You know you

can you explain to me.

All right.

My name's Holly thank you.

I've been working the epa tax of graphene
lab in the physics department for

a little over a year now I've been working
under the supervision of Dr Edward Conrad

and this isolated terminations
of stock motors the face.

Graphene was a collaborative
project between some of his

grad students or myself.

So to begin about quickly thank
my advisors and the Ph D.

in undergrad students who
help me model and program for

funding the project to begin with and
also Georgia Tech for the facilities.

To give you a brief overview.

When I want to go over I'll
start by talking about

the current status of
silicon technology and

how we've kind of reached a threshold as
to how far we can push what we have and

that's led to a resurgence in materials
research over the past decade.

Graphene has come up as
a potential candidate for

post tonics it's basically a plane of
carbon atoms or ancient hexagon lattice.

And what makes graphing so special is that
because of its electronical properties

electrons flow through graphing as
if they have no effective mass.

And this has a lot of potential for
future electronic applications and

bio applications as well.

Our lab particularly works
on epitaphs of graphene

which is graphing grown from a crystal.

So when we grow graph in the manner which
we do we find that we grow several layers

and when we test these
layers electronically we see

that they act like
individual graphene layers.

So for some reason the layers
are not interacting which is

very good we see that even though
we have several layers there.

They're all displaying the individual
properties of graphene instead

of graphite like you'd expect.

And the purpose of our study was
to figure out why our planes

are layers graphene are interacting
with each other and

we think it's due to some
built in stacking order that.

In layers of graphene
a different orientations

such that they don't interact.

So can the brought us a long way from the
first transistor made in Bell Labs in one

hundred forty seven to the integrated
circuits we had today more plotted out

the efficiency and affordability
of computers as a function of time

our computers have gotten cheaper and
faster exponentially over the years but

we know that we're getting we're going
to hit a limit and that limit is that.

When you get silicon transistors too
small into the nano size they begin to

allow the tunnel through them.

So we get transistors that leak
current and that lead charge.

And because of this.

Many of the transistors that we use in our
computers talk about if you get a Hertz

and if we want to continue this trend of
faster and faster computers and lower and

lower prices.

We're going to need to find some kind
of new material solution of this and

this is where graphing comes in and.

As I said before it's a two
dimensional array of carbon atoms

in a honeycomb lattice and two dimensional
systems are a little new to us but

we see that we've seen graphing before.

If you call it up.

It's a bucky ball if you roll it up.

It's a carbon nanotube when
you stack it it's graphite So

it's one of the most studied
substances we know of but

we never have been able to isolate it and
just a sheet before.

And we do that we find it has
a very special characteristics.

The carbon atoms are bonded together was
thinking of bonds and those are the some

of the strongest bonds in nature and
that makes every plane very very strong.

It also makes these rules that extend out
into the plane and that creates clouds for

electrons to travel through.

What's very unique about
graphene is its band structure.

If you see energy plotted out
as a function of momentum we

see that it's linear.

And this is something we've
only seen before with light.

This is something very very special our.

Electrons travel through as if they have
no mass almost as a photon would and

this gives a very high conductivity
very low resistance a very high thermal

conductivity and this is the big hallmark.

Graphing is a linear band structure.

As I said are all our works with
every tactile graphing in particular.

So we start with a silicon carbide
chip which is one still cannot and

pervert carbon atom.

On this face.

I should begin by saying that
silicon carbide is a polar Crystal

we have one face that ends in carbon
atoms and one face that ends and

silicon that I'm so when we say See
face graphene we mean graphing grown on

the face that ends in carbon atoms.

No we find that when we heat them up
in a radio frequency induction furnace.

The silicon begin to leave the surface and
what's left

behind is a bunch of carbon atoms that
will will reform bonds to become graphene.

And over time we have
several layers growing.

And I want to give you very qualitative
description of what's going on.

These are atomic force
microscope be images.

So the colors aren't real.

They're just to give you a sense of the
height how high each step is on the left.

We have a silicon carbide crystal you
can see the steps there very smooth.

This is before we've done anything to it
and those are actually actual steps in

the crystal itself by layer steps as
we heat it up the steps are eaten away.

If you will as the silicon leaves and
we see steps retreating and

changing shape and
leaving behind carbon in its wake.

