Philip Kim – Materials in 2-dimension and beyond: platform for novel electronics and optoelectronics


thank you very much for coming and thank
you for the invitations and it’s my honor to be here and give it this
distinguished lecture series in both Institute so I was really impressed in
the past two days the visit here and this exciting research going on as well
as the facilities and people here thing is that very impressive in many aspect
originally when I was invited to give a talk that I didn’t give a talk title so
until quite recently but then I found that somebody put in my talk title ready
and the title that they’re listed here that as well you can find the
poster is actually the title that I didn’t give I don’t know somebody
actually assigned to me but then I reassigned my talk title as a bit more, some of
the research that related to my research but then actually the reason that I just
bring it back here is in past two days that my experience here I start to
realize that a lot of things in this title is pretty much a generic title
that also related with something going on this research center that whether I
want kind of make a little pitch on to the my research which is a two
dimensional systems tied with this quantum frontier and that a nano scale
so we’ll see the how it goes alright so as Holger said that the main topic I want
to discuss is about the particular class of material there I’ve been working on
it’s called a two dimensional material or the van der waals materials system and
it’s material that we’ve been known for quite a long time nature provides us
a lot of so-called delated materials graphite is a good example in this
material that the chemical bond is a within atomic layer but between the
layer there is no real chemical bonding it’s just the weak van der waals which holds
all them together because of this weak van der Waals force nature the worst
first thing you can realize that many of the
this material that simply can cleave quite well using scotch tape that
peoples in about 12 years ago demonstrate you can break the graphite
down to a single layer of the atomic layer of graphite while you hold the
graphene and that is possible simply because the material is is stable
because all the chemical bond is within the layer it turns out that’s only the
beginning and soon we start to realize there are all other type of the
so-called layered materials or we can call the van der waals materials that you
can increase we take this material down to one atomic layer or in other word
you can also grow this material using CVD and be in various different type of
synthesized method you can grow this material stabilized down to one atomic
layer such that you can effectively create a two dimensional material system
and that has been quite exciting first of all the material comes with the
various different flavor graphite that i mentioned that turn into the graphene but if you just go above this at the time below the periodic table there is solisten
domains and all this if the number four families can generate something like
graphene like the structures they can be stabilized down to one atomic layer and
they tend to usually shows the interesting electronic properties
so-called semi-metal or zero gap semiconductor all durán materials in
modern terms but also some of the material becomes an insulator boron
nitride that I’m going to discuss is a good example or some of the materials
especially transition metal dark or denied which consists of the transition
metal with charcoal gene atoms such as sulfur selenium and tellurium
those kind of materials often demonstrate as a semiconductor pure anti
or sometime we become superconductor if you just use an item for example or some
materials becomes the strange matter such as charges wave system and the list
actually grows every day and we’ve been just enjoying to see in past five or ten
years that what kind of materials we can create in the two dimensional limits and
one can study a lot of their interest in electronic property in this
the in a sense quantum limits of this 2 dimensional system now one another
advantage e we can see is in this that captured in cartoon that which was a
listened by the way physics today last year it’s kind of following tom not only
you can study them as a down to one atomic layer limits such that you can
understand their interest in quantum physics in principle you can stack them
together this is relatively easy in part because again it’s a big van der Waals
force when you just put them together you don’t have to worry about their
lattice mismatch and just commend your abilities they simply just kind of put
them together and because of a diverse interaction they just can hold them
together and yet there is no real chemical bond so they just kind of exist
a data stack system so this ability allows us to the following things not
only you just kind of work on single layer system you can just bring it back
and stack them together to put some of the these complicated structures either in
plane or even out of plane no plane but even the same plane even
further because as I said this material comes with a various different flavor
you can think about this is a unit that just displays some of the interesting
functionalities right so if you just choose by just interesting design like
as if the you just block the kind of fact it’s a Lego block in principle you
can stack this Pandarus materials to try to come up with quasi three-dimensional
structures with some of them interesting the functionality basically this idea of
a Hector structuring of this functioning system actually bring us a quite
exciting new moments such that we can probably study this material down to
atomic length scale and look at this the how the quantum mechanics plays all but
on the top of that one can actually build distant into this some of the
functional devices we’re not even individual layer see that
that the properties that if appearing in the interface actually exceed that they
are expectations coming from the silicon individual materials right it’s not only
just kind of science fiction’s but indeed
in past five years people have been moving on to demonstrate some of the
devices starting from the simple field effect transistors or interconnect or
biosensors the tunneling device memories LEDs and now kind of more of this
electronic and optoelectronic devices based on this combination of 2d
materials and even just try to build up this flexible devices there are least
actually growing up and a lot of the in tech interest to use this material as
some of the application has been come around in quite recent years I’m not
going to discuss about the detail about the application simply because I’m not
good at this kind of real applications but at least I want to kind of address a
part that what kind of the capability that we’ve been stuck kind of developing
and moreover what type of the interesting the physical property we
