical axis is the ratio of the s2 to the s-1 signal the ratio of the charge signal to the initial light signal and in the upper graph we've we've lighted up the detector with with radioactive tritium actually in the form of of taking methane which is ch4 replacing one of the hydrogen atoms with radioactive tritium and actually introducing that radioactivity into the xenon stream in the experiment which is a little bit scary when you've gone to all this lengths to make a super pure experiment and then you have to calibrate it using something that's radioactive and you know you do all these cross checks to say yeah it'll come back out we'll be able to purify the xenon and get rid of the radioactivity and it worked and the challenge here is that you know with smaller xenon detectors you know in the days when you were running just say a 10 kilogram detector how did you calibrate it well with when with a 10 centimeter detector you could just bring up a radioactive gamma source that you could like get from a smoke detector and it would light up the whole detector but now if you're on the 50 centimeter scale and you really care about the detector response right in the middle what we said you can't get the whole point of this big xenon detector was you can't get gamma rays to the inside itself shields so how do you calibrate it you have to light it up from the inside how do you do that you and Jeff gain radioactivity and and make sure that you can remove it so this was actually a pretty bold step as far as we know we're the first called low background experiment that deliberately introduced radioactivity into the experiment to do this to do this calibration anyway big song and dance this to us is just like this gorgeous spectrum of simulating background events throughout the detector volume right so we can have like orbit you know a million tritium induced low-energy electron recoils throughout the detector volume and so we can study the detailed response of the detector as a function of position as a function of time measure the stability do it at the beginning of the experiment do it at the end really nail down the systematic response okay so this one graph of the tritium shows us exactly what an electron recoil background would look like in the experiment this is like having a million background events in the detector okay what about signal so these are the background like electron recoils to make signal like events remember that I said you know we go deep underground to get away from cosmic rays so that we don't have any neutrons in the vicinity of the experiment because neutrons are dangerous they're neutral they also cause a nuclear recoils just like a wimp so what we can do though is bring up artificial sources of of neutrons and have those sources just for a few hours interact near the detector and that that gives us signal like events okay now the thing that may be a little hard to make out here is that so we're using this graph of nuclear recoils that are induced by calibration neutrons we've we fit that that band of events to this centroid the solid red line and you know sort of 90 percent of the of the neutron events are contained by the dashed lines then we take that centroid and we plot it up here same scale and we plot it here and you can see that very few of the electron recoil events make it below the red line okay so the game we play is that in the actual run of the experiment will only accept wimp candidate events that occur below the red line we'll throw away half of our signal the upper half but will suppress 99.5% of the background and we'll be able to do that in a very controlled way because we have these wonderful calibrations okay okay so now what does a typical wimp candidate event looks like look like well wimps don't pack very much of a punch so unlike that calibration event I showed you earlier you know will accept a wimp candidate event with as few as two individual hits on photomultiplier tubes this is so we pushed a very low energy so we can look for wimps that have very low mass then nothing happens now this is this is like a macroscopic object it's 50 centimeters across 50 centimeters tall and for this period of a few hundred microseconds absolutely nothing happens it's just dead quiet there's no internal radioactivity everything's been shielded we've gotten rid of the Krypton the tritium source has been purified away it's just dead quiet this is like the least radioactive place on the planet is this inner 100 kilograms of xenon so it's really pretty cool I think so anyway we do a lot of you know detailed event selection on you know how big the s1 signal should be how big the s2 should signal should be how quiet it should be in between those two periods of time we're in XY the event is and so forth so we did a first run in the experiment we collected 85 live days of data and searched inside the inner 118 kilograms is about 250 kilograms overall we throw away about the outer 10 centimeters and we and we preserve the inner 118 for the WIMP search so again calibration tells us what to expect for nuclear recoil signals what electron recoil backgrounds will look like and what we find from that for from this run is that our there are about a hundred and sixty events in that in the in the run in in this inner region 160 events and there they're entirely consistent with being inside the electron recoil and we didn't get any significant excess below the red line where when candidates would would show up and this was the longest exposure you know terms of kilograms times days of exposure with this low a background that anyone has succeeded in doing yet okay now we have you know very very talented postdocs that subject the data to very fancy you know statistical methods and they simulate what a wimp signal would look like if it was eight GeV and 100 GeV and everything in between and do all this you know very detailed mathematics and then they then they tell the professor's what the what the result is so the this first science result from Lux is captured again in this cross-section versus mass plot all the previous experiments to date are the red curve and above and then our contribution after six