And as you heated up.

The carbon graphene film just
drapes over the crystal and so

what you see in the other images a graphic
ties sample has a film of graphene over it

in the background you can see these steps
and how they've changed from this nice

straight step edges to these curved and
kind of jagged step edges and

also you can see that the film itself is
transparent to atomic force microscope.

But the white streaks are pleats or
bunches in the graphene film.

And that's how we know it's there.

That's how we know there's
a blanket over the entire surface.

And as I said before when when we
see these multi-layered films and

test them they display the same
electronic properties as

individual graphing layers would.

This leads us to a little bit
of a quandary because in nature.

Stacked graphene means graphite and
it turns out that nature stacks graph in

the one way it doesn't work that is every
layer stacked with sixty degree rotate

rotation with respect the layer before and
this makes Adam sit on top

of each other so they're there interfering
with each other's P. orbitals.

And we have layers that
are slightly bonded to each other

which destroys destroys all these great
electrical properties we were looking for

in graphene to begin with and this is our
main question why are our multi-layered

graphene films not like graphite which is
in essence a multi-layered graphene film.

What makes them different and
behave differently.

And we went about solving this is by
looking at three primary sources of data

we looked at lower as you
looked under fraction and

the result photo emissions
spectroscopy and surface X.

ray diffraction So we took everything we
knew about these films and combined them.

And tried to make models for

how are these layers ordered such that
we're getting these kind of properties.

And then we went back and
tried to fit our models to X. ray.

Data.

So begin with low energy
electron diffraction we basically

shoot a beam of low energy
electrons at our sample and

they diffracted off and they were
collected a phosphorescent screen.

Now wherever the way forms constructively
interfere we get bright spots

on our screen.

This is the sample.

This is the lead system we have at
Georgia Tech you load the sample.

In here and it eventually gets
pushed into this main chamber.

So you can see the main chambers of
the window is kept in very low vacuum or

electron gun is in this canister
it shoots down at the sample and

the deflected beams are then.

Scattered back up and
the phosphorescent screens at the top in

order to see these things you actually
have to climb onto a step ladder and

look down into the viewing
window at the top.

But when you do this if you have a very

highly ordered substance
something with very regular.

Gladys the the spots that you see

should be indicative of the geometry and
the orientation of your sample.

So we do this we get the following
results are still can face

graphene we know that it's ab stacked and
that's why you only see bright spots at

sixty degrees you see a hexagon just like
your graphing lattice and so we know that

we only have on the silicon face graphing
stacked with sixty degree rotations.

The carbon face which of the fills we
deal with so something very particular

you see these planes but
you also see this ring of other rotations.

So not only do we have planes but

we have this variety of other
angles that occur between.

Our thoughts are that
somehow the interspersed in

these layers are stacked such that the
layers aren't coming into contact as much.

So.

To get a little better of
an idea what's going on.

We looked at angle result
photo missions spectroscopy.

And this basically allows us
to take a cut across that cone

you can imagine if we
were to cut across that.

Band structure where the cones
converge we should get an X..

And that's what we see when we
do some of our thicker films

we get several cones from.

The different layers.

Every now and then however we see
this band structure open up and

so if it looks more curved it's not an X.
anymore.

And this is where to be plain have
come into contact with each other.

This is why ab stacking
is detrimental to our

films because that's no longer
graphene like that has a band gap and

that's that's not the electric
properties we were looking for.

Unfortunately our pets can only see
the first few layers in order to

penetrate farther you have to go to
something a little higher energy and

that's where we get surface X.
ray diffraction.

So if you remember that that A and

B. point I showed you
before on The Lead image.

What actually diffraction
it allows us to do.

It's almost as if you're looking down
on a rod that that point is actually

the top of the rod and
then straight surface.

Diffraction data allows you to
turn that rod on its side and

look at the intensity at
different spots on that rod.

So you have X.
rays in your sweeping up this rod.

At different angles.

And you get the following profile curve.

As you as you sweep up an angle.

We get three three main intensity peaks.

And because we've picked a spot where
only a and B. plan to contribute.

We can figure out how to remake this this
graph just by figuring out how the A and

B. planes are ordered in the sample.

If we can figure out where the A and B.

points are as a function of height
in our sample we can remake this.

And so we sat down and
thought of three different models.

