Thank you.

Well I want to thank the cloakroom
committee for inviting me and my host.

And did you hear me.

It's not here.

It's not on OK.

Which one.

So in any case I'll be
presenting some of the work that

may many collaborator
students postdocs and

others have been working on for
many years now and.

Starting more than a decade ago now and
from a from a field that was

completely unknown to anybody in fact
I had to explain what graphene was and

I was told by many that it was really
a waste of time to work on this so

many years ago and then finally now
it's become almost a household word.

So a little bit of luck
a little bit of insight and

I explain a little bit how we came to
this whole idea of doing graphing based

electronics which I can probably say
actually originated in this building here.

And so many years ago.

But in any case the story has slowly
developed it originally I thought

we could produce something significant
in graphene within two or three years or

so and that was a decade ago and
the course that's developed enormous but

you'll see a little bit with the states
that are the state of the art is

right now and perhaps also why we stay
very enthusiastic about this material

even though I must say that the hype and
the smoke is so

thick right now it's sometimes difficult
to figure out what's true and what's not.

Fine so this is a little bit of
an overview of our original and

collaborators and there's lots more but
we have very strong connections with.

With France by the way.

Mainly due to Clare and
what we in fact at Conrad is right

now it's all a making some measurements
at the synchrotron right there.

We have.
A very strong connection with the C.N.N.'s

is seeing a hit with the take
Sukey as you all and also a with

well here in the AND
thousand who was here a little while

ago in context of this collaboration
doing spintronic with graphene.

So what is it all about.

Well actually something also happened
essentially in this building many years

ago was a very simple observation of
unusual properties of carbon nanotubes

which was actually the precursor to all
this graphene stuff and what we actually

found was that carbon nanotubes had a very
unusual behavior in their resistance and

that was if you take a long can
a carbon nanotube or a short one or

change its length the conductance is
constant and the conductance is turns

out to be close to as you know this number
one here which is one East squared over H.

or actually two East H. which is a quantum
of conduct and soups and want to do that.

The quantum of conductance fundamental
units here and it turns out to be

if you hurt this to a resistance that
turns out to be about certain kill on us.

And that's very striking because the
quantum of conductance is key to a lot of

Mesa scopic transport resistors this
this is the unit of resistance on

the quantum level might be able to explain
a little bit where this comes from but

if you see this number pop out of an
experiment you know that what's happening

is the electrons are travelling
from one side of the river

at the nanotube to the other without any
resistance actually and all the resistance

you see cause it has thirteen kill
ohms to sitting up the contacts and

this contact is not FICA It's a funny
kind of resistance it's almost.

Due to the geometry or.

Going from three dimensional to a one
dimensional system actually its intrinsic

in the Quantum mechanics of transport
more than actually caused by scattering.

Ansin purities So
again there is once you get the electrons

in this carbon nanotubes they
go without any resistance and

that's of course very very exciting
to see and that's how we thought.

Wow Is it possible to harness these
properties in another material and

carbon nanotubes because none of the tubes
could make wonderful Lek Tronics but

they have a pretty bad shape there.

I don't I don't have a picture of them but

they're basically nano scopic tubes
of graphite and they're all round and

like little like little pipes so
how do you connect them together.

How do you solve them together.

So the idea came quickly that perhaps we
could do something different with them and

I'll come back the previous
light in the second.

Maybe if we cut them into ribbons then
we could there capture the properties of

carbon nanotubes and indeed this turns out
to be true if you take a graphene ribbon

you can quickly calculate and this is
graphene these are the hexagons of carbon

and you could take a graphing sheet and
cut them to such a ribbon and

then you could quickly calculate that it's
electronic structure of such a ribbon is

very closely related to that of carbon
nanotube So the big insight here was

basically the way that whatever may carbon
nanotubes have those ballistic properties.

Is it possible that we can mimic that with
an other geometry perhaps a graphing cheat

that we would cut into
ribbons like this and

could they have the same property and
if so wonderful.

If not why not to walk proceed from there.

So again that was a long time ago.

Actually this paper about
cutting a graph into ribbons

was already earlier was done by just
a house of the Early in the mid ninety's.

She thought about this but

the this was simply a calculation very
easy to do calculations by the way.

So so now the motivation for
all of this of course.

Right now we always want to have
stuff have grand schemes and

do great wonderful things.

So we thought well one thing we
can probably do with graphite is

graphene in the this case make a new.

Kind of electronics of graphene based
electronics so we looked at this very

famous plot by Gordon Moore who says Well
in about now actually supposed to do but

graphing silicon is going to run out of
steam and cannot continue to have its

constant progression that has been doing
since the one nine hundred seventy S.

toll now it's going to flatten out and
and there's going to be limitations and

speed size and heat and so the industry
has electronics industry has for

a long time been looking for alternatives
to silicon and other materials and

there's a whole host of insulin
are now carbon nanotubes or

it came onto the scene and then graphene
and that's what the story is about.

So our work actually led to some work or
earlier work on the ninth in two

thousand and one and following down the by
two thousand and three we have enough.

Experimental evidence to prove that this
whole scheme works and this is quite

interesting because we decided the
graphene not electronics could exist and

this is a very first patent on graphene
electronics actually before anything else.

This is what we did with Phil and
myself and Claire and

a decided that some structures
could be very interesting.

Now what I'm going to point out is that
the structures have to interest things

about them interesting thing is
first we were considering normal so

field effect transistor so you would
take a graph in the ribbon then you put

electric fuel on it and you could change
the flow of electrons going through it.

That would be one way of doing it but

we quickly realized that there's much
more to it that the good luck comes in.