years of work is this blue curve here so all the space between this red line and this blue line we're now saying has been searched for dark matter for the first time and we didn't see anything and that's the sense in which we're the best at seeing nothing okay so what are we doing now we're continuing to run the experiment we're trying to accumulate about 300 days worth of data and if we don't see anything in that run then this dashed blue line will turn into a solid blue line and then we'll we'll keep going okay so the next experiment that we're working on is called LZ so it's the the Lux collaboration combined with the Zeppelin collaboration which is a European group that was running also liquid xenon detectors it's going to be twenty times as large as luck so it'll live inside the same water tank but instead of a quarter of a ton of xenon it'll have about seven tons in the in the in the inter detector another thing that we're doing is rather than just surround it with a water shield which is you know gets rid of all the gamma rays where we're surrounding it with a new type of detector a scintillation detector that's doped with gadolinium what will that do for us well one of the things if we if we did see some events below that red line that suggested that they were wimp candidates you would say to me rightly so dan how can you be so sure those are wimps and not some residual radioactivity so by surrounding the new apparatus with another detector one that's very good at seeing gamma rays one that's very good at seeing neutrons we'll be able to continuously monitor the residual background of the experiment all right so we have some radioactivity inside that we haven't fully characterized in building the experiment not only will it interact in the Xenon but it will interact in the outer detector so we have like a continuous monitor of background so now you really have to bend over backwards to say you know you saw an event below the red line and you didn't see anything in your nice souped-up gadolinium scintillator detector you know now now we should really be able to convince ourselves that if we see something it's real so really in the jargon of our field we're calling this this is a discovery instrument well of course Lux was a discovery instrument but this is even this is even better okay so we're doing a lot of this work at SLAC we're in the process of building a liquid noble test platform so these are sketches that were completed some months ago and it takes forever to do lab renovations and just in the last month we're finally starting to you know install some of our equipment so we're gonna use this to you know try to master a number of the of the technical challenges that are involved in in making these larger detectors the lab was interested in having us do this because there are other applications of liquid noble detectors using argon or xenon for other applications so there's already been interest at the lab you know from other groups that are that are interested in working with us we got to do the whole Krypton removal thing all over again a factor of 100 better so we're gonna build you know another chemical plant to do this that you know better faster not cheaper pick two usually they say or one the the site that we've been given to work in is I know if people are familiar with the Bob our experiment this was the B maison experiment the last particle physics experiment that was done at SLAC using using the accelerator there so that was was dismantled and the interaction Hall I r2 is the the interaction Hall where that experiment was housed was cleared out and we were given space you know in this structure and along the side to build our test platform and then the main footprint and the building is a brand new cleanroom that's being built for the LSST experiment which is a to do a dark energy survey they're gonna be building the world's biggest camera right next door to us now you know it's funny working at the lab they they did sort of a little PR write-up for I think if you go to the slack homepage it's one of the one of the little stories that pops up about dark matter and we wanted to talk up all this cool stuff where you know the Krypton removal and all that and they're like well you know we don't want to say that you're removing radioactivity from the xenon because that will kind of you know make our neighbors nervous the only well how are they gonna dispose of this radioactivity so I was really trying to say you know we should use this as a teaching moment because I went and calculated the total amount of radioactivity in the 20 tons of zine our 10 tons of xenon that we're gonna process at SLAC is equivalent to about 20 bananas yeah okay so getting rid of the that residual radioactivity it just kind of underscores how clean the environment is that we have to make for a dark matter experiment so I probably should I should probably wrap up and just put up this last plot which shows just to put LZ in context so this is the result we have so far from Lux this is in the absence of a detection the result that we'll have from the longer run of Lux and then this is the the increment in sensitivity that we should see from a three-year run of LZ eventually we start to hit irreducible backgrounds from neutrinos which if we don't see wimps by the end of LZ or maybe an upgrade then it might be game over at that point but I have job security at least until retirement so that's good put it in the context of the field a study was done a couple of years ago to kind of put on one graph all of the Dark Matter experiments that have run solid lines and have been proposed and then you know where where the neutrino floor will kind of fully start to kick in and LZ is kind of the boldest of these promising that you know we can get almost to that a neutrino floor with this 10-ton experiment so I hope I've convinced you that dark matter is an exciting topic with an exciting future that liquid xenon time projection chambers are playing you know a role being at the you know at the frontier stay tuned for more results from Lux over the next couple of years and then on a sort of ten year time scale see if we hit that neutrino