But here the the bold lines are A and B.

planes and
the dash lines are the rotated planes.

Our first model was random in which we
start with a completely A.B. stack plane.

So everything is ab stacked and

we throw in these rotated planes
these dashed lines at random.

And you can see wherever in aid
of the plane are left in contact.

You get ab stacking the second model we
considered was caused by ordered which

is where we expect the rotation layers
are happening a very regular intervals.

We should see a rotating layer every other
or every two or every three layers and

we did allow for some statistical snafu.

So sometimes you can get ab stacking but
generally you have very very ordered film.

And then interrupt it is
very similar to random.

But interrupted we say that after
a rotated layer the film can switch

its stacking So instead of a B. a B.
at the rotation it switches to be A.B.A..

Just a little variation to try
to cover all of our bases.

When we do the Fitz I
program these in matlab.

And when we did the fit's
we got the following.

Brenda model that we knew wasn't
going to work pretty quickly.

The backgrounds too high and
we also weren't able to fit the correct.

With the sample that we did this this X.

ray diffraction off of was only thirty
three layers that and our peak with

in order to fit for the random model
we had to go down to eighteen layers.

So we're pretty sure that's not the case
and it also gives us forty percent maybe

stacking which is a large chunk of our
films and we don't think this is the case.

This doesn't match with the electric
measurements we've done.

On the other models because they
were model fits much better.

It was able to lower that background and

also match the peak with at
the correct number of layers and

it gives us nineteen percent A.B. stack
which which is much better number for

us and we also think it matches more with
the electrical properties that we see.

Is interesting to note that even though
the interrupted model is based off of

a random model.

It was able to fit
the correct peak with but

we still weren't able to get
the background down in that case.

So what we're able to conclude
from this is that our graphing

films act like graphene films
because there are these

rotated layers interspersed
between the A and B. layers.

So that the layers can interact.

There's they're separated too much.

And because the quasi order model
fits best we think that when grafting

grows from our crystals.

There's some mechanism at the surface
that forces these orientations in a very

ordered manner that there's some kind
of built in anti graphite mechanism.

If you will.

And the third thing that is that most of
our films are graphene like they're not

ab stacked and the Ab stocks are the
exceptions and so now that we know this.

We can go on and
say that we bake our chips and

can immediately make devices out of them.

We don't have to worry about
transferring or worry about.

Well is this film more graphene or
more graphite.

And so our lab is always
working on doing things better.

And one of the one of the ways we do
that is how do our graphics comes work.

We're always working on characterisation
of films under different can.

How do our graphic films behave

the the other thing we're always working
on is figuring out how to grow better.

So how do we how do we make
their films higher quality

how to make it more uniform
over a larger area.

And also device creation as
I mentioned now we can just

take a sample out of the furnace and
immediately pattern it.

Mike sprinkle from our group actually
has worked on a ten thousand

transistor chip where he took a sample and
has used

to pattern ten thousand transistors on
a single out of a single graphing sample.

And there are other groups in
Georgia Tech that use our samples and

work on thermal biosensors we think that
graphing films have a high potential for

being able to detect very low
amounts of glucose and proteins and

dangerous gases so
that's another real world application and

any questions you

interrupted you know you have your brain.

Interrupted was plotted with
the course I ordered as the red line

for us this morning was really

like this.

I'm not sure that they've done that
they've done different S.T.M.'s you know

studies but I guess as you said
a lot of these places are far away.

And so I can definitely ask about that.

We're going to how do we
control how many layers grow or

is there a limit on the sea face and this
is a problem we started with very early on

the C face you get anywhere from five to
one hundred layers depending on how long

how long you cook your chips but
we see that

as you get to very very high layers the
graphing grows kind of splotchy that is

some places will have you know seventy
five layers and others only have seventy.

So when you get very very thick it's
hard to control the uniformity.

Which is why they're trying to grow
thinner films now now you can get it to

three or four layers and
they were trying to make model layer and

by layer is what they're really
working on it would grow.

Not that we've seen but
we do see it specially there's

lots of growth it's hard to get electrical
measurements because you're not actually

measuring an entire film you're
measuring chunks of film.

So it's a little difficult to tell but
we can't see that in the point that

there's a threshold where it starts
to be graphite all the time.