Graphene could be coherent
if there ballistic.

Then they could show the wave
properties of electrons and

perhaps even at room temperature.

Now that is a very exciting story if
we can harness a wave properties of

electrons in a real material we can do
sayings that you cannot do with standard

silicon because that's a diffusive
conductor I can back to it.

So for those reasons these other
structures came up this mock center switch

which looks like a Michelson
interferometer and

it really is it's considered to
be a Michelson interferometer.

Or a funny thing with a junction or
something with many junctions but

basically playing with nontraditional
electronics that what's going to be a key

part of what we were going to be doing.

And of course none of us important here.

So the whole point about can we make now
a structure and play with this graphene.

In the meantime life.

The science went on and
the Scotch tape method for

peeling graphene and
putting it on stuff was found and

we did it a different way and
I'll have a lot to talk about but

new methods have been developed to
try to create graphene we do this.

It's called up attacks of graphene I'll
introduce you to that in a few minutes but

to just so you know because you've
probably all heard about the Scotch tape

method of peeling graphite and putting the
flakes down on the substrates and measure

make measurements on it looked exactly
the same reasons influencing electronics.

So this was a very interesting.

It's a very simple way to do it.

We don't do it this way but
you can try to do that and

all kinds of dream situations were
thought of quantum dots with Graph and

cutting in certain ways high frequency
transistors and other things like that but

I'm going to go into a rather technical
slide here a big surprise came out

if you take graphene and this is jumping a
little bit ahead I hope we can get you on

there and what you can do is you make a
transistor So what's a transistor graphene

a transistor graphene a simply a graphene
strip with electric field on top of it

if you put the electric field on graphene
you can change electronic properties but

what people soon found out that
the graphene made graphene ribbons were so

disordered that you
actually can't use them.

If you cut them.

It's too disorder to use and so
this is a little bit was a major problem.

I'll get back to this in a second.

And why happens well this is what a lot
of people might expect that if you make

a graphene ribbon and
you try to flow electrons through it.

Then what happens the electrons
rather than flowing through it.

Ballistically as we had hoped but
they really do if they hop.

Around from a because there's so much
edges and stuff electronic and trapped and

then rather than flowing
in in a beautiful ballistic

flow like they do in carbon nanotubes they
scattered off of all kinds of edges and

roughness and
we do get some kind of activity out of it.

It's something called the mobility
gap that's not so important.

But the basically the point about it
is if you take a graphene sheet and

you take it to say a lift for
another form of lithography and

you try to make up something that
mimics a carbon nanotube you will fail

because ultimately you make a rather rough
structure that doesn't do very much.

This is all for background.

Now.

Graphene electronics is developing
in Europe very quickly and

people have this idea of what you can do
with this graphene they kind of thought

well you know we can make a lot
of new interesting materials I

assume you have a little bit of background
in graphene since it's been here for

a little while but an ideas work perhaps
you can make a back touch screen

things actually use the graphing
conducting properties to make touch

screens some papers all LEDs and other
things not going to say much about that

other than saying that there's a whole
field that's trying to use graphene for

its great conductive properties.

Now if you know don't know
anything else about graphene you

probably have heard that graphene
should be a quite a good conductor.

So considering a conducting sheet or maybe
you can do something with it and this is

what people plan to do from nineteen
twenty twenty thirteen to about twenty

twenty three and but the trouble with that
by the way and again we're not doing this.

This is but the trouble with that is
there's a lot of already going on in

the so-called flexible
electronics business.

So if you hear somebody say
you know graphene is going

to be a new flexible conductor you can
make great transistors out of it and

you can do all kinds of
fancy stuff with it.

You can just go to the Internet and
look at a look things up

like a transparent conductors or
flexible like Tronics and you.

Quickly rant run into the so-called
plastic or organic electronics story and

you find out that all this
stuff already exists for

example flexible displays made out of
organic electronics was a ready being made

by Sony Samsung I just heard
has a huge production factory.

So the problem with this is that graphene
at the moment even though it's considered

to be a great wonder material for
electronics is not going to really find

an easy Nish or an easy market
in the in the so-called flexible

electronics world because we can't
really make a good graphing transistor.

First of all.

And second of all because
the stuff already exists.

So this is kind of the the ugly
side of graphene you know if

it went with an enormous explosion it's
going to revolutionize electronics and

the you know you could could printed and
all that but

it quickly stalled out all the fact
it's very difficult to make good working

graphene transistors and
so that's a problem.

So yeah well that's again just
to show you how far the other

electronics school and now I'm going to
talk about this other stuff the stuff

that we have been working on since
actually the beginning days.

And that's called up attacks of graphene
and potential graphene what we do is we

start with silicon carbide and we have
a silicon carbide wafer that we simply put

in in at high temperature in vacuum it's
a simple as this it's a little furnace and

you can see several those in our
laboratories you heat the stuff up and

silicon carbide is literally that silicon
and carbon and if you heat it up.

What happens is the silicon evaporates and
when the silicon evaporates the carbon

This left and when the carbon is left it
forms graphene Now that was known for

ever so this is an alternative
way doesn't involve.

Well that involves simply you taking a
silicon carbide Crystal and heating it up.

That's all you really got to do and

this was discovered actually in
the one nine hundred seventy S..

This method and

that it makes for Afeni was known
since those days so but we thought you

know it's a very interesting process to
have graphene growing on silicon carbide

crystals because silicon carbide actually
is a very important electronic material.

It's used in all your L.E.D.S.

and all that it's so
it's a well known and established.