start to see especially electronic and electro optical properties that might be
used for some of the realizing conventional device I would say but even
further beyond this conventional device where the quantum mechanics plays quite
important role the method that we’ve been using is a rather simple things
it’s basically what I call the van der waals reverse a hetero structure this method
has been developed out of the collaboration that I had at Columbia
University a few years ago especially together with the Jim horns group at the
mechanical engineering but this can be easily generalized quite a bit so of
course you can get this deep materials using various synthesis method including
the hetero structures such as MB MOCVD is a method that people try but even
before that if you are not worrying about the scaling about the idea
following idea works quite well so I told you that we can cleave this
material down to single atomic layer on the surface say silicon oxide surface or
something and then you can prepare the polymer tapes or polymer the membrane
specially designed such that it may not lived a lot of residues and we can
control the origins and then you start with pick up some of van der waals system
you want to start with typically boron nitride is a good system because it’s a
good engineer and then you bring down this boron nitride on to the another the
single layer system in this particular case the graphene and you just kind of
change the temperature of the substrate
controlling the adhesions between the silicon oxide this material and mature
to the boron nitrite you can controllably pick up this much yours say
from the silicon oxide and attach to in boron nitrite right and then you can
start to repeat this process and over and over again and other materials in
this cases you can pick up the boron nitride again or other the layer the
materials and then in the end of the day you have the distal stacks of the
materially different compositions in there now it depends on the sequences
you have the different type of things but the important part is if you just do
it in really controllable way as you see in this cross sectional TEM image here
is a graphene encapsulated in the boron nitride important part is that this
interface especially this hetero interface between graphene and boron
nitride is the extremely clean there is no single atomic defects you can see in
this cross sectional TM images all this a dot and impurities basically squeezed
out from this vanderbass interface such that you have atomically shaft
interfaces possible not only you can get this mb type of the screen interface in
principle you can just kind of cut this these texts at a very careful manner and
expose certain atomic layer that you want to make the contact and there has been
some method that developed that you can selectively contact the the layers that
you want in this particular graphene is contact with this gold and make the
extremely good contact so today you can make this a device that based on this
the quantum Etro stretch structures where individual atomic layer can be
contact right so that’s very important part so in a sense you can mimic that
semiconductor heterostructures people having made have been synthesized using
mb technique but here there’s a poor man’s mb technique but nevertheless it
works rather beautifully especially in the graphene now we can
make this type of devices such a clean limit the electron mean prepared
measured in this type of the system is basically exceeding several tens of
micron in fact that electron mean free path is set by the sides of sample we
can get a list in the low-temperature means that we can make extrema clean
interface between graphene and boron nitride this type of the clean sample
immediately allows us to do a lot of exciting physics this is a good example
that once you apply this strong magnetic field you start to see that your
transport shows this quantizing effect what you call often the quantum hall
effect but it’s not only quantum hall effect it turns out there is an
interesting interaction between the graphene and boron nitride what you call
the more apparent forms and give us the super lattice which actually quantized
the graphene band even further and create this all these kind of
complicated structures I’m not going to discuss anything in detail here on this
one because it’s Arabella’s static subject but nevertheless this problem
actually gives us exciting these new physics that how this the commander
ational in combination lattice structures in fact with a magnetic field
to create this a fractal like the energy spectrum in the system and this type of
the experiment is our only possible that when you have the extremely clean sample
that available in the extreme quantum condition so this is a good example that
what you can do when we have the clean sample but this is the beginning
it’s not only graphene but you can make this all this hetero structure in
various semiconductors and the other system in this particular example we
have the boron nitride encapsulated sorry graphene and moldy disulfide is
encapsulated in between boron nitride make hetro structures that graphene
onto the moldy disulfide I serve I didn’t make the good contact between them so
graphite in this fuel a of the graphene server is a good content on moly
disulfide and nevertheless moldy disulfide is encapsulated in this boron
nitride give us a really good channel and entire
these structures can be work as a transistors and their mobility is quite
good something like easily ten thousands under the magnetic field issues are
gained quantum oscillations so you start to see that this example actually
extended beyond the graphene one can make this come interesting these quantum
devices as well not only just kind of this magnetic field and road temperature
quantum device but you can also make this some of this known the other
quantum device a beyond the CMOS here is another good example in this particular
example we have sorry about these small characters but here is a bit tungsten
di cilinide p-type semiconductor and another
tungsten dicilinide but in between that we insert a very thin layer of the boron
nitride as a tunneling layer right again this made made out of this successive
stacking and then you can clearly see that between this two layer because it’s
a Saltine boron nitride is a tunnelling device now this type of the tunnelling
device is what we can make we the so called the quantum well resonance
tunneling device that out of semiconductor heterostructures big
different see here is we have the gate from top and bottom such that we can
control this band alignment between these two materials right in the net
when you measure tunnelling current through this device the tunnelling current
at the fixed the gate voltage for example shows increasing tunnelling the
current as a bias voltage and there is a peaks and there’s tips and going up
