JQI Seminar 2/5/2018 - Philip Kim

JQI Seminar 2/5/2018 - Philip Kim

SUBTITLE'S INFO:

Language: English

Type: Robot

Number of phrases: 1612

Number of words: 10951

Number of symbols: 49939

DOWNLOAD SUBTITLES:

DOWNLOAD AUDIO AND VIDEO:

SUBTITLES:

Subtitles generated by robot
00:00
thank you very much mommy all right thank you very much for coming to the seminar and thanks for the invitations again I think it's always kind of good to be back here in the Maryland and share some of this the recent result I'm going to tell you about these stories that going on in the electronics and auto electronics in this van Della's heterostructures let me start with and what I mean by the van de receta structures I think
00:35
probably you heard about a couple of times over this type of story from various different the researchers we've been flooded by this material called the two dimensional materials or vanderbass materials in past probably 15 years or so ever since the graphene has been experimentally discovered and a lot of study has been happen quickly that people start to realize it's the graphene is only one example out of the host different type of many different
01:05
type of the two-dimensional layer the materials nature provides us materials that can cleavable meaning that the older chemical bond is within the layer atomy layer and artemon layers are stacked by the banderas force therefore using scotch tape or some other method you can even grow them and point here is you can get a really thin atomic sheets the chemically stable and yet this atomic sheet can be a very different kind depends on the choice of your host
01:36
material you can get the metal semiconductor superconductors and various different type of things right their ideas goes build that because is this interaction between the layer is a very weak in a sense they don't care too much the what's on the top or bottom chemically so you can actually bring them together put them together make the stacks and if you just choose the right elements of the DISA stack of the material you can build up the so-called functional hetero structures where the each of the component can do the
02:07
different type of activities right so in principle this can give us some of the design abilities of the putting some of the hetero structures to make this something useful it's not only just a scientific district or but already the engineers our colleague engineers that kind of bring up this material and put them together and try to come up with a various different type of electronic device in dr. electronic devices and so on so we know there are implications of the applications based on these 2d materials as well I'm not going to talk about these stories today
02:39
but I'm simply kind of showing this table that emphasizing that as a physicist also this material start to give us a new platform so we can realize interesting physics simply two different dissimilar materials with a different type of the electronic or magnetic or optical properties you put them put them together then create these quantum hetero structures can be can work as a platform that realized some of the interesting physics there I mean starting from very simple things right
03:09
the p-type and n-type semiconductor put put them together you get PN junctions and PN had a semiconductor draw structures except that in this technology give us that make them really really thin and atomically thin such that the quantum mechanics is a basic element so you have to start with so such a simple hetero structure lawn already give us some feeling that one can realize somewhat different hetero structures that cannot be realized in conventional way easily all right so that's basically the idea behind of that
03:41
and now that's not only the power point and over the years we know the how to create them at least in the reasonable length scale about microns or tens of micron length scale such that we can do somewhat experimental experiments it's not up to the yet that you can realize it for the real technological application unless you can grow in the largest scale but at least you can just do interesting physics studies the method that we heavily rely on is so-called this the core laminations or
04:12
mechanical exfoliation followed by the state Pandarus at Rose stacking method method is more or less extended version of the what you what people can prepare the graphene using scotch tape here that of course over time that we know the how to just manipulate and cleave this material more controllable way but the essential parties basically cleaved out cleave the material downturn monolayers and using these texts of the specially designed polymer you can just pick up the one by one and you can just make the stacks out
04:43
of it and more important parties once you make that the stacks as you see in this cross-section at TM images this is particularly graphing encapsulate between the boron nitride all this atomy interface is a super clean there is no impurities embedded into the desert between the layer such that you can make this atomically shop the interfaces by this kind of this type of method the boron nitride but that you see in the few example in the later we can just extend it into the any 2d materials in
05:15
principle and a lot of interesting the combination of the hetero structures can be done so once you make this type of stack there's another techniques that we have developed in collaboration with Jim Hans Gruber Columbia in a few years ago is basically selectively etch this the hetero structures exposing the particular part of this the atomic layer in this particular cases graphene encapsulating boron nitrite boron nitride act like a good dielectric it's an insulator but graphene is a kind of
05:46
reasonably good through the electronic system you want to make the content on the graphene in the encapsulation you can just attach it down with a specific chemical the recipes then expose the graphene and users can put the contact from the side making this covalent bonding between the graphene and the dis marrow turns out although this is just a one-dimensional edges you can make the extremely efficient contact in such a way and create the devices and has been shown that you can get the extremely
06:16
high mobilities and very high homogeneity in the sample and such that allows us to build the interesting devices based on this of the Vandellas hetero structures so that's basically where we are in terms of technologies I'm going to tell you that probably time allows about three stories based on these techniques that I just present right so all the stories I select today is basically related with some how interesting combination between
06:47
electrons and holes or more precise electron hole pairs that can be correlated in various mean and I have three topics I won't cover the first one is cross and reflections where the soup conductors can correlate the incoming electron two outgoing holes into the defender of a system in particular the graphing system and once you just bring to the quantum Hall regime you will see that how this acrylate electron holes in the quantum energy states can appears in there next example is again that
07:18
correlation between the electron holes but here that this exit only form is under the magnetic field across the Landau level and that's basically the magnetic stands and the quantum or drag experiments is something I'm going to discuss and the final the topic I want to discuss is the expanding today this the other Vandiver cetera structures especially PN junctions we want kind of show you that how one can create this long-lived external van der versatile structures now most of the results I'm going to tell you today is actually