Technology but the main reason that we're
using silicon carbide as a substrate

is because we want to copy what silicon
this doing and whether silicon doing well

you know in the silicon industry you start
with a single crystal of silicon car of

silicon which is this thing here
a single crystal and you slice it and

you start manufacture and you have to
work one where ski suits to do it.

Apparently or bunny suits as they're
called but the fact is this is

a technology is very highly developed but
why do you single crystals.

Well there are several reasons.

First of all the properties of a single
Christe are electronically better but

the other is you have a perfectly well
defined substrate No one big key about.

Graphene and that's going to become
clear and clear it's going to

show its potential mainly in
nanostructures and nanostructures means

you have to make things on the nano scale
and making things on the nano scale

means you have to have a surface at
least well defined on the nano scale.

So if you want have an analogy like that.

If you want to be a fine
paper like Rembrandt like and

he starts with a burlap sack you know he's
not going to make a night watch on that

were married adventure and Mona Lisa for
that matter you really need a material

a substrate that can stand that kind
of precision and that was very clear.

So the four that we work with
single crystals of silicon carbide

certainly expensive but
that's a reasonable starting point.

If you finally want to makes this
stuff work on the nano scale.

So again because this graphene layers that
spontaneously form all these crystals and

then after that we can do
all our processing them.

But now the of the thing
are already told you are indicated

that one of the biggest problem with
graphene be it on silicon carbide or

you know using scotch tape and
putting it on the surface is once

you start cutting it using standard
techniques like lithography and

whatever other techniques you might use in
A.F.M. needle or something like a cut it.

You end up with the rough edges.

So how can we get around
making rough edges.

So that was really the big question
how can I make nanostructures

with edges there are not rough and
well defined.

Well I'm going to paraphrase
a little bit how that works.

The idea was if you take your two year
silicon carbide crystal which later we

heat up and grow graphene on but
just put a little stuff on there.

What's going to happen.

We happen to know that the silicon.

Primarily or easily evaporates from
the step edges and so it's reasonable

to assume that the graphene will grow
rapidly on the steps and they do and

there's various other reasons but
this is where the growth is and so

what happens if you simply cut a step
by some kind of etching technique so

whatever way some twenty nine of
meters deep or something like that.

So this is a step you silicon carbide and
you heat it up.

What happens and
you heat it up really hot.

This is fifteen hundred degrees so that's
beyond white hot really and then what you

happens is this with this square step you
put in there turns into a nice facet.

These are crystal facets.

It has actually indices that
are given over here and

what happens next is
the graphene layer grows there.

Now what's very key to everything
we do now is we've built carbon

Graffy nanostructures starting from
single crystal silicon carbide patterning

the surface actually made in a three
dimensional pattern heating it up and

after that the graphene grows on the edges
and it's beautifully in the old.

So if you get that part.

You'll see how we do stuff different.

We are working with nature to actually
assemble these perfect grafting struck.

Yes we're saying telling nature.

We're heating the stuff up to some
blazing hot temperature in back you

figure it out and it's going to find an
equilibrium condition and that equilibrium

condition turns out to be crystal faces
because you know all roughnesses that

appeared because you're you didn't cut it
perfectly straight or whatever vanish.

You get facets.

And one of those facets you get
this graphene layer growing.

So this was a guess that this
might work and it did work.

OK And so the rest of the story is going
to be how these ribbons that you grow

edge is simply cutting a step on
silicon carbide and heating it up and

evaporating the silicon from that and
having this graphene layer growing there

starts having properties are very similar
to those of carbon nanotubes and so

that took us to many years to
figure out but we finally did and

I'll show you now how
that story progresses and

there's going to be a little bit of
other other bits and stories about.

Well the country cartoon figure of
what happens is here now here you see

the silicon carbide and here is graphene
she draped over it and you see the edges

are sort of ducking into the silicon
carbide this is a cross-sectional T.M.

pitcher actually not made of our devices
but by somebody else some years ago

where well there's more than one
layer it doesn't really matter.

But you can see how this graphene layer
grows over the step and then terminates

into the silicon carbide we are able to
control this growth in such a way that one

maybe two layers grow at a somebody tells
I really don't want to talk about it but

so this is the graphene on the step and
just fully in the old and you see this.

Well you look at here you see
the silicon the carbon atoms in this

in this transmission electron
microscope picture and

you see the graph in sheets side view of
those so this is what we're able to do.

Actually this was done by a very
very good scientist nor him out.

So who's an expert in the so called
cross-sectional T.M. tech technique.

Well then we said well that's very nice.

So we can make these steps a.

We can make its ribbons and
can we make something out of it so.

Almost for fun but actually more for
proof of principle.

For our former student of ours Mike
sprinkle actually made ten thousand

transistors taking such ribbons and
putting the various electrodes in place

and it showed to be quite easy to do so
now we can make many many transistors of

these of these of these graphing
ribbons on a small chip

six by four millimeters a chip ten
thousand transistors that was kind of.

And actually we could have put millions or
billions on them.

The trouble is you want to
be able to contact them.

So we have them made big enough so that we
could go with with with a microscope and

know where they are locate them put the
probes in place and then measure them and

that's why it's ten thousand and
ten million.

But you can make as many as you want the
transistors are not working very good but

it shows that the principle of
mass producing many many grapheme

transistors or structures if you will
using this so-called structured cross

technique really works and
that was published some years ago.

Now you can do more than that you can take
little little rings for example and so

here you have a silicon carbide again and

you you make little pillars and
you heat them up.