again so this peak appears in the terms of the bias is what I call this the
negative differential resistance in a sense there is a negative slope in here
is they actually indications of this resonance tunnelling between this so
2 dimensional system so has been demonstrated again
they say many of the discs on these quantum hectro structures as a key diode is a good
example but important part is those kind of this this signature of the resonance
tunnelling peaks can be movable by the gate voltage we apply in the system
showing that we have the control complete controllability in those kinda
hetero structures and we can build up this type of interesting devices as well
so not only this type of the electronic device it turns out you can also turn
this in the electron opto electronic devices in this particular example that
we have P PI by n type semiconductor with them together again very similar
fashions and you can see that this is the thinnest PN diode that you can make
either P or n Junction is only one atom thick materials but exciting part is
when you just bias this one is that kind of illuminating the light like this the
LED light enemy emitting diode and that the spectrum can be tunable by the gate
voltage again simply because we can tune this parallel alignment right or if you
want you can use these are these photo sensors if you shine the lights and
measured you get the photocurrent actually is
different it’s a kind of you can view as world thinnest a photovoltaic for
photovoltaic diode but variously this type of the device operation can be done
in the extreme that when you make that this sample is extremely thin right so
this is kind of good starting point that what kind of the application we can just
kinda launch you on this extremely 2D limit of the system now I know that
up to here that is a good demonstration but probably a lot of the my physics
colleagues may not be too happy to see this well this is electric engineering
again may I know that you can just make it thin but what we can learn from here
now exciting part of the this two dimensional system and the hetero
structure is actually allows us some things go beyond that what I just kind
of showed you at the example of the conventional electronic devices we’ve
been known right so let me show you a few example that I am pretty much
excited about the opportunities it’s by far its ongoing project but nevertheless
was you will see that some play but what we are excited about just kind of using
this type of to this system so you’ve been heard about especially in this
institute that you’ve been a lot heard about a lot of this mijorana fermions and quantum computing based on the mijorana fermions and mijorana particles and
so on right so I don’t have to explain or a lot of these things but in 1930s
this Ettore Majorana the Italian physicist come up with this intriguing
solution of the drug equations which is brand new equations to describe that’s the
electrode the relativistic quantum mechanical particles and that defined
the solutions that quite unique solutions such that it’s a the self
antiparticle as a solution it turns out this majorana is a formal solution
one of the foremost solution of the di-equations but it’s a weird innocence its
particle is its own antiparticle especially if you just think about in
terms of electrons it’s basically half electron it’s a half electron or half a
whole like and they always comes as kind of pair right what has been just
realizing what a rather recent a year is if you just realize this majorana
particle pairs and especially if they isolate pairs and if you just can just
kind of play around because they’re inherently entangled between these
two majorana particles and they are rather robust because it’s partially
protected one may be able to use this as the elements that realize a robust
quantum computing and out of this idea of the past five years that in the
condensed matter system there’s a massive search of the material platform
to realize this my own on a particles to be used as a quantum computing right and
there are many example that I don’t want to go through but some of the example
that a lot of commonalities a you need some sort of spin orbit coupling or just kind of something that twisting the band structures around and
important ingredient is you have to interface with a superconductor to
proximatize and realizes 1/2 or 1/2 electron type of system there are quite
a progress in this field in the condensed matter field and just
realizing one of the materials many topological insulator it turns out this
all the two-dimensional like and that’s kind of good interesting point but the
other part interesting part is and you can just also work on the non
conventional superconductors such as the P wave superconductors and quality
particles all the vortices in this the P wave soup can often consider as a
mariner particles the problem here that is there’s another easy way that we can
realize those kinds of system in the reliable way the one example I want to
share in that this a two-dimensional world there are many things that kinda
touch upon this so to demand rate system is at the last part it turns out there’s
a way that you cannot approximatize the higher states of the quantum Hall
state so quantum anomalous so here whole states and try to realize this robust be
the topologically unique particles in there
how it works basically we need to know that how to approximatize you are the
quantum or say to be the superconductor and that has been really difficult task
in conventional semiconductor system part of the reason is when you just
apply the magnetic field the magnetic field basically compete with the
superconductivity and furthermore in typical semiconductor under the magnetic field
is a very difficult to make a good omi contact to do just a quantum or
measurement with the superconductor which is usually reflected in materials
here comes that kind of interesting new directions I told you about the graphene
and this is a single layer and a high quality treatment system under the
magnetic field we know that it developed a nice quantum or effect but even
further because it is the geoweb semiconductor nevertheless once you put
down the superconductor there is a good way you can engineers a fairly highly
transparent or highly effective superconducting in there so it turns out
graphene can be really good candidate one can realize both superconductivity
proximitizations as well as the quantum hall effective indeed this
experiment so we made this very narrow the finger like the electrode onto the
graphene channel and apply the magnetic field under the magnetic field once the
Landau level forms and everything is quantized it turns out the transporter
is only carried by these edges states and simply measuring the edges state the
potentials one know that how the quantization is happening so basically
we just did very similar experiments like the typical quantum or measurement