07:49
developing stories except that probably the first example I just briefly share so it is the not well-cooked and many part is just kind of bit of the raw data and not much of the physical interpretation yet but certainly I want to kind of share some of the excitement that the way we are standing now rather than just kinda show you some published data but anyway let me start with some published data first so the first story is a related with the the combination of
08:20
the superconductor with the normal marrow and there of course the well known physics we know is andrey reflections so when superconductor is in contact with a car the normal conductor when you try to send a current through that is normal to the superconducting junctions then there is a one page problem so we start kind of the encounter is where in the normal conductor that major carrier is electrons or maybe holes in the in the missing electrons in
08:50
the this conduction band but this electron need to be converting to the Koopa pay eventually inside of a superconductor the way it does I can can happen is basically when incoming electron from the Cooper pair by grabbing another electron from the normal narrows and make the Cooper pairs and but then creating holes in this conduction band and this a hole in the conduction band if you just think about the current conservation along this direction should move in that directions right so you can clear it create the incoming electron then create this the
09:20
outgoing horse in conduction band and this is what we call the under II reflection and the main feature is well basically you expect to see that this retroreflected hole into the normal metal right and this actually explained the many features we are seeing that interface between normal to superconductor and this electron hole with a certain distance can be correlated such that one can use this as a effect of this correlated electron hole pair so cause particles are and repairs and often that
09:53
in mesoscopic physics the behavior of the disc related electron holes can be quite important right is there a way that we can actually directly observe this reflected the hole well in this geometry is a rather difficult because it's exactly tracing back to the day incoming electron however there has been way that one can observe this reflected hole and that's basically applying the magnetic field and this experiment was done indeed all the ladies are in the Russia as a Soviet Union back then it was following ways so is bismuth
10:25
it doesn't have an even low dimensional system it samatha it you know mostly long mean free path of the electron in the low-temperature and under the magnetic field basically when you inject the current the and the here is a voltage probe whenever this cyclotron radius of this electron meets this distance between currently injecting the voltage probe you get the district of in your signal voltage probe basically pick up the chemical potentials injected the electron from there so this first one is basically force of the orbitals and second peak coming from basically this
10:57
bounced back with a certain specularity you get the reduced de peak but still you get the peaks and you can even get the third peak speed from from third bouncing of this small effects on videos so as you increase magnet field that Peaks actually appears is reduced because of finite the specularity of the power the interface but you start to see the peaks what they did in this experiment is basically put the lad on to the example bound and what interesting thing happen is when you just cool down the sample
11:29
such that the lad becomes the superconductor then they start to see the signal actually the first signal force the the magnet focusing is still there but the second peaks actually turning to the negative Peaks and this negative focus the signal is very interesting and that's actually how it works for the we measuring this reflected Hall what happen is then when electron comes in and bounce it back from the superconductor precisely this
12:00
and reap the reflection is happening and horror comes out right but then will just kind of go through the same path as I click on path and just arrived here but all carries basically the opposite sign of this chemical potential that's why you get the dis- signals and this observation of the negative signal actually tells us that indeed bounce back the bouncing quasi particle is a hole and we can detect with the voltages right and it complained objections and
12:31
this old experiments well obviously when I just first see this these things on another it doesn't make sense to me the electron is a circulating in this way how come the hole is going to the same direction it has a negative sign of carriers now one thing that we have to notice that is a hole that created here is not just a semi conductive whole day we usually familiar with their creating something in valence band this is things say the hole in conduction band or the pair concept you can think about this a missing electron right so or actually
13:02
the group LS the hole is the same direction as the electron velocity another way to think about this is a hole is a really weird hole that you created conduction band on natural hole which actually has an opposite sign of the charges but also possess I know mass mass of the hole here is negative so therefore you actually negative with charge negative sign no mess we'll make the same direction as a electron directions anyway so this is experiment a beautifully demonstrate that you can measure this reflected Hall of course now one can do very similar experiment
13:34
in the graphene especially that I told you that we know that how to make the good content on graphene not only just normal metal but also superconductors so here is a device that made out of the graphene channel and here is my injectors here here is my current detector and here is a denying that contact we made here right so with a reasonably good engineering of this contact with an IV we can create those kind of clean under a bound and rave reflections there and indeed the measurements shows that that here the
14:05
negative magnetic field is basically distractions because if you flip this one the first peak is basically first focusing peak second peak at the a carbon basically as a positive Peaks are the background or once you get down to 1.5 Kevin we're now having become soup Condit a pig becomes a deep and negative signal appears alike similar so we know that basically we can repeat the same experiments what has been done in the 30 years ago with this bismuth right so okay so this actually demonstrate at
14:37
least we can make the a fairly transparent a we can create the under your reflections right the next thing we want to do is now because now we can realize in two dimensional electron system we can ask following things what if we just increase magnetic field and bring this system into the quantum limit so all of this description might magnet a magnet to focus a semi classical physics so we're just kind of moving around where pockets but instead of doing that what if we just can bring them into the quantum Hall limit such that now we just kind of put everything
15:07
