Thank you very much.

I really appreciate that gives me more
time to talk which I love to do and

thank you very much for having me here.

It's been a real pleasure
I've really enjoyed my stay.

I'm going to tell you a little bit about
the work that we've done with Syria.

As the title says Before I begin
let me just acknowledge that

a lot of this work would not have been
possible without funding from the O.E.

and the Air Force collaborators
polyphenols year with University of

three S the John Bose in my department and
the students who really did the work.

Mikael Carney yellow
from Kevin McCarthy and

they learned so I hung though from Penn.

Some going to talk about Syria I don't
know how much everyone remembers back from

their freshman chemistry but Syria is
one of the rare earths here it's not

rare turns out to be more abundant
in the earth's crust then nickel or

copper or supply issues but

I'm not going to go into that it's a very
important material in callousness.

Primarily because of its use in
the automotive three way catalyst.

So in automotive callouses for

the last twenty years people
have used this technology.

They have essentially if you
want to remove three components

that's why it's three wake
a callousness you want to remove C.E.O.

hydrocarbons and
knocks in the case of C.E.O. and

hydrocarbons the removal is an oxidation
reaction in the case of knocks it is

a reduction reactions are trying to carry
out oxidation and reduction simultaneously

that's why this makes it somewhat
difficult if you look here on the right

this conversion as a function of the air
fuel ratio the oxygen slick Yama tree.

You can see that.

If you don't have enough
oxygen going in your engine.

You cannot remove the hydrocarbons and
C.E.O.

if you do have enough
oxygen going through.

If you run lean you can remove C O N R and
hydrocarbons just fine.

But you don't have enough for duct and
to remove Knox these catalysts are not

selective and so
there's a very narrow window

air fuel ratio where you can get
the Select Committee that you need.

That's really the basis of three
way to Tallis the way this works is

it's essentially a control problem you
have an oxygen sensor in your tailpipe

that controlled sends a signal a computer.

Which controls the air fuel ratio which
controls the fuel intake it cetera.

And so you can maintain the story
through that control mechanism but

you can't control it closely enough.

That's the problem control
people are very good but they.

This is a tough problem because the person
who's running the car doesn't know enough

to run it steady and all the rest of that
right is Excel orating fast and cetera and

so what you do is you add
a capacitor to the system and

the capacitor in these systems Assyria
the idea is if you have too much oxygen

Syrian reduced forms of Syria
will take up that oxygen.

If you don't have enough it will donate
oxygen so that's really the process.

The problem is that it
shouldn't work if you look

at the thermodynamics of this you've
heard it go in your C.R.C. Handbook.

It should not happen.

So I'm showing here some thermodynamic
measurements that we made where

I'm showing the oxygen to the Syrian
ratio so this is C E O two right.

Syria.

C E tool three would be
at one point five So

this is the action star
is a function of P O two.

Now when you get up to ten of the minus
eight twenty eight atmospheres.

I mean that would be one molecule in this
room so obviously we're not talking about

real partial pressures where that partial
pressure comes from is it's an oxygen few

gas that he that comes from an equilibrium
between reductive and oxygen steam and

hydrogen or C O two and C.E.O.
And if you were to look at.

Six hundred.

Sea which is not too far off from
where a catalyst my operate.

What you see is that is that for ten
percent water ninety percent hydrogen and

guess what exhaust environments are never
that reducing OK they might have

ten percent water but they would
never have ninety percent hydrogen

you still have a stoichiometry of C.E.
one point nine eight.

So in other words that whole process that
I just told you shouldn't work right.

That's one issue that
I wanted to deal with.

The other thing about Syria which makes
a very interesting is that it's a catalyst

promoter and I'm showing here some
work that we did many years ago now

where we look at the rate for
the steam reforming reaction so

methane being oxidised by
steam to make send gas and

this is a log of the rate as
a function of one over T.

what you see here is that Palladium silica
there's a one percent Palladium silica

catalyst is not very active Syria
not very active Palladium Syria.

We couldn't even measure in the same
temperature range these are two different

loadings one in five percent.

Played him but

in both cases the activities were
about ten to the fifth higher.

So there's a huge promotion effect
associated with having both Syria and

oxygen together.

And if you look at the mechanism that we
proposed I think there's a lot of evidence

to support this the proposed mechanism
involves Syria oxidizing Palladium So

you're centrally making P.T.O.
the P.T.O. then can be reduced

by methane to make C.E.O. and hydrogen
that's not a big deal in the important.

The other important step is that you
can oxidize reduce Syria by steam.

So you can essentially go through
the catalytic cycle with that mechanism.

And the key step here is that you can
oxidize the Palladium using Syria.

OK so that's a key step.

I should point out there
was some work at Ford.

Done by George Graham where he looked at
taking a palladium film put a plane film.

A series or Kone a substrate and
he just watched in vacuum.

What happens is he did and he was able
to show that at about one hundred fifty

degrees you could completely
oxidized the plate him so

there's an auction transfer as I'm showing
this mechanism and it clearly does happen.

So it's not just my fiction.

Here.

OK So one of the things that we did to
follow up on this very recently was

we looked at trying to understand what
happens at that metal Siri interface.

And in particular what we did was work
with one of my colleagues chemistry

Chris Murray who makes very
well defined nano particles so

he can make these of any metal he wants.

I'm showing here some pictures that he has
taken the point being that these particles

these metal nano particles are so
well defined so uniform.

That he then allows these to condense into
a solid You get what looks like a crystal.

So this crystal is actually
a crystal of nano particles.

This is Palladium nano particles.

Those are not atoms or

lattices that is the actual metal and
now parts and you can vary the size.

So I'm showing here data for
platinum palladium nickel and

you can essentially change small to large
particles very uniformly you can put this

on a support now and I just want to point
out that let's say with Palladium here

small medium large particles.

You can put those on there by
essentially depositing them and

you get a very nice uniform distribution
of particle sizes on your catalyst and

the important thing that came out of this
was that we looked at SEO oxidation rates

on this and if you were to put these
particles on alumina So these are all

aluminum supported catalyst what you
see is that their rates are almost

identical these are light off curves
the conversion is functional temperature.

You can put these on to.

A rake versus one over T..

And the point I want to make
here is simply that on alumina

this is a structure insensitive reaction
doesn't depend on the the metal particle

size that just occurs on
the metal surface area.

So only the surface area
the metal is important.

OK.

However when you put these on to Syria.

You get a very different answer.

So again the light off curves
change very dramatically when you

change the particle size and
they're quite a bit lower temperature

the other thing you can see here
if I look at the So this is

the pull of the Palladium I believe so
this is the.

Small medium large Palladium
you can see you can

get about a change in the order of twenty
in the relative rate of these materials so

it's a very strong size dependence and

what we did was we then took this and
compared this to models

where