except that one of the electrode is replaced by this niobium nitrite which
is known to be type 2 superconductors where the HST 2 is a large enough such
that when you have the quantum Hall effect
still the superconductivity can be reached preserved in our experiment we
found that that this sample actually did show that nice quantum Hall effect
down to say few Tesla nicely developed all this Venn diagram
with all nice quantum Plateau but not only that when you just
carefully measure the chemical potentials of the edge states we realize
the interesting part some part of the chemical potential image in the circuit
only to the negative and that’s very weird moment because we just applied
only the positive voltages 0 to positive voltage in these circuits
but at some point of the circuit you start to pick up negative voltages
classically or semi classically it cannot be explained and the way that it
happen is when this edge states carry the electron and hit the superconductor
across the superconductor basically this electron
into the whore out of this interesting procedure what we call the and reprocess
right and from there that the electron turn into the hole leaving the Cooper
pair behind and we start to be able to read this negative voltages in the
circuit and this fact that we just read this negative voltage tells us that
there is electron and hole conversion is happening in this system and in terms of
this more complicated world this is what you call the and reprocess or especially
cross and due process and it happens when you have the really thin superconductor where the electron is allowed to turn into the whore across this very
thin superconductor thinner than the soup canoed clearance length which means
that if you just make the discipline electrode thicker and thicker basically
we start quickly lose the signal negative signal and this length scale
tells us about how much actually the superconducting currency in this system so
it has been nicely done right I mean you may say that all right so I can see this
one but why this is a relator is so called a majorana particle up to here
this is basically description nice description of the experiment of this 80’s
and 90’s idea right what have that we have this recent tweak
is there is a new way that we can just view this problem so I’m just kind of
showing you that that device but let me just zoom in there what is the device to
look like right follow moment let’s forget about this is a superconductor
right and then quantum edge state comes in if there is no graphene
underneath indeed we just make the trench out of it that the quantum
edge state will turn around and coming like this right so across the distant
trench that we had the one quantum edge states to run this way and the other
one is opposite way this is basically counter-propagating States and then we
feel that state in between the state with the superconductor which can couple
that these are two different edge States can’t prepare any agitates through the
superconducting the superconducting interactions basically this is a basic
ingredient of the any majorana physics any majorana on a particle is that you want
create in the previous the material platform works in very similar way that
you have the counter-propagating states topologically protected right but then you
just couple them with a superconductor approximatize them at the end of the
day basically at the end of those can
approximatizations you just gapped out those kind of the counter propogating
United States at the end of the diskette are two state you actually expect is
that they localized majorana and that’s case basically more than view and this
is not my idea my saying that all this theorist actually worked it out this
type of things a couple with the counter-propagating states that this is
what is the expectation interpretation of the our cross and the reflection
therefore is basically there’s a resonance state that through that this
majorana that we have the resonance transport through that edge state which
actually turn into the cross and the reflections why this is exciting
basically this is a way that now we can start to see that how we can engineer
the majorana states and then how to manipulate them in some sense rather
than we started with the one majorana I can demonstrate it we can just have
the descend we can make there another fingers and then indeed you can ask what
if that I start can turn on the interaction between these two majorana
or maybe I can just try any layer were created and this is basically the
basic ingredient we can just present not only you can create a majorana
particle we can start to the interactions as you see here this is a
extremely preliminary data we just start to make these things that we at least
know that from the cross annual fraction they exist of the majorana and we
started to pick up the idea that we can create them we can manipulate them very
preliminary data but promising there is indeed when you just measure the
Josephson coupling between these two through the majorana we start to see
that whenever we change in this quantum Hall plateau we start to see that this
Josephson critical current changes with same steps that we expected from
this be the superconducting gap indicating that these two states are closer linked
together and hope that just give us the promise or hope that that we start we
must be able to manipulate them some more efficiently so at this point is actually
preliminary data but already I give you some flavor that just combining that
advantage we have in this treatment system such that we can create this
clean system we can also make the contact proximate idea superconductors I
give you some flavor that what kind of non-conventional device we can create it
right as you see in the title that I mentioned
also this not only electronic device but we can just go for the opto electronic
devices so let me just kind of dwell on this now optic side well I’m
not this really optics guy I know that I’m in danger that they’re in this
Institute I’m talking about optics but let me just try to do that anyway so in
my simple mind that the important object we have to discuss in when we just
discuss of the optics are the optical properties of the electronic property in
semiconductor its accidents which is basically one can create in this field
band with cap the system of semiconductor shined in the light and we know the
photon can create the electron hole pair as the excitation right well this
electron hole can Coulomb contact each other and they can form the bound
state so what that’s what we call the excitons and this bound state shows a lot
of the interesting properties like the lewd read bud excitons is a good example
but absorption properties and all a lot of the optical properties governed by
the exciton now unfortunately this exciton is a rather short-lived because it
said it’s not the ground state it’s a transient States so depends on the