floor thank you very much village to ask the first question so when you say hitting the neutrino floor and beyond that you would have to design a different kind of experiment or you would show that there is no wimp there's no wimps yeah so so really what would happen is so that we would basically encounter a background that a we couldn't shield because we don't have light-years of lead and B we could no longer distinguish from a wimp and so by the neutrino floor I mean at that point there are Astrophysical neutrinos produced in the atmosphere produced by the Sun etc that will just create background in the detector that we can't do anything about but wim says and Dark Matter are still in the play it's just we wouldn't be able to search for them using any known means so ah so the other thing is that you know according to the theorists if you look at this graph there there are dark matter candidates that you know are further down the graph we so wimps might exist in nature but no one's thought of a way to try to detect them once you start to encounter these neutrino backgrounds so game over in the sense of you know we may just simply be beyond our reach yes very good talk I wonder is there any way to connect what you've learned the final graph LZ sensitivity with one of those first curves you showed supporting the existence of wimps that what the radial flat curve and somehow what you've discovered in your experiment so far at sensitivity is there any way to infer what the nip distribution might be around like let's say the milky way so to give the rise rotational velocity measurement yeah I mean until we until we discover wimps we can't really say much at all about the astrophysics about the rotation curve and so forth if we were to discover wimps at the current scale of experiment there's a there's a follow-up technology called a it's a low low pressure actually a similar time projection camera but it's a gas that's run at very low pressure the idea is fu instead of having a condensed so a condensed medium like xenon is nice because it allows you to pack a lot of detector mass in a small volume if we were to discover wimps then what we would that what we would do as a field is to is to employ a gaseous detector and actually you'd have to make it well below atmospheric pressure maybe about a tenth or twentieth of an atmosphere and now those nuclear recoils would actually travel a macroscopic distance in the detector and you'd be able to image those tracks and because of the kinematics you know the solar system is orbiting the galactic center there's kind of a preferred direction we like to say we're headed into a wimp wind so we could measure the the direction of those recoil tracks in a low-pressure tpc then we would begin to learn something about the the kinematics and and and know for sure that you know we were we were sort of headed into this rotation curve but just measuring energy energy depositions and knowing something about the cross-section that would give us kind of a first clue that we're kind of in the right kinematic regime but to really tie the whole thing together you'd like to go into the area we call wimp astronomy right now we're just kind of a discovery experiment is there anything there at all and then what what follow-ups might be technologically possible yeah that was almost exactly my question would it be possible to distinguish between your trigger event and your signal event to get some directionality of the recoil back yeah and there you know there was a time when you know people were building these low pressure tea pcs at the comparable mass scale to the condensed detectors like the germanium detectors and the xenon detectors but so much easier and cost-effective to scale up these technologies that unfortunately now the the low pressure tea pcs are kind of being kept alive as you know a follow up technology yeah dan a buddy of mine in Holland is working on non Newtonian gravity and I suppose he's he's hoping that you'll just continue to fail I'm just sort of wondering what about the AMX you know the Alpha Magnetic Spectrometer on the International Space Station this guy Tang whatever his name is wasting its time because it sounds like you've got got him beat in terms of sensitivity you know that there you know there are regions where where when you know this we call this direct detection you know setting up atomic nucleus in the path of a wimp and seeing seeing the interaction the class of experiments that you're that you're mentioning we we call indirect detection where we're looking for you know if wimps are present in the galaxy or you know collecting in the center of the Sun or what have you then then they should eventually reach sufficient concentrations where they annihilate and then we could look for you know that back reaction they would turn back into protons and antiprotons or positrons and electrons we might be able to see them as an indirect detection Astrophysical signal and then then there's also making Dark Matter in the laboratory and we go back to that original that so-called Feynman diagram we had you know quark and antiquark turning into wimp if you if you if you turn that on its side that's the kind of experiment we're doing we're trying to see a wimp and a cork collide if you turn it backwards then you're doing indirect detection you're trying to see a wimp in an anti wimp turn back into standard model particles and then if you run it in the forward direction like they're doing at the LHC you're trying to make dark matter in the laboratory all of those are completely legitimate enterprises and for any of them to produce a convincing result would be wonderful for the other two things because there's places where we overlap and sensitivity and then there's regions of parameter space where you know we have access to high mass wimps but the LHC runs out of energy so you know and likewise with indirect detection so you really want to look for all different ways and then to really say you've solved you know the dark matter riddle you know you want information from all those different areas converging because they all contribute you know different information about you