This was done by who won who graduated
last year and that when you do

the same thing you heat it up the graphene
will coat just to add the little rings

around here see little rings of graphene
and you can make huge arrays of those and

those are interesting and for for for
various applications of this is what.

Also we can do.

So a whole variety of structures you can
make a ring circles lines what have you.

And so it all involves patterning
the silicon carbide heating it up and

having the structure having
nature take care of itself.

So one thing I want to talk about is
a work that's done by Ed Conrad and

so what now we want to look at with
the electronic structure of such a.

Bracken ribbons are and what you do
is you make a race of these this is

a time of course my cross the pitcher.

This is what it looks like if you look
from the top few These are a whole whole

bunch of ribbons in a row and

from that you can actually probe the
electronic structure and we do that with

the synchrotron light what at this doing
Conrad's doing right at this moment.

And here you can map out
the momentum of the electrons in

the graphene on such a ribbon versus
the it's energy and this is so

called dispersion relationship and
I don't want to say much more about that

except you can see that the energy if this
is momentum which it is and that's energy.

You see momentum and
energy are linearly related and

what does that mean that means that energy
is Linear A leader related to momentum.

Not quite radically that means the speed
the group philosophy actually which is

a derivative of momentum to energy or
the Omega the decay if you will.

It's a straight line and
this shows you that the.

Electron the band structure of this and
other words the electrons that are in this

graphene are moving with a constant speed
you maybe know on that and that speed

is approximately the speed of light but
you can see a direct measurement of this

of this property of the graphene electrons
directly from such a measurement.

So the measurement really consists of
bombarding the the graphene ribbon with

with high energy photons and
looking at the electrons that come out and

you measure precisely the energy and the
momentum of the electrons coming out and

ultimately to produce that picture
just to show that these ribbons

really are graphene But
this is actually a beautiful image.

And actually meet with us for a second
reason to is because this was a Nature

article that appeared last year in
the some details I want to talk about but

this so-called structure here
is called the diff cone and

you can get a beautifully represented
simply by in this experiment but

now let me go on to what I really want to
talk about is the electronic properties of

these graphene ribbons so.

The first thing is let's look at
the structure using another probe not

synchrotron light but now simply
a scanning tunneling microscope and

if you do that you can actually
image the graphene and

here you see the graphene sitting on
the side wall which has the slope and

we can do something else calls can't get
incoming spectroscopy and we can show that

it's no graphing on the top or
on the bottom but just on the sidewalk.

So here you see some my cross could
be pictures or electron microscopy

I'm sorry scanning telling my cross
the pictures of the graphing ribbons on

the side walls made by our collaborators
in France with our samples.

I don't want to talk about that now we're
going to come into the coming little bit

to the heart of the talk and that's
transports or in the very beginning.

I showed you what happened
with carbon nanotubes and

the carbon nanotubes they're
ballistic conduction properties and

I'm going to show you how we do
that kind of measurement and

it's going to be some real surprises
here and it's going to get interesting.

For those who are a little well it's
going to get really interesting.

So the the what I'm showing here is that

is a device made by my former student
being Ron where he had graphing

ribbons that are draped here you see
a little rock in the ribbon go here and

buy some magic he was
able to fact touch for

electrodes to this and
also put a so-called top gate on there.

So the whole point about that you can to
drive a current through this ribbon that

you goes from here to here.

Actually yeah the ribbon is here.

The current is flowing through the ribbon
and the same time you can apply electric

field on this ribbon to change the charge
density and changing the target charge

density can change the fairly level and
can change the conduct of any and

that's an important parameter
in the in the transport story.

So this is just a very simple diagram but
the point about this graph the ribbon

electric field current going through
it you measure measure the voltage and

the current through it so
you measure the resistance basically.

This is really nothing more than
like having a very fancy ohm meter

on a graph in wire that you put electric
field on that's all it really is.

So what could we see.

So this is going to be
a little bit tricky.

And if you lose me here.

Don't worry it will catch
up a little bit later.

So what happens is we take this gate
will teach this gatefold to choices able

to apply potential to the graph in
the ribbon and we can make a positive or

negative when we make that part positive
what happens you're sucking and

you're putting electric
field on the graph and

we haven't that will suck charges
onto the graph in the ribbon and

the conductance goes up because the
graphing ribbon has more charges in it if

you use a negative potential
Actually you suck electrons out but

what's left is holes and the same story
you get some conductance increase.

But we have no charge on the ribbon
of all it's actually charge neutral.

So there's no charge on it yet it's still
conducts and that's very interesting

because the amount of conduction we found
it to have this is squared over age.

So I want you to get this we have
a graph in ribbon we can change

of charge density we can go positive
negative or is Iraq where there's no

charge density on this ribbon it's
still conducts and it conducts

with the conduct of the squared over age
which is one over twenty seven kilometers.

So it has to actually have to conduct and
so the carbon nanotube but it's quantized.

And you'll see it's really point nine not
one but this is a very very interesting

and so this is a narrow graphene ribbon
that's conducting So we've found back

again this property we saw in the carbon
nanotube now I think I have for reference.

Again what happens in
let me do it over here.

Well no I showed it earlier but
it might come up again.

So if you would do this
with a little graphically

pattern driven a ribbon with really rough
edges this conductance would go to zero.

So the way we did.

With these ribbons with these nice and
healed edges.

We could conductance of
the square over age.

If you do it by going using standard
lithography methods what happens to

the conductance goes all the way to zero.

So that's the difference but
any case we have regained and

we found this this conduct is quantum and

there are some equations here
don't really talk about that.

I want have time to talk about
itself forget about it but

really this is a key point here.