system exciton lifetime can be something like even less than Pico seconds
something like the microseconds depends on the all these the mechanism how the
exciton is a recombine there right but you can ask the following questions where
external as a composite particle they must have if you’re I just streetlight
the pollen particle should have the same symmetry what is their symmetry where
it’s pair of the electron and hole which is both of the pheromone is called a
spin of the 1/2 so exciton must have the integer spin which means that we know
that this is supposed to be bozon composite bozon so idea is well if the
exciton is bozan if I pump there a lot of exciton then in principle they can
condense down into the Depot giants and condensation and form some of the
macroscopic quantum states maybe that can be called that one can use something
only the problem is exciton is a rather short lift so the first thing you have to
do to move in that direction is it just to try to create a long-lived exciton
one part we can do is you and just walk on this semiconductor
quantum well this is something that Professor Na Young Kim has been working on
for a long time right you just create this quantum well and then nice thing about
this quantum well especially in the double quantum well is you can create
the exciton in the one over those as well but also you can apply the electric
field and you can kill this quantum well and such that that now the valence band
bottom and the conduction band top of the conduction band bottom and the
valence band the top getting close each other such that exciton can be now
addressed in the even this ground state in principle all close to ground state
right and this idea of the creating this spatially indirect external indirect internal
has been around and people actually using this type of the spatially
indirect excitons or maybe put this excitons into this cavities to create
extra proton in Y or whatever way that there isn’t some demonstration you can
create this the collection of the exciton and there is a signatures that this
exciton start forms at least some collective states such as spontaneous
coherence actually start up here’s some of the example that recently
demonstrated in this semiconductor hetero structures but nevertheless
external majorana today is still another ground state such that it
can be only changing and only exists in the pumped the system non equilibrium
states so whether this you can call this extern condensation or not that’s a lot
of the debates in the system now there is a way that though one can make that
this exciton as a ground state and there’s a possibility then you can
condense them and it’s going back to again quantum well system so idea idea is
following imagine that i have the two quantum well system then just one top and
bottom each other and apply the magnetic field in there right if i just feel this
then then there’s a Landau level forms nicely Landau level formed but let’s
imagine that instead of that I make that these two landau completely full
completely empty whether you expect quantized Hall effect
appears what if that I just partially filled this landau level and partially
filled on landau level is kind of bad metal I’m the nothing special things happen
it’s just kind of poor metal right but nevertheless if you just put this
to partially fill landau level very close to each other and then I just kind of
deliberately put this the density of the each of the layer each of the layers
partially filled but it’s complementary partially fill means that if I just put
them together their full rundown level and in this particular case especially
when the lathe delay is close by then you naturally expected that electron in
the top start to see that electron in the bottom layer and of course the
exchange interactions or Pauli exclusion exclusion principle tells us they don’t
want to sit in the same space they just kind of want to experience
such that you want me to this public scrutiny exclusion principle works which
means that as the put the layers close by they it’s a self-organized such that
they just avoid each other right now if you just look at this system from the
top they look like something like this right if you just project lead out it’s
completely full landau right as if they behave like completely full
landau level so that’s basically idea behind of these things that you have the
partially filled to landau levels together and put them together and then
you just form this completely full landau level why this is a related with exciton if you just go back to this picture basically here’s the electron
directly related hole and electron directly relate the hole such that you
start to see that this exciton sub pair is actually formed in this picture right
but as a whole this is a full landau level which means that there as a whole
they actually we store back the quantum Hall effect as if these two layers
behavior one layer right and this beautiful arguments actually work it out
already in the garden arsenide and from the 90s that Jim and Jane Stein’s group at
Caltech is basically the pioneer under this view
demonstrated when you have this two quantum or layers together is that kind
of talk to each other and form this quantum mechanically quieren states can
be detected by just carefully measuring the transport for example now again you
have the two layer that you send the current in the top layer and just
measure the voltage in the bottom layer and define the resistance this is not
the director resistance this is what we call the drag resistance and they start
to see that drug resistance is quantizer and if this two layers doesn’t
talk to each other there is no reason they a connected each other but in this
picture basically because of this quantum mechanical process they are
connected and they do see that that there is a direct resistance appears and
this direct resistance is also quantized and that has been strong evidence that
even in this these two layers just kind of talking together and make this exciton turn and not only make the exciton they actually condensate it into the
quantum mechanical system showing this a quantized direct resistance so this has
been beautifully done in in past 20 years and well demonstrate it now of
course in the two dimensional electron system such as graphene boron nitride
all of these things as as long as your layer is clean you can repeat this one
into the graphene right well here is it devices so we have the two graphene
layers separate by the boron nitride with the top and bottom gates and or
each of the layers got contact as I mentioned that this can be done and then
you just kind of well the real device images like this and then when you apply
the magnetic field and measure their direct resistance and indeed we do see
that like the what in gallium arsenide we did see that drag resistance is
quantized right so quantum or effect appears in one layer to the other layers