know about the parameter space yeah a couple months ago at Stanford there was a talk about another kind of Dark Matter thing and it was the attempt to detect accion yes up at the University of Washington I think yeah he did not refer to those as wimps and in fact that's right yes later yeah accion's are extremely light particles and they're detected in a completely different way so I referred to wimps as one of the leading candidates accion's being the other so again it's a particle that was predicted by a riddle in particle physics if it could use just you know baldly use jargon for a moment in the strong interaction we expect there to be CP violation hypothesis was put forward for a new symmetry of nature that suppresses CP in fact Helen Quinn athira said slack was one of the co proposers of that model and so people have been looking for the acción in the laboratory and as a Dark Matter candidate you know to try to see so again it's an interesting place where you have suggestions from particle physics about particles that might exist in nature that have nothing on the face of it to do with cosmology but have just the right properties to be to be the dark matter yeah so so accion's very very legit yeah so are you at all close to proving that wimps cannot explain the dark matter issue or what would it take to prove that there are not enough wimps to account for the dark man see ya so you know they're on Stein said subtle is the Lord and the rest of that quote is but malicious he is not I'm not so sure because it could be you know could be down here and we could never rule it out you know I mean you can make you know with the with the kind of collisions that if you if you arrange the particle physics just wrong you can make lots of dark matter in the early universe by arranging the annihilation physics in just the right way that you freeze out at the relic density that we're looking for but arrange it such that the putt the probability of a scattering event as opposed to annihilation event is is just essentially invisible to you and so will never be able to prove that wimps don't exist but our argument is that we should at least look in this very reasonable region of parameter space and hope that nature cooperates nature gets a vote I just say that some of the great triumphs in physics over many many years have been null experiments which are really well done and you're in that game I mean you're doing a very good job of establishing what might turn out to be a null experiment the importance of that I think would be it would be a revolutionary influence on cosmology and and and I want to call to your attention if you're not already aware of it there have been developments in theory in the last in the last few years which are beginning to indicate in a very in internally coherent way inconsistent way that dark matter is not needed to account for the excessive rotation of the outer in the outer reaches of galaxies is not needed to account for acceleration of Hubble expansion and it is not needed to account for what is known about Galactic halos I have a paper here published a couple of months ago of which I'll give you a copy which makes that point and so that's also quite possible in this theory produces numbers and the numbers are consistent for these very phenomenon right thank you how deep could we go to be able to get a better you know opportunity to be able to be a deep that are the deeper minds that you are looking at yeah turns out this is this is kind of deep enough in other words given even a ten-ton or even a 50-ton scale experiment it's deep enough it's an interesting question though the so working deeper and deeper be start to become more and more inconvenient because you start to have geothermal heating and so it's just a dip more difficult environment more air handling energy that you know needs to be expended to keep physicists right and but then eventually because neutrinos Astrophysical neutrinos are also coming up through the earth when you get to a bell it's about eleven thousand feet so about twice as deep then you then you start to reach diminishing returns because the neutrinos are interacting in the earth producing muons and they're just gonna come from from above the hypothetical limbs have a habit of annihilating each other so they would dissipate over time does that create a constraint on how dense they can be does that still be around I'm not trying I just say the question again that they dissipate over time with a rapidity of proportional to the square of their density right right so they're talking like in the in the context of this already given that we're doing this experiment fourteen point seven billion years after the Big Bang does that constrain the so the whole for the experiments that we're doing though the whole idea of that freeze out argument is that you know once the wimps have frozen out there now sufficiently far apart that on their cosmological density they they will not find each other to interact with okay so they won't the the cosmological density of wimps you know is unchanged from what was imprinted you know back in the early universe so they the only one one concern you might have is or the wimp particles themselves stable or do they undergo some decay process and so one of the requirements for web candidates of course is that they're they're stable you know on the timescale of the age of the universe yeah so there were you know we're trying to get to the trying to get to as if we have some control over the knobs trying to get to the point where they're sufficiently collected that they'll annihilate away at a detectable rate but they're still so weakly interacting that those interactions are so rare that the densities won't appreciably change as far as I understand it okay let's thank our speaker again then every speaker gets this special setting mug with alien robots talking to each other and if you have further questions please come up here thank you

Channel: SETI Institute

Views: 15,448

Rating: 4.8857141 out of 5

Keywords: dark matter, slac, Cosmology (Field Of Study), WIMP, LUX

Id: ol2YZ4KnX6w

Length: 70min 1sec (4201 seconds)

Published: Thu Jun 18 2015