That's what I'm going to do about this
is a little bit of a bounce structure

picture.

Now I can go back all the way
to our very first picture

where I showed the original
calculation of a graphene ribbon done

by trustful house in one thousand
nine hundred six already.

And this is a little bit complicated but
let me see if I can kind of simplify it.

So here we have the graphene ribbon.

It has a within that has a length.

And what happens is because it's
a real if you believe narrow ribbon

the ones we're using maybe twenty to forty
nine a meters or something like that.

You get quantum confinement in other words
electrons bounce back and forth here and

it's just like having a particle in
a box and electron in a box you know

how you quantized up it's in
your quantum mechanics classes.

So that's what you see and what happens is
this ribbon actually behaves like a way

for those people who know about
waveguides know it has most and

so just like electromagnetic waves and
wave guides the electrons

flowing through graphing ribbons also
go in modes and here you can see them.

This is momentum this is energy
this another way to present

modes in the cavity and
you can see that when you're at.

Here you have only one mode
participating to the transport

as you're going up you'll cross
these so-called sub bands and

every time you cross one you and the other
mode becomes assessable in the wave.

But when you're near this point
zero It is only one mode that

participates to the waveguide.

That's a ground mode.

I'm being very.

Crude in my explanation here but
that's basically correct.

What happens is these are electrons
that basically go whose wave number.

Wave factor is parallel to the axis
it doesn't bounce back and forth but

it goes straight through it doesn't
bounce but it goes straight through and

that's the most are represented here and
these are the most of actually will give

you the transport that we see squared over
age and it turns out if you calculate out

where these modes live in a graph in the
ribbon it turns out they're on the edges.

So the point about this this.

If you have a graph in ribbon that's
neutral the chart the current should be

running along the edges and
we find out that it's quantized so

that's really interesting if you increase
the charge density than these other most

start to come in and you can see
a staircase every time you increase

the charge density you move the family
level you used to the next coming in.

The next more coming next coming in and
that somehow that's what you can see.

So.

This is basically a story again but.

Now let me talk again about the response
of the gate already show you haven't been

a member here and what's what you see next
is kind of interesting when we warm it up

the can talk to us actually increases
systematically This is in the fact that.

Was not expected and
I don't think I'm going to spend my

time on this because I've wasted
a lot on my introduction you know.

So what I'm going to do is going
to skip over this is not not so

important for the rest of the story.

But this is.

So again I've hammered on this several
times that when we make our ribbons and

we change the electric field on the
surface of the ribbon to put charges in or

take charges off a charge neutral neutral
when the charge neutral we have this one

squared over H..

Now let's again.

Compared to record what
happens if you have one of

those with a graphically pattern
treatments and you can see.

What happened.

So this is a ribbon not made by our art
technique by simply but taking a graphene

sheet and cutting it with and
with an fogger for something like that.

Now what do you see here.

If you see you change the gate fault
that you see the conductance goes

down by six orders of magnitude in
the comes completely resistance so

that is exactly the fact of roughness and
that we don't see here.

OK while we did were doing all these
experiments we were collaborating with

a group in Germany who who was were going
to do some experiments on our on our

samples and they ended up making
measurements on graphene ribbons

that we had so here in the background
you can see the ribbon This is

an electron microscope
picture of such a ribbon.

But they had a very special machine and
this is an ultra high vacuum

chamber and they have the little

graphene sample sitting here and the
electron microscope is looking at it and

the same time they can put four contacts
to it under the electron microscope.

So this is kind of.

A simple experiment in principle because
what you're doing is you're putting four

electrodes on the graphene ribbon that
you can barely see here but it's here.

You put it run a current to
the outer two electrodes and

you put a you measure the voltage of the
inner two electrodes This is called four

point measurements for the current
is always constant with a ribbon but

you put your electrodes here and
you measure the voltage difference.

And from that you can get the resistance.

So what do they find so they measure
the resistance and now this is H.

over east where it is this is in
units of inverse of the conductors

Kuan this is twenty six kilometers and
they found a very surprising thing

no matter what the spacing was if
you had the well prepared ribbons

the inner spacing of those inner two
electrodes are resistance was constant.

So if you have a constant resistance.

No matter where you are what
length of ribbon you measure.

Now if you remember where I
started this whole story up with

the carbon nanotube story and
its key feature was that the resistance

its resistance does not depend
on length and that was a key.

A key.

Phenomena and here again we see it
you change the interval spacing and

you see the resistance is constant.

So again there is apparently four

with some exceptions these
are made different ways.

It doesn't really matter
how we did that but

you ended up with the condition after
we have properly and Neil that to have.

Mean free pass as they're called
which in order of Mike of order

of hundreds of microns which is tenths of
millimeters so electrons are going through

these ribbons without scattering four
tenths of millimeters perhaps even more.

So that's the very first
interesting thing we saw So

again resistance was transport in contrast
of the ribbons that were made by other

methods where the resistance is so
high that since the the mega on reaches is

a really six to seven orders of
magnitude better than those ribbons.

The other very interesting thing
we've found is that the resistance

of these ribbons does not depend on
temperature which is very unusual

you expect a resistance typically
depend on temperature but

independence of temperature is again
a property of ballistic conductors.

On top of that we change the bias
voltage for those who care and

then factor was very little dependence on
the bias voltage also an important test.

So it passed all the tests.

But now I'm going to tell you a US story
that people knew theoretically for a long

long time that had never been observed
actually not in the way we see it.

And that that is a property of
ballistic electrolysis that's really so

graphic and so visual that it's
actually fun to tell you about it.

So this is a let's imagine you
have a copper wire here and

I put two electrodes on either end and
I put a current through it.

Voltage on it then measure the current
the resistance is simply the ratio of

devolved into the current and that's what
it is now let's say you take another

electrode let's say and simply touch
the copper wire just touch it.

No currents flowing through it.

The resistance should not
care at all about that.

I mean just imagine that I
take a wire like that and

measure its resistance I simply touch it.

That shouldn't change its resistance.

But that's not true of
transports ballistic and

why is that because here the electrons are
injected from one electrode and they fly

to the other electrode and have to select
road sitting here just touching it.

They go inside of here and they get
confused and they can come out either

this way or that way that is so so
electron coming in here goes in there and

temp scatter back simply because you have
electro there so it can fly back and

so that technically what you expect if
you have really a ballistic conductor.

And what you expect if
putting a contact in there.

The resistance should double because
some of you causing scattering.

Now there are other ways to well let's
like a well that's that's the prediction.

T. is so
called The transmission coefficient and

let's say you have a probe
which is called invasive.

If you would contact it here in the
probability the electron goes from Comes

out how to here goes past us like and

continues its path is
fifty percent point five.

TS The half in that case and
half of them go the other way.

So the scattering is is the scattering
will cause the resistance to actually

double and so we wondered Will you see
that and indeed you do so in this this

all try vacuum a scanning tunneling
microscope the experiment was simply done.

We put two contacts down we measure
the resistance of this ribbon and then we

simply out of the contact embalm the
resistance doubled for East squared over

age to approximate to a square to rage and
you put two on there it goes up to three.

Now that's a very interesting and
against a firm.

This time this kind of
effect has been seen and

this is done the really rather
interesting lens of many many microns so

this is something I thought was
a spectacular demonstration of ballistic

conduction Now I don't
know how to impress.

How exciting and important this is.

But it really is.

And again it's an experiment we have
not done with carbon nanotubes but

I'm quite convinced if we've done
the same thing with carbon nanotubes who

would have been able to
see the same thing but

we had never had the access to
the tools that we have right now.

Then comes an even more
spectacular observation that

that actually was quite confusing and
it remains confusing.

So what we found is if
we made that point that

the resistance measurements on such a wire
and we kept increasing the length like

a told you it was virtually independent of
length and it continues to be constant.

But not all the way suddenly after about

sixteen microns the resistance
starts exponentially increase.

Now.
OK if you think about a normal wire.

If I take a length this long.

Let's say it's one ohm if I double
length that should be two arms

are triple lengths it should be three arms
but it should be the resistance should

be linearly proportional to the length
it shouldn't do anything funny.

And here it's doing something really funny
and this is done at room temperature.

It's exponentially increasing.

As far as we can tell there's not enough
in points to really tell you if it really

continues that way but
this is really has no simple explanation

was seen on two different ribbons by
the way I guess in the meantime by others.

Again by our German collaborators.

So at sixteen microns it
sits on the diverges.

Now those people in for milieu are with
major scopic transport anything beyond or

a lot beyond what I'm
telling you right now.

You know something called.

Anderson localization which tells
you if you have a one dimensional

conductor eventually the resistance
should exponentially increase but

we looked at that and.

It doesn't really fit and
this is also done at room temperature and

by the way if that were true that would
mean by the way and it might be true.

It's just I've a little uncomfortable
with it it would mean that we have phase

coherent transport at room
temperature up to sixteen microns.

Now that is a spectacular
effect that means we have

almost like an electron laser.

I mean these electrons are going up sixty
microns is really a huge distance from

Microelectronics point of view and staying
in phase with itself like like a laser

light it's coherent now lot
of wishful thinking will tell

me that this is actually going on here and
if this is absolutely revolutionary.

I think something equally
revolutionary but

slightly more subtle is going on but
I won't have time to talk about it today.

But but the effect is spectacular.

And I haven't really mentioned that but
there's a second step or actually a step

that happens a similar stuff that happens
a little bit earlier where the conductance

goes from well let me show it here
also a second step happening in

the much shorter length on the order of
point two microns we see also a step.

So I didn't tell this whole
the whole the whole story really.

I told you the story mainly about these
electrons but there seems to be a second

species of electrons doing the same
thing but a much shorter length scale.

The point about this is really at
this as as we are right now or

going to get this light in the second
is that we have stumbled on

a subset of electrons in
graphene that seem to happen

when graphene is neutral that her having
incredible properties that the other

electrons don't have this is really
specific to these charged neutral ribbons.

And and again.

Ballistic conduction resistance was
conduction if you will over a sixteen

microns distance this is no we're
seeing this at room temperature.

This is.
And

these systems are not even that clean
you know they're exposed to air there.

It's there's something else going on and

I'll tell you some of the thoughts that
people have about what is going on but but

for me this and these these these
unusual resistance increases that

we see also or
are signaling something very spectacular.

But this is a little bit of
a compas a diagram to show

where we stand now I've talked about
the conventionally produced graphene

ribbons with the standard lift techniques.

If you measure their heat resistances
which is a measure of two

dimensional conductors if you will
the standard ribbons have something on

the order of ten kilo almost
per square down to one kilo or

depending on their with the few so this
is what you typically see typically if

you make a graphene ribbon using standard
techniques you see the resistance.

You resistivity of the material keeps
increasing and increasing as you make

the ribbon narrower and narrower and
goes into the tens of kill on the range.

If you use two dimensional graphene and
it sheet resistance

as it calls varies peace between
something like well something or fifty.

To almost a thousand ohms
per square if you will.

This red line here is
a theoretical limit calculated for

graphene It's just absolute maximum.

Actually it's I'm talking about
resistance here the absolute

minimum resistance you can taking
all the factors to account.

Theoretically if it's perfectly clean and

the only scattering is is phone
on say it's resisted he should.

Bottom out at about something
like thirty ohms for square.

We are already at least an order
of magnitude below that.

So the resistivity we see
exceeds theoretical limits and

the only way that can happen is
if the theory is wrong is not.

Looking at the same thing
as we're looking at and

again to our measurements are not
more complicated than simply

O.-Meter measurements by you know if
you do if you push comes the Soviets.

So you take your own meter you put it on
the thing and you measure its resistance.

Their light technical logically
a little bit more complex but

ultimately that's really what we do so
not much we can go wrong here and

mobility is for people who can handle this
phrase is sitting at room temperature on

the order of ten to the seventh us ten
million which is also a president that.

So we're looking at properties
that are really on usual.

I'm going to throw the slide up
only to show you that people have

actually even predicted things
called perfectly conducting

graphene that are such a thing as
a perfectly conducting channel actually

exactly where we find it in this
so-called region near charge neutrality.

But you know and for Originally we
actually thought that this was the was

the origin of the of the properties
we saw but a necessary condition

is perfection that we simply
cannot believe we do that.

I don't believe and
nobody else believes we really have.

So we were actually more or
less forced to abandon this from

our theorists friends and others who
said look this this can't explain it but

I do want you to be aware that that
there had even been a prediction

of so-called a perfectly
conducting channel.

Who knows maybe have some bearing
on what we're seeing right now and

want to do with this other stuff
that's going on is people also

calculated Steve Louis from Berkeley
has looked at these channels and and

also looked at various aspects
in terms of perhaps these edges

are special because the current is going
along the edges according to The predict.

Is that are there.

Perhaps you know something special
is happening actually he was

looking at an ad in the interaction
would cost feral magnetism.

But it would cause a lifting of
the degeneracy now I'm going to

speak a little bit beyond what I have just
been talking to you about Up till now.

But there's actually a big problem with
the fact that we're seeing one eastward

over age one conducting channel one is
great over age and not to east great over

age for the conduct of any you should
that we should have simply seen too.

And this is a little bit
of a technical detail but

we can only see that if all
the degeneracy are lifted and graphene for

those people who know a little bit more
there are two under one called a volley

degeneracy and the other is called a spin
degeneracy So apparently in our systems

we have to lift all the degeneracy S.
and it has to happen spontaneously.

So one of the things that can happen is
feral magnetism and that's why I'm putting

up the slide that could possibly
explain why we have one square to him.

Not two but the jury's still out.

We really don't know why it's true and
about people are thinking about the stuff.

We're going to forget about this now.

The last little subject

I want to talk about is some very
recent work that was done on

the A ballistic ribbons going on in this
laboratory at this very moment in fact and

these the slides are very
very very recent and

actually is whole and John and
as you gung and Claire were working

on collecting more and more of this
kind of data and this is what we found.

If you'd run the current
through a graphene ribbon.

You can modify it.

Well let me do it differently
if you put a kobold

contact on to the graph in the ribbon
Cobalt is a feral magnet as you know.

So what happens if you have
a cobalt contact on top and

actually if you space a little
bit from the graphene ribbon so

they have to tunnel from the from the
Cobalt into the into the graphene ribbon.

Well you can do special things.

What happens is the electrons coming out
of the Cobalt since Cobalts ferromagnetic

are actually spin polarized.

So that's the trick that people
use to inject spin polarized

electrons into into any structure and
what you can do then if it

if if you put it in this in the magnetic
field and you spin polarized electrons

inside the graph in a ribbon as well you
can see that the electrons can recognise

if they're going to spin
polarized graphene ribbon or not.

And this all comes under
the name of Magni to resistance.

So to make a long story short what we can
do is we can measure the resistance of

the current current going from the Cobalt
magnetically polarized Colwell contact

through the ribbon into the gold contact
in this little loop the other electoral

For get about them but you're having
just current going from here to here and

you're changing the magnetic field and
what you see is the sum of the big

resistance change and
that resistance change is is.

Is associated with when
the spins are aligned or

and the line coming from the cold
bolt going into the graphene.

Now this whole area is
called spintronic sender.

Ferric was here some time ago talking
about this your magnetic read has actually

worked on this put this
principle of spin spin

polarized transport and this whole
phenomena of Maggie to resistance but

the interesting thing about this is
we have a cobalt compact here but

the graphene should not be magnetically
polarized the indication here is

that the graphene ribbon is magnetically
polarized so it's like a feral magnet.

And that's very interesting.

So this.

We don't know exactly how the sparrow
magnet works but there are and it's so Sol

it could be very simple that we have just
all the edges have have spins on them.

They're aligned one kind a way and that
gives you the feral magnetism perhaps.

That's what's going on but there's
certainly something magnetic going on and

on top of that what we also see is
when you change the current around

apparently the spin
direction flips with it.

So this is another phenomena that we're
trying to figure out what it is but

I'm throwing the slide up here to show
you that not only do we have ballistic

transport in this graphic ribbons.

We also have some spin effects probably
spin polarized we north to bliss to

conduction is part of the even persists

until room temperature and now it's
been polarized transporters too as well.

Now what does does is opens
up two entirely new fields

that we can exploit in graphene
electronics One is that

of coherent ballistic electrons may
be coherent at room temperature.

Maybe even up to fifteen microns and
that's on her behalf and the second of all

is spin polar ice transport that might
even persist to very high temperatures.

Now these are the point I'm
trying to make here is that these

are different forms of transport that
don't happen in copper and silicon and

all of that this is unique to graphene So
what we're looking for

is a way of doing electronics that
gets away from this standard model

of select Tronics which is
diffuse of transport in this

is could be coherent polarized
Valley polarized spin

polarized transport in graphene nano
structures that are very well made.

So in my opinion.

We've actually getting close to what
we set out to do in two thousand and

one here.

We wanted to develop a new kind of
electronics that works on the of waif

properties of the electrons and
not simply their diffusive properties and

other new spin variable.

So to speak a state variable as
it's called in the business and

and we've succeeded.

And we have at least so
isolated that we can produce these

polarized electrons the next step of
course is to make devices of them.

So I have many ideas and right now
we're writing proposals about that and

hopefully.

Next time we can we'll meet will
give you a little bit more of

on that I'm just going to throw one more
slide up for those people who who care to

hear about this but I've talked about all
this happening on silicon carbide but

I do want to mention we fought very
hard how we can integrate this

with with silicon technology and
this is how we plan to do that.

So this is the next year's talk when
you know a whole world is full of

graphing electronics what we like to
do is start with a silicon wafer and

do all our patterning in the normal way.

What we do is make all these ballistic
ribbons somehow mac make transistors by

the magic of technology or whatever.

We're going to do next and

then simply we're going to paste on top of
that silicon wafer that's what you do but

you say well that must be very hard to do
well SNOPT it's the industry knows very

well how to do that and
we've succeeded in doing that too.

You can actually cleave off of a silicon
wafer a very very thin sheet and

you can glue it on to something else and
that's called the silicon on insulator or

S. or our technology and and Tom quo and

I have been working on this and
succeeded in doing this.

So we're actually looking ahead to where
we hopefully eventually take this.

This that to the next level where we have
a whole layer of a boost to graphene

electronics down here and then we paste
the silica sheet on top of it and

we very very thin maybe twenty thirty
nine emitters like that and then

we make Vyas In other words little holes
in the silicon to connect to this and

then we can do all or our grid or
silicon seam OS technology on top of that.

So that puts us a little bit at

the end of this I give you a rough
overview of what we're doing.

And I hope you got a little
bit of the flavor of it.

OK thank you for that.

Thank you.

Yeah.

Yeah.

Well I'll tell you what I
personally think as long

as it doesn't leave this room and I like
the camera to be turned off at this point.

No I'm kidding.

I'll say.

No I actually think the whole
story is very very different and

I deeply believe this.

I believe these are life times.

I honestly believe what we're
looking at is the carriers and

what we're look let me expand
the story a little bit.

Graphene was very interesting
before it became popular for

theorists because they realized that the
ground say the graphene must be a quantum

critical system that should collapse and
have a correlated ground state

by some spontaneous symmetry breaking that
is all this talk about quantum mechanics

on the table top and doing experiments
that can only be doing done it.

Sorry it was all alluding
to that property.

So what I really believe is that there
is a next atomic ground state and

somehow we are seeing those electrons and
they're very short lived.

And the lifetimes that are associated
with sixteen microns and

and this one hundred sixty.

Now the meters are in the ten
to the minus twelve second.

So I believe what happens.

We have some kind of very unstable
particle that lives just long enough and

these things are going
with close to speed.

At what street one three hundred to the
speed of light or maybe even faster and

we're just seeing there.

The case and that's what I believe and and

I'm the only one who believes this by
the way and they haven't locked me up yet.

But the other thing about it
is the feral magnetism and

other magnetic properties that
we have seen and fill and

Stroh's you have seen and they also see
in the fractional quantum Hall effect

that don't fit the normal pattern there
seem to be magnetic moments involved and

large magnetic moments on the order
of five more magnetometer or more.

My guess is that's an intrinsic
property to of these compass it.

Let's say not simple electrons but
they're somehow.

Dressed with with other
ones I think to try X.

of Tom's basically I think there
are two electrons on the whole

is what I think they are but
again me and nobody else.

So that's what I think and what I've
done by the way I took the bright Robbie

formula and I simply adopted it which is
the decay time of a pile of positronium

and I put the scaled in the numbers and
it comes up the right values so

you know if you have an exit
tonsilectomy in the hole and

if the case because it's unstable and
it's the right time.

So that's what I think.

But if they behave very different
the other electron so there's really two

kinds of electrons all these kind
of electrons and the normal ones.

So one immediately thinks well in terms
of things even like superconductivity

etc I mean you know if you if you make
a superconductor you have the normal

electrons that have a resistivity in
the superconducting electrons that have.

No resistance and why a small
energy gap opened up as you know.

And because of the correlation effects.

So that's what I think.

OK.

OK Are you talking about the gate voltage
experiments where you see the staircase

going up when you put all this.

So there's an exponential rise.

Yeah.

Yeah well it's consistent
with exponential.

Yeah.

That's correct.

Really exponential.

Well if you want to do it this good and if
you want to prove an exponential I think

you have to go at least
two decades maybe three.

So the problem here in our case was that
the limit was really apparatus limit

the view of the electron microscope
used in this experiment was only

twenty five microns.

And so we were hitting that limit as you
can see the data stop at twenty five

microns these data were
taken about six months ago.

It's relatively new and
obviously a lot more work has to be done.

We're very interested of course to see
how this if this this exponential growth

continues to be that way and more
important how it scales with temperature

because if any of the stuff I said was
true then these lifetimes should depend

inversely on temperature and
that's another thing we're looking for.

Thank you thank you.