is kind of connected although electrically they are separated out
right so two partially filled band Landau levels shows a drag quantum
quantization indicating indeed that we are seeing the condensed magnetic
exciton in the system right well on the top of that since we have this more
capability on to the tuning the density the gallium arsenide that people did
seem a high field landau level but in this case we see that those effector into the
many different Landau level for example and two half fill landau level to the
full landau level we do see again that is similar drag effect is showing that we
can also created exciton condensation in various places of the difference between
different Landau level but perhaps the very different things that we cannot see
in the gallium arsenide but clearly see in the in this particular system is
following case I told you that in the magnetic external condensation
is a very close some the excellent condensation except that as a whole
layer it is still quantum effects such that we have this the agitated we exist
right but in principle that we can make this rather than this coupling between
electron Landau level to another electron landau level in different layer
what if I just using the gate that turned this on at bottom layer of the
graphene into the whole system right so I have the electron landau over to the
whole Landau level and ask the same thing
now in this particular case it is a real hole and there is a really electron and
still it kind of this is a real extra forms right but the important idea there
is especially if you just put the same amount of these landau level
feeling so half fill landau level hald fill landau level and there are edge States or
they’re adjusting the the edge estate is counter productive directions meaning
that this becomes on exciton but via insulator without any edge States it’s
very close to the real exciton condensation we are creating if you just
can put the electron and hole togethers and make these things indeed they all
experiment I showed you we can just choose now electron in the whole side so
especially this is the filling fraction of the top layer filling friction the
bottom layer along this line basically we populate half full landau level top
and half full landau a heavier electron Landau level top have failed hole under
level in bottom right what is the feature we are seeing is all these would
be drug resistance and drag conductance we imagine goes to zero basically it
becomes really good insulator and if you just can simply make a two terminal
resistance call together it turns out your conductance actually
drop to zero when when you have the half full Landau levels together and this will is
rather weird situation so I know that this is esoteric but let me just
kind of explain that something really simple
half full landau level I told you is a bad metal right so I have the electron
half filled landau level bad metal the electron have a whore half full landau level which
is bad metal right so you have the bad metal but it’s still metal conductor what if we
you just put them together you make the parallel plates
and you just kind of measure them together what is your expectation
classically okay it’s a poor conductor another poor conductor it is probably
still poor conductor but slightly a better conductor right what I’m saying
here is if I just add a two conductor together but suddenly it becomes
insulator although it’s parallel connected and why is that because it’s an electron
and hole coupled together when I just try to send the current electron is flowing
in that way if he did strongly couple holes are flowing the same way it
canceled the current right so that’s why it becomes an insulator another way to
say that is now the carrier that here is basically electron hole pair it’s a
neutral object they cannot carry the current anymore right so this is
basically clear demonstration but something really happened when we have
that really just accident condensed excellent condensation here there will
be more of the exciting part is the other way that you can view that is now
I send the current in the bottom layer but simply that imagining the voltage in
top layer I’m sorry that I’m just as mad at the bias the voltage on to the table
to send the current on the top but if I just make this short in the bottom
layer what happen is in the bottom layer there is no voltage source but then you
start to see the current is flowing another way to say that is where exciton
is here if the exciton is a dragged by this the bias voltage in the top
basically that if the hole is flowing there electron is flowing there in other
columns the flow in there amount of current I’m getting here should be same
as amount current there indeed if you measure in this situation or cross here
that where action is condensating exactly we getting this the dragging
current is the same as a driving current it’s what we call the politic the drag
right and this is another indication indeed we have the condensed X in this
system right now we can just do a ribbon no more things I don’t want kind of do
it on this one or a bit more but you can actually carefully look at this how this
exciton condensation is happening in different magnetic field and different
temperatures and in all of them indication let me just skip this one
because it’s a but the okay so I will do it here that then you start
see that there’s interesting the phase diagram I can draw that how the extent
is conversating as a function of magnetic field and temperature I don’t
want to go into too much of detail but once we see these things we just realize
this is not only just to what we are seeing the experiment but has been
discussed in long time in exciton condensation communities it’s basically
that phase diagram you expect see when you have the excitons in electron hole of
course if the temperature becomes high and high exciton becomes unbinding and then
becomes the electron hole plasma or if you just put a more and more exciton in
there there’s so-called emal transitions happen in other word that
basically screaming between the charge carrier increases such that you start to
lose a full on binding energy that the electron hole the exciton system turn
into the electron hole liquid just kind of binding States right and all of the
system as you go down to low temperature they can condense of course in if you
have the exciton, the exciton is boson it condensed into the bose-einstein
condensation if you have the electron hole plasma that condensed as a
superconducting state BCS turns out that in the low temperature there is sure to
be some BC to the BCS crossover happens and the other is what is the expectation
indeed what we measure into the day of our system is very under Louis when we
just measured we are seeing the very similar things happening when you look
at this a gap that out of there measurement we are seeing that this
conversation gap or transition temperature shows a nice dome shapes and
that kind of eject the crossover what we expect to see in the BCM bcs so indeed
the system that we discrete out of this magnetic sense really follow
through the what we what we supposed to see in this section state however as i
mentioned that i want to talk about optic optical states but you already
start to see that is this guy is already digress of the his convenient the pace
pace of the graphene not to mention about any of the properties right and
this is nice the demonstration makes and conversation but only appears in the
strong magnetic field right how about the real excitons to of these systems we
are dealing with actually is a good system in principle that we can discuss about
that some part of the reason is in 2D semiconductor TMDC it has a decent gap
and because of his safety atomic limits when you shine the lights they
form the excitons but they’d have the very strong binding energies simply
because electron hole pair you created all of the field aligned going out of
the materials without screening and their chrome interaction becomes very strong
so it turns out experimentally you find that half electron of this half electron
both of these exciton binding energies because if they have the strong binding
energy in principle you can also attach this exciton with a charged particle
create so-called a trion or positive negative charge at trion states they can
be also stabilized and this type of the beautiful idea has been tested out
already in the three dimension electron system by the many other groups here is
at Antonio Heinz group and Xiao Dong Xu’s group they all demonstrated indeed exciton
do exist in two dimensional system they are very robust and strongly bound and
you can create a lot of different type of optical species and make the
interesting quadrants and interesting on other parties strong in a spin orbit
coupling in the system made these valleys gas spin split which means that using
the light you can also control the spin state of the particles or excitons in
the system so we know that this is a very exciting system can you actually
couple this with electronic device indeed it’s a relatively straightforward
right you just make this a transistor I just so to show that before right and
then you just put the gate as if it is fill that effective transistors rather
then you just measure transport you can just measure the optical spectrum and
see how you look like under there all these kind of different electronic
conditions so here it’s a good example here is that we have a transistor made
of the tungsten di cell and monolayers we have top and bottom gates
as you see the Y and the source and drain and we can just act or to study
about their transport properties but we can also study their optical properties
as a function of a gate using the gate we can control the electric field edge
where as a charge density and this particular diagram I am showing you
there is a photo luminescent data as a function of the density we can
change in the system if you just apply the same polarity of the voltage onto
the gate we can change the density without applying the electric field or
we just apply the different polarity on top and bottom gate we can change the
electric field without changing the density and what you see here is their
spectrum the bright the red is basically peak and blue is a deep and there are
the peaks appears in the optical spectrum we can assign following to the
previous work this is exciton this is a negatively charged trion on this is
positively charged trions as a function of the density extreme quickly dies off
outside of the gap region their energy has shifted as you change the density
because they are screening property got changes so we start to see that a
spectrum got changes but important part is this exciton we create or try on we
create in two dimensional system as a function of the electric field in the
vertical directions there instead intensity got modulated but what you see
in that the energy position is that just constant right and this tells us
whatever dipole moment of this the species accidents or Trion you create if
this type of moment is they should be in plane such that it’s orthogonal to the
electric field we applied in the vertical direction another way to say
that is truly two-dimensional object we can create optically created and
potentially we can also electrically manipulate it right
while this the two-dimensional system we create is a completely the
two-dimensional electron default this system one can create this the rather
semi three-dimensional system by hetero structuring I told you that we can
create the PN junctions by just kind changing different type of materials in
this particular case I’m only thyself I sell and I tungsten Dyson and I put them
together it’s a type 2 band alignment we can create this the structures I showed
you before that this behave like PN diode we can create the
photoluminescence and we can get we can get the electro luminescence and photo
currents out of it we actually put a little bit more efforts in here we can
put complete control top and bottom gates each of the electrode
each of the layer and if each of the electro need to be gated to make the
homie contact but in the end of the day we have the device by the way this is
kind of heroic efforts of the couple of students and making this type of device
takes us sometimes to build in the working device but nevertheless once you
make the device you get beautiful optical spectrum here I’m showing you
out again photo luminescent curve so this one is that when we have the
tungsten disseminate along this is where the molten dicelonide and I are on a low
temperature the peak width is about Mille level 4 almost a radiative a
lifetime they’re limited but the interesting part is interlay exciton when
you have that this part of this the transitions we do see that the peak
appears in heterostructure area indicating that indeed now we have this
intellects and forms and that’s important kind of beginning point right
I told you that interlay exciton can be long-lived in principle because now
electron holes are separated right and over the on the top of that because of
this band alignment is not the three-dimensional structures by just
again we can control this band alignment which means that this energy scale got
changed indeed that’s a case when we actually show this intellect some peak
changes with the gate voltages again as you change the density we see the
intellects and the energy got slowly slightly changes but outside the gap we
quickly lose the intelligence so intellection only appears when thermal
level is within the gap but important part is unlike this is a two dimensional
system in disappeared junctions in the atomic Alateen PN Junction as you change
the electric field by controlling the gate
inter exciton and energy is linearly changing just kinda linear structure
meaning that dipole moment is now out of plane and just looking at this
the magnitude of linear shift we know that this is a completely interlay excitons
we just created now I told you interlayer exciton can be long-lived indeed if
you just look at the external lifetime external lifetime is a reasonably long
it’s something like 200 nanoseconds the half micron half microseconds more
importantly energy you just change Gate voltage as you just apply the
electric field along the exciton direction such that you start to pull this excitons
away you start to get this electron and whole wave functions overlap and less
and less and you expect a lifetime becomes a longer and longer and that’s
precisely what you see as you apply the electric field
lifetime becomes longer because you pull this exciton ends up right and such a long
leave the exciton is already forcing gradient we need to create the exciton
condensation right the first thing that you start to see is this because exciton is a
long leave that when you just create the exciton they can start can diffuse out the
sample rather fast right in this 10 micron size of sample you just kind of
shine the laser there about this spot and then you start to see that the
exciton can be picked up or across the sample even further you can actually
look at this how the exciton get diffuse out by just just posting this exciton and
you can get this exciton actually diffuse out to rather fast and just the time
dependence this extra measurement we can start to get this in the numbers it
turns out diffusion constant of exciton is a three centimeter square wall centromere
seconds which actually correspond this the the mobility of the carrier do we
measure in the system so we know that it works more important part is that this
in homogeneity that tells us the exciton we create can be also trapped in some
part of sample due to the imaginary and the density inside of trap can be
controlled by descending the current through the desist one of the layer so
this starts tells us that in principle using the current in one of the layer we
may be able to drive this exciton in the system this is very analogous to the
magnetics and I showed you right when I send the currently one of the layer that
I said get the perfect drags in to the other current
unfortunately we are not in yet get that new range of the measure
but clearly the optical spectrum start tells us that sending the current in one
of the layer apparently affected systems in the way
that excellent distribution is something’s it’s a very kind of
important part are we in the exciton condensation regime I think let me just
skip this one I think we probably not we don’t have
the really good evidence yet but at least we are getting it there how do we
know if you just pumped in the exciton with the high density the exciton and it’s
that blue shift and that’s because exciton and exciton and interactions and this blue shift
actually give us a sense that what is exciton density we can create in this
system without heating the system it turns out exciton and ends to be creating
about 10 to the 10 10 to the 11th they’re centimeters we don’t know that
yet that what is the density that exciton can be created but if we just rely on
the mean field type of the theory calculations already here okay so this
is a BKT line where the exciton conducts BC condensation is happening so we is
about here right I did it and then you see the temperature is something like
this we measure this is a 4 Kelvin so in terms of this density we are not really
far from this mean field of the line of the exciton condensation we hope that
once we just create a little more careful experiment especially direct
type over the experiment we should be able to really pick up this quantitative
quantitative analysis of this exciton condensation features so this is another
example I think at this point it is just the demonstrations of that at least we
start to see the very first step that what we are looking for that just can be
on the convention of the electron device such as the photodiode
or the photovoltaic device and here is another example that once you create the
so kinda exciton condensation in principle you can create exciton and
condensation even a really elaborate temperature as long as you you just make
the long-lived excitons in the in this system and that’s kind of another hole
that maybe it there is a way that we can create this quantum electrical optical devices
based on this type of and there was hetero structures
technology got evolved not only you just kind of create the excitons but you can
create this excitons and they can one can mode they modulate the exciton and the various
positions and some things I don’t have the real time is going through and just
kind of create this at the some of the localized excitons and trials by just
the look at the engineering the important part is important message that
I want to deliver in some sense as a summary is this the new opportunity that
we are just received just arise from this availability of the various
different type of materials combined with this the mesoscopic experimental
technique subtle ways to give us some of the exciting the the the new quantum
physics especially new many-body quantum physics startup here in this type of
system now the next level question is is there a way that we can tame this type
of new physical phenomena into the kind of device going beyond of the CMOS type
of applications so that’s basically next many years of the answer that we have to
work on finally I should probably thank my group first that they just kind of
make the oldest the various part of the presentation is possible but also not
only is within my group there are strong collaborations both the superconducting
inside optic side and theory side that most of the my local collaborators as
well as they might collaborate in the Japan who provide high quality of this
boron nitride crystal thank you very much

2 Replies to “Philip Kim – Materials in 2-dimension and beyond: platform for novel electronics and optoelectronics

  1. Are there links to the slides used? While I appreciate the dynamic camera angle switching, it was rather hard to clearly see and focus what was on the slides. Thank you as always for sharing and uploading quality content.

  2. Yet another series of seemingly interesting presentations totally ruined by an "oscar winning" cameraman. By far the most interesting part which is the slides on the screen, is barely distinguishable. However the filmmaker is "creative" ! Time to time the view swiches from the lecturer's mimics to the conference hall, showing just the backs of some anonymous people sitting in there. This way, the IQS videos lose >90 % of their potential value. What a waste of resources because of only one "movie maker genius" 🙁

Leave a Reply

Your email address will not be published. Required fields are marked *