scan into the edges state pictures so if you just imagine that what happening a quantum quantum or limits the following things happen right so if I have diseases my soup contain copper now I have this sorry this is my biasing the electrode that that is missing and this is my voltage probe so it's a whole bar but one of the contains made a big circle right and then ask what happens when the reflection is you have in the superconductors what happen is when you just buy us here and drain the Curan and measure the voltage in Hoba of
15:40
course the electron is coming into the truth energy state create this the Cooper pairs and make these unreal reflections and then the support is coming on and speaking if you think about skipping of it the hole that bounced back again into the soup and create the whole Cooper pairs and make the electrons and again that chemical going through right so in order just going through the peripheral of the discipline do this many many times and every time you create the electron and hole Cooper pairs if this happens many
16:11
many times in the end of the day you have more or less cancellation between the electron Cooper pairs and whole Cooper pairs and actually when you just live here basically this is some where mixtures between electron holes and likely 15 50 percent right so if you just measure the the out the outgoing or the down downstream of the diskette States to measure the chemical potential what you see is basically we all made the basically averaged out T vs - forget
16:43
about the geochemical petition right so unfortunate parties now if I just bring this a focusing experiments in quantum limit so like this we may not be able to see that downstream of this depart occur because we start trying to get the mixture of the electron holes throughout this candle own peripherals right however as you see in the next next plot we can actually change the story a little bit if we make that this soup conditioned electrode is a really thin so what happen is that we just can make that this superconducting electrode
17:15
really thin the chopping of this and rip convergent process or under a reflection process in following way so usually incoming electron we just color the collect electron from the same side of the conductor but what if that micron superconductor is so thin and contact with another normal matter is in the other side I can grab the hole from electron from the other side of the conductor's and create the hole and this hole will flow the same directions right so if we just create that those kind of
17:46
things soup basically we can create it the and reflection is happening in the downstream side and the let whole flow and of course all flow with opposite chemical potentials of the electron therefore I can just read off the negative chemical potentials and downstream so that's basically experiment a propose can I just make the dis electrode of a lake in and thinner than superconducting quadrants lens and create this system that whether we can detect those kind of negative the chemical potential in downstream yes
18:20
this one sure this one sure right doesn't matter because as long as what is the important part is okay so the I would say this a propeller length scale right because in this way that often people call this is what we call the unrealistic basically this is the skipping orbits that consists of the D about equal probability electron holes right so important parties even if you make this shorter but if they're I have the in of the lengths that I can create
18:52
the many the skipping orbits they're similar physics it will be very similar okay so of course if you start kind of reduce down the length scale right then then probably you yes I agree and then you start to get these things but usually here the magnetic length scale which actually is more or less in the quantum limit closest a chrono beam length scale that's a scales over the tens of nanometer so we are talking about really mesoscopic size or conductors which actually go into very similar story like this in a moment
19:24
yeah so we can even we can make the indeed those kind of devices here is the graphing device very similar to before down now I just can't rotate into the 90 degrees in counterclockwise and this is my suit conducting contact but as you see this is finger shapes and the width of that this finger is something like the 50 100 nano meter length scale or smaller than hundred nano meter length scale matching more or less equations lengths of the Niveen nitride in this particular case and indeed if you just measure downstream potentials here which
19:56
actually normalized by the current such that comes onto you the we units of this resistance and this is an equal one equal to the equal six and that's typical sequence of the quantum all appears in this the single layer graphene in reason behind magnet few you start to see that these downstream potentials that go deep below to the this is zero showing there we can just detect the negative chemical potentials coming to the downstairs downstream and this tells us that indeed we can create this and revé conflate the convergence
20:28
and we can even detect that you start to see that this signal is a relatively small it's only about the 10% of the what is expected signal which means that this reflects the conversion efficiency is probably quite a reduced down rather 100 percent but nevertheless significant come out the scanner tells us that this conversion is really happening right and to make sure that it is really related with the previous pictures at present we can make that this the finger with a la Paix large and larger so there is a different type of device the 2d to the
21:01
edge you just increase this width of the finger that negative signal we just observe is quickly drops down right and this exponential decay tells us that the calculus the length scale is about a 50 nanometer which also match with the superconducting clearance lengths expected in Niveen nitride so tells us that indeed we can create those on the downstream of this conversion of the hole in this system as long as it's a finger width is a small so that's nice
21:29
yes no yeah okay yes unfortunately I dropped out all that the following slide because I want kind of stuck into the other talks but I can show you the later so there are a few tests we can make right so if you go to the above above the T see how this signal changes up if you bias it outside of the super-clean gap how this guy changes every time so you just kind of bring it out to bring the superconductor from the super can you see this becomes severely dead and
22:06
dead on zero actually becomes positive the reason it becomes positive is remember that it's not a formula measurement doing anymore this is basically we are measuring the downstream of the chemical potentials there so there is a contact contributions out this this lead actually contributed a positive signal so evident I admit that this is a kind of few tens of all right but we have the reasonable confidence that this is indeed negative especially if you take account all this
22:38
put the contact which is now in this part right so I can show you that later in there separately but this is a nice demonstration but actually there's a kind of new spins that tells us kind of what direction we can move on on here right so there's a kind of slightly moved on way to interpret the same data and this is a following way right so if we just look at this so super cool this is a very thin I'm this close to 1d superconductor because I think this width is correct
23:07
will in the dimensions of the saprykin in quadrants length and then you wonder why I actually make the distinguish Apes why is long because there is a theoretical proposals associated with this so if you just think about just quantum ready States to create this finger first of all by the way we actually make that this trench please can remove the graphene and we backfill with this de su / conductor so you can think about without superconductor please get a graphene quantum or energy so we're just follow follow like this right so there is a
23:39
quantum or energy and there is a quantum or edges and quantum Reggie is basically we know as a chiral the 1d states but then if you think about where my superconductor is basically copying this between these two chiral edge states with the super containing order and we didn't hear about this quite a lot that's basically basic ingredient create Amaya on estates right indeed and the one over the picture is a PI just kind of coupling this to Cairo edges you can create this mir energy state one especially here is localized and the other one is probably reasonably
24:10
localized probably delocalized one way you can think about this cross and the reflection is basically those kind of resonance through this neuron a state we can create in this way right in fact this theory even just bring us the even further if you just realize this is not only integer come on affect the way you can create the smile on estates if you just bring to the fractional quantum of state theory tells us that well maybe you can create even pallet arameans with the varied weird non abelian status see in the system so that's basically exciting new direction
24:41
probably you can push along we actually have the very first step it's a preliminary data but I want just to show you so one way you can prove that is okay if it is a we create a mile on the states how do we prove that we create another one just bring them close and create some sort of Josephson junctions between them and just try to read up to their phase and braiding off and those kind of thing is kind of new direction indeed we start creating those two finger facing together we start to see that they create the Josephson junctions and already that very preliminary data
25:12
start to shows that this Josephson junction critical current can be modulated by just two quantum states and that's kind of interesting so as we just get into the different fielding fraction the conductance of the normal conductance through the system is stable changing as reflecting ponta most physics but once they get it to superconductors and their critical current also follow these steps and more or less a change of the discrete record instead is related with the quantum conductance with the gap of the superconductors up to the some kind of
25:43
some numerical constants so this study tells us that at least we can create this ravinok particles and maybe we can create this interaction between them so that's kind of exciting new directions maybe this give us a new platform to realize interesting this so topologically interesting their species based on the 2d system coupled with the superconductor okay it's a kind of example now i told you that that there's another way that i want the we can create a electron hole
26:16
pairs and i think this this is a more traditional way i you to say through this a way that we can create the electron hole pair is basically in the semiconductor or in the gapped system put the lights and create this electron hole pair and by Coulomb interaction they bound together and make this X tone often that X Tony is a reasonably long-lived although there's a tension of and all those kind of electron hole pairs they govern a lot of important property or a lot of important optical
26:47
property of the solids so we know that how to just kind of understand these optics in conventional condensed matter physics languages one issue here though is extern is always a transient object it's not the ground state properties it is excited state of the materials so depends on the material it can be very short-lived but imagine that if you have the long-lived extern and we can just kind of think about as a quad particle or composite particle you can ask what
27:17
kind of statistic this external fellow of course x10 is constituted of the two fermion electron hole so their spin must be integer therefore he should be boson right but then as a boson if you create a lot of extern and if they long leave that they may actually create the interesting the many-body quantum states such as the XM condensations and those kind of paths right problem that we have here is the usual X initial lives so we want created something long-lived that
27:47
son one way that in semiconductor industry or semiconductor research has been worked it out is make this the action long-lived by separating electron in spatially or the in momentum space such that they be suppressing the the recombination right here is a good example you if you make the semiconductor heterostructures and cuted semiconductor structures by biasing often that creating this so-called inter layer the one electron
28:19
is 1 million hole is another layer this is often called in collection formation can be stabilized and once you create those kind inter Lexton in samarkand other theater structures there has been some report that you can create this sum of the macroscopic quantum mechanical States transient but nevertheless self self-organized the clearance appears in those kind of system has been done so that is kind of good motivations we can just adopt from this American data structures and just ask
28:49
that can you realize very similar system and accidents and can long-lived right by the way there is actually way that you can make the Exxon as a ground state properties it's not through this semiconductor heterostructures creating electron and hole by the light but applying the magnetic view so let me just kind of starting with that topic first imagine that I have the two conducting layer it can be semiconductor or structures it can be two piece of the graphene doesn't matter right so I have
29:21
that this through piece of the conducting layer two dimensional system and apply the magnetic field and under the magnetic field of course the Landau level forms right I can visualize this form of the land Oliver is basically a bunch of the cyclone orbits is occupying in my sample right and imagine that I just kind of partially occupy this Landau level in both of them right once I just fully occupied this land Oliver each of the Landau levels developed a quantum of states and then current will flow through the only the edges but once
29:50
you have this partially fill under level meaning that only just kinda partial number of the sector orbit got occupied in the post of the layer it is just kind of bad metal right I think it's conducting but it's a weird marrows and it's basically not a good metal but it's just dissipative system right however I just kind of put just these two Landau level or two layers in in the such a way as an individual layer they are partially fear but as a tour or they are
30:22
full and Oliver in other word I just feel this land up in two layers in compensating way such that if I just induce the grab the mold all of the electrons there you can fill up at least one of the level right and visually what it looks like is following things if I just kind of bring them close of the dis to layer what will happen is then electron in the top and electron in bottom start to see each other and when you just bring them close each enough then there is a the exchange
30:53
interactions that happens right which means that Pauli exclusion principle star can occur kicks in what does it means they're actually electron that underneath of the despair immerse that avoid electron from the top and that then is scale is based gloominess skill mediated by the exchange intentions so if I just put them really close by then you can imagine that all the electron in the bottom layer will avoid each other since we just put them in the compensated way so you will see that as this distance
31:24
becomes a smaller you will see that there is a profit correlation between top and bottom layer will happen one way you can view that is you just kinda make the projection from the top and just to look at it right if you just put the D very small this is what will happen right and as a whole and Oliver if I just projected this is a kind of complete fill under level which means quantum are in fact wise where I expect to see that there is edgy state happens and we should behavior like this the one single and our level of the quantum of states right but of course in
31:55
microscopically if you just look at them there are small separations and then basically the bunch of the hot that typo should be their electron holes down there the dipole moment were just pointing out they up and down but complicated or mathematical expression is something like this but nevertheless as out just coarse graining out they should behave like the one full under level right in fact that this experiment was done from the 90s there is a heroic experiment was done in Gmail genistein Gruber Carter they made
32:26
this quantum or bilayer and make the individual content on each of the layer and they see the following things so basically there are few different measurement you can do you can descend the currently one one layer and measured current in the same layer the voltage in same way or different layer and if you just measure in the different layer this is often what I called the drag experiments so send the current and measure the voltage another layer if this two layers couple there something's happen right so what they see is kind of quite amazing things of course they have the all the quantum or in fact in
32:56
individual layers but when each of the layers about half fear suddenly some things happen so if they measured in half a lambda level they actually behave like the full under level as long as they are compared have fear and fear then the system behave like that again it's an equal one from quantum of states but even more exciting thing is if send the quarantine one of the lair measured voltage another layer it's a drag measurement especially if you measure a whole drag measurement they see the whole drag is quantized as the
33:29
same content quantum Hall quantizations and showing that this Thule is a kind of complicity couple together this is quite exciting part and later experiments they actually showed it is also if you just measure the one layer in the one direction and measure the other layers so called the counter flow geometry they see the perfect conductors if you send the current in one layer and measure to try to measure another layer there is a cough occur on to drag in opposite direction is happening so there is a some sense that this is some sort of super fluidity of the system happens so
34:01
that one of the interesting the interpretation is indeed you have this magnetic system forms under the magnetic field crosses the the logged out level and this magnetic stands condensed into the superfluid and created the superfluid states of the magnetic stone is interpretation of the desert the Seminole experimental data that they have of course we can do the very similar thing into the graphene right again that replacing the gallium a set of hetero structures to the graphene
34:31
make the content of the individual layers and we have the top and bottom cond are decades unlike the gallium arsenide we can actually have the huge' to nobility of the descale to create very similar experiments not surprisingly we see the very similar result when we just measure the individual layers we have the nuke 1 equal to the signature of fraction is also appearing but important parties when they get into the heavily under level very similarly that soup the quantum mode effect becomes a vent
35:01
re-entrant but at the same time if you measure this drag hole drag hole is quantized is the same value indicating the very similar experiment as the the gallium arsenide case actually happened so not surprising as long as this physics is a general for the any 2d system under the main net fear this is what is happening in the graphene is of course comparing the gallium arsenide and as the scale is much higher this is expected in a sense because if you put this
35:32
and to graphene layer together you can just put them together very close by only two or three nanometers away so you can think about electron hole can be really close by there by the energy strong right so all the energy scale becomes a stronger such that this excellent formation will readily happen the condensation is happening so unlike the guard oh my sin aware that you can only see this 100 million Kelvin ranges this experiment can be seen even 10 Kevin ranges you start to see that signatures of the dissection
36:04
condensations what is more exciting part is we can also repeat very similar experiments in gallium arsenide in here that we can just the send the current from the table layer but then we just we connect that from the to the bottom layer I think this first experiment was done Ian Spearman here in gallium arsenide but we just repeat very similar thing the graphene but then you just kind of measure the voltages in this layer or door layer doesn't matter so basically we want to kind of see that whether there is any dissipation component in this this configuration
36:36
this is half a lambda level without this if you measure just kind of this patient there the drain there you you get huge this patient but once you use this to return this and drain from there and measure this patient that this patient is gone when this the magnetics and condensation is happening right so tells us that it can indeed kind of feeling of the super fluidity is happening there right one other important part is actually in graphene since we can tune this the density in wider ranges unlike
37:08
this the effect is seeing only half-filled and álava in gallium arsenide we can see that it happens on the many different under level for example in this particular case is a bilayer graphene - the bilayer graphene and we see that the nice drag effect appears her fill and Aleppo does the same thing happens for heavy land Oliver and the twin - and Oliver as long as this land Oliver has a right symmetry in the system we can always create all those kind of the drag effect this is a single - single layer this is it most
37:40
recent experiment with the even higher quality because we start to use graphite packet and up gain and you start to see that that this is a direct signal and I can highlight a few part and this is unique one you could do total equal minus one and one terrico two or three and whenever there is at the signature of the direct signal appears if he zooming you see the very similar the quantized whole Drake appears we know that the effect the
38:10
magnetics turn a condensation happening and for example here that new total you want if I just cut it through that when the the tool and error level is a compensate field in companies compensate Tilly then you start to see the important parties that what is it okay so drive and drag and everything is more or less go to zero in here and the whole whole drag and whole drive actually goes so they write values the quantization so we know that it actually happens for any
38:41
cases now the real story that I want to tell you is now this is background but following so let's zoom in one of the states here for example new Total Recall negative one and although very similar physics can appears in other places I want to measure the count of flow threat and that's basically as I said I'm sending current in the one direction and we turn into another directions and measure the the longitudinal resistance that's a counter flow resistance xx component and you can also measure
39:12
counter flow X Y component but let's focus on X X component I paint the pictures that if the magnetic Sun condensated it's actually view as a superfluid right and one of the evidence is that this quantity goes to 0 right so I want to use this counter flow X X as my experimental tools whether there is a super fluidity happens or not that's basically experimental observation so if this goes to 0 that's a superfluid if this is finite I know the system becomes a dis purty
39:42
right and then I showed you that's a that happens in low temperature and high magnetic field but we can ask that what happens if I just try to do this in the different temperature in different magnetic field right I can just draw the following phase diagram so this phase diagram is constructed following way I controlled the table am back at table and bottom gate voltage is such that as we tune the magnetic field I try to stay on always both of the layers a hair fill under level right so of course as you increasing magnetic
40:14
field I have to increase my gate voltages increase the density to make that but assume that I'm doing that right and this measure the counter flow xx XY and just record that and then I just change the temperature so right and this bluish region and our polar region this is basically where a counterflow X X goes to 0 so I know that okay so as it should I'm seeing that there is some superfluid condensation is happening there right but as you increase the temperature well that eventually this
40:46
becomes a higher resistance something like the tonic law basically this becomes a bad mirror as I said turning to the it's bad marrow in other words condensation is a breakdown on a high temperature which is also expected right but you start to see that have the feeling that okay so that's not how they break it down has interesting shapes here right so it's like the dome like the shape as you change a magnetic field and you start to see that roughly here the change is gradual but change is a sharp I can show you that or even more detail is there so I call that this is
41:17
condensation per phase and this is normal phase and it's this kind of phase transitions happening with different behavior and if I just kind of cut through it from the magnet constant magnetic field and temperature behavior this is a behavior right I say I say this is gradual which means that this blue line is gradual this is a rather steep sharp changes almost look like the superconducting transition right find a conductivity resistance drop down to the zero suddenly right and something is the middle right in fact if you just plot
41:47
this one in the the RNs plot you start to see the interesting low temperature a low magnetic field behavior is more like the activating behavior in fact this is what has been seen in gallium arsenide when you look at this counter flow or any of the this kind of experiment all the parameter there always activating right so that's not surprising we see the activating behavior but the energy increasing magnetic field you start to see the activating behavior becomes a worse or you can see that even just the two slopes or whatever right
42:18
but sharp transitions but certainly something is happening kind of change is happening and there's interesting phase boundary we have to understand this type of things right so this part of this the experimental phase boundary immediately actually vocals very analogous pictures we have a scene in other system I just borrowed this slide from the Ogawa's talk it's actually that a phase diagram that sketch it for the xvo accidents and
42:50
extent condensation or further of course for the rear extern if you just put a lot of externs axons that the screen each other such that xn becomes unbinding and that's why we call the mode transitions so if you have the extant gas if the to high density basically xn gas will dissociate with electron hole kozma now if you just think about those kind xn system if you cool down low temperature you know then of course Exton as a boson can condense into the bose-einstein condensation of the X turns but electron hoard plasma
43:21
here well if condensed there if it's in the right condition may actually condense into the superconductivity in separate electron holes right so there is a kind of bcs type of superconductor happens and in fact the theory tells us in this system maybe there is a smooth crossover between the specie mbc s on this axon system in the low-temperature there are some some things that developed in this line although experiment observation is Arabic controversial in this directions right very similar and although they actually can be applicable for the magnetics and I was talking about right the theories
43:52
are a bit more complicated but you can it more or less see the important part is here that is theories governed by the two important line scale one is what is exercise a Bohr radius of the Exton and versus what is a distance between the extant because if the this distance is a large enough the way the Bohr radius extends the well separated we can just view as action as individual quantity but if you increase X intensities such the Bohr radius versus this X and distance becomes comparable then then xn is not a good quantity anymore
44:23
there is a screenings and all of these things very similar way if you just described the the magnetic stirrer in similar way the important to line scale is one is distance between the layer which actually determine the X and size excellent distance is governed by the magnetic length basically by tuning that change of magnetic field you can tune this a distance of xn because that's basically distance between the guiding center of the section orbits right so very similarly we can see that D versus
44:54
a D which is fixed by the experiments versus lb magnetic length scale which actually change tunable by magnetic field will plays important role and we can actually tune over this type of this the interaction between the X turns right indeed if just kind of look at in my experiments that I showed you here you can do the following things right in this the when magnetic field is small in other of the magnetic lines is law in this a t / lb t is much much smaller
45:24
than magnetic length what you see here is a smooth transitions and that's where that I get this activating behavior as you increase that the the magnetic field will become small and becomes comparable you start to see this activating behavior start is poised right one way you can just do that is it just a draw that how the activating energy see if you just get the slope that changes as a function magnetic field low magnetic here we have the really well-defined deactivating behavior as you go to the high magnetic field you
45:55
start to see that there are multiple slopes appears if the dive kind of spread it over although that maximum flow is following through this similar line and they can you can even just do this experiment with a different thickness of the layer and you can nicely can extend this curve but nevertheless we start to see that they're activating behaviors what is governing in this ranges but that Sotka changes right I mentioned that high field there is a stiff changes of the resistance right our current flow resistance almost like
46:26
the superconducting transitions in fact actually is a funny that if you just measure carefully IV characteristic secend the current I measured voltages and this IV characteristics precisely follow the powers role at the low temp and as you or at low bias and as you increase it the high bias it becomes ohmic behavior in fact there is a sensitivity changes with the temperature and all this behavior of the IV anyone work in the 2d superconductivity immediately notice that ah this is IV for the 2ds conductors where Katie
46:57
transition is happening indeed precisely that IV characteristic measuring can be described of Katie type of the transitions the power-law behavior IV and from there from this exponent we can extract this Katie transition Co Celeste doubtless that transient temperatures and plot it out nicely actually scales with the D over L in proportion to the magnetic field scale right so we start to see that a lot of things that we just seen can be framed in terms of this
47:30
super conductivity or superfluid type of the transitions as kind of indicating from these pictures so I don't have a lot of time and this is a kind of brewing story so I don't know what kind of definite tom but roughly this the experimental observations that can propose following things right so basically as you change a magnetic field that we just going through this one region today the other regions right more like that there is some crossover or behavior changes this only
48:01
suggests maybe we are getting to the dis BC type of region where the axon is a well separated and the well described by the individual action to the where that in this that the the magnetic length is small such that all the action is basically all incoming go where the action picture is not good there must be some behavior changes seeing the our experiments third one is temperature dependence only indicating indeed at least in the dis high field the limits there is a kind of cleared behavior of that this KT type of transition is happening
48:32
so whatever other parameter it is that indeed it is two-dimensional system and it's condensation phase so getting out from this condensation phase basically observation of the circuit if physics is rather natural things we are seeing of course that this is not all and we have a lot of the unsolved problems here and what is the reason of the disappearing behavior might be there is a kind of some sort of Maron type of physics if you think about this is to the XY models that in the limits of the co2 symmetries
49:03
in the equal zero limits that's maybe natural things or maybe the other one is what is a real nature of the dispise limits of the ground states in this one how they created is dictated fringes there are some of the theoretical the predictions there I don't know that whether how well they are matching with this experiment at this point but certainly this is kind of big open questions we are facing in here right so I have probably five minutes that that quickly go over the last project now
49:35
everything so far I just mentioned is involving the graphene but in the beginning I just mentioned the graphene is only a good but one example of the 2d system in fact there are the series of the 2d materials especially semiconductor we called the transition metal nature coordinate it consists of charcoal gene atoms surrounding that the marrow that usually the tungsten moly and those kind of things then it becomes semiconductor this is what we are just
50:07
if you just look at the projection you look like the more like the graphene look like the structures or more like borign like the structures the inversion symmetry is broken here but also on the top of that there is a strong spin orbit coupling because it contains the heavy metal will make that this spin orbit cup the band structures is spin splits right in the end of the day what you find is the conduction band and valence bands are kept at the beryllium zone corner
50:36
and also they are spin split because it's depend at the brilliant corner the value index is very good so K and K prime index is a very good way and they can be optically addressed because you can use a circularly polarized light to address one delivers to the other but because there are such people arise which means that using the optics one can address a spin of the system so in their regard there's a lot of excitement maybe if you tame this material so one can use the combination of electron electronics
51:07
and optics and realize some of the quantum operation to the D is the system furthermore because of this is so thin and all this electric field when you create the electron hole pair all the X tons coming out from the demuccio's and just in principle go through the free space or some dielectric so if you so choose your dielectric in free space right there you can minimize the screening Coulomb screening in other words that your X term becomes strongly bound so indeed experiment actually
51:37
tells us that you can create the strongly bound X tons and because X tone is a strong rebound that you can also use you cannot touch another charges on next and the so-called 'trying can be stabilized and one can use and I don't need just a neutral object but discharged of charge it Exton you can just create it and many plate in the system and there are in the field there are a lot of reports such as the not only accidents or trials by extends and they are stable they can be manipulated
52:09
by circularly polarized Estonia integrals visual challengers choose a group result they all report that there are a lot of interesting of test properties one can many plane in the system as a transport group what you can do is we can actually make the device electronic device and combine with the optical properties and measure these things so this work is in collaboration with Duncan talks group and Misha look in scrub at Harvard basically showing that once you make the device transistors and we have to put a gate
52:40
you can just not only do that transport measurement but you can just do the optical spectroscopy as well here is I'm showing you the photo luminescent spectroscopies as a function of the gate voltages in particularly in here that we have the two gates which means that not only I can control the density but also I can control the electric field in the vertical directions and that's important because a field is a readily copper with axton's right the result is interesting here that in this plot this is the energy scale of the photo luminescent
53:11
and that bright red is red is basically peak of the luminescent blue is background what you see is when you put that zero density in other word when you put the chemical potential in the gap you see the multiple peak so here's one two three or maybe more right so you cannot sign this is extra neurons and other trials depends on this or the sample preparation it has been working out quite well people but then when you apply the gate portage is such that you
53:42
start to change your density in the electrons and holes and all this the optical spectrum modified right especially exon will quickly dies off the one so you drag out from the gap region because now everything becomes screened by electron hole and that quickly becomes a try and type of the pigs and their energy got modulated as you should write because the screen protocol changes but what is interesting is following graph here that we apply the electric field in the vertical direction by charging top and bottom
54:14
voltage gate voltage into the opposite polarity right and fix the identity and then what you see is that that intensity are modulated but exit on and trials they a peak position doesn't move with as you change the electric field and this tells us immediately optical transition dipole moments of the dis optical species we created is all in plane therefore auto plane make an electric field will not couple with these energies right so it immediately tells us this is two-dimensional object
54:45
okay so that's that's kind of good or surance but then of course we can create this auto plane object or the three-dimensional type of externs now by putting this a two layer together right so if you just put the P layer and layer together and make the district hetero structures right you can immediately see that you can create the exon of the each of the layer but then if you just move the carries one layer to another you can create this inter layer externs there right what is the experiment we want to
55:17
do is basically we want create those kind of interaction by optical means but also we want to control them by electrical means so we construct the device with the PN they're together and but then we have to put the all the contact in or surround the soil each of the individual layer like the graphene device but also we have to put this top and bottom gates so that this is kind of finer device in shape is it's a really hero efforts many many electrode in each of the layers and control on the top of that way of the top gates and varam gates matching gates
55:50
so many many control but in the end of the day when you just take the optical spectrum you will see following things this is only that constantly die selenide only this is molded at I solenoid all inner layers where that they are on there at some Peaks appear in the photo luminescent what is interesting part is when you just look at in the middle ranges where the two layers overlap you basically suppress individual layers optical the photo luminescent and get a photo luminescent of the intellection zoning and not only
56:21
you get this intellects and kicks they can be modulated by the gate voltages very similar way as you change the density they got modulated but more important part is now when you just kind of change the electric field in the vertical directions unlike this the single layer this the two layer intellect stands they shows huge energy changes just kinda linear stark effect tells us that basically action is aligned with electric field we are directing so we know that this isn't alexinnz more exciting parties in fact
56:53
the day a lifetime is a fairly long int early enteral a extent is only just kind of set attends a picosecond of the lifetime here that this is hundreds of nanoseconds almost kind of microseconds along lived excellence and this is idea again the electron holes are leaving the different layers right but the important part is their lifetime can be also controlled by the electric field in a sense by applied electric field we can just kind of move the electron hole pairs actually move further away and that basically make their lifetime long
57:25
when longer so we know that the lifetime can be controlled in the very long lifetime and long lifetime is very important because that's where that we can make a lot of electrons and eat this is a good example this is our hetero structure area we put the lasers there excellent generated and they start to spread it over because they're long lifetime we can see that how they actually spread it over since time is a kind of rather short and and from from there you can just kind of measure their what is a diffusion constants of the dis
57:55
the extent the fusions and so on so I think that's kind of good directions we know that we can create a long long lived axton's we see the how they actually moves around one thing that noticed here is you see that that this is a folding edge of the distillation post and then xn does not disappears uniformly but there are some spots in the sample and as you see here that in homogeneity actually appears so clearly our sample has a lot of image in homogeneity we don't know what's their nature but they are there and they
58:27
actually control the X and dynamics we believe they are some sort of excellent trap because they are brighter but when you see is once that extent is that trap we see that X only is a rather long leave so the next level of the idea is can we actually engineer this excellent trapped to make the deceiving long believe and we can have the book control we don't have a yet path stages but at least that when we measure this external density by just looking at the blue shift of the photo luminescent we start
58:58
to get the sense that excellent density is something like the 10 to the 11 X 10 per square centimeters before we heat up the samples in substantial level and that's important because this X intensity is exactly tells us how much actually we have to generate action and what is the time what how low temperature we have to go to create this X and condensation according to that this mean field type of the simple calculation has been done on to the descent electrons by the Fogler we are around kind of disre zooms so we are not
59:30
yet this condensate a regimes but it's not too far so the further engineering of the de sexton to make this a higher density and long-lived is important and we believe that by engineering tools we can just do that time is a really op so i don't think that i want to go over to details but it turns out that we can not only generate action by the optical meanza but since we have the contact by PN Junction we can create action electrical means electrically generate excellent they also long-lived but also more important part is a let me stop
01:00:01
enormously this slide we thought actually put the gate local gate every start just to demonstrate that indeed we can just move around this optical species try on for example to create the externships we have to create the dielectric traps but this is already demonstration we can start kind of engineers optical species in that the kind of several occasions something like that demonstrate here alright so let me just quickly summarize so I just try to tell you about it the three episode over
01:00:33
the electron and hole pairs in here right so using either superconductor or the quantum or bilayer or VI PN junctions the kind of demonstration we can create these X tones or the electron pairs that can be strongly correlated but important parties each of them actually is that provides us some new platform to realize interesting physics for example this correlated or whole pairs may actually going for the new platform to the the qubit table topologically protected qubits ideas or
01:01:05
some of the other exerting many body states and so on so I think this is in a sense a very exciting field in their regard all right so my final thing is I want to really thank T my collaborators without collaborate this never work and my students and postdoc we actually did the work the first part I just mentioned the cross and the reflection was done by the kill Holly who is now in post in collaboration is awesome ER cookies group magnetic stone was done by the charming Lu in collaboration with the Cory Dean's group at the jelly is
01:01:37
actually co contribute and Peary wise we got a lot of help from Portland have both happily an interaction was done by the Luisa with in collaboration with a mission looking around campus group thank you very much [Applause]

DOWNLOAD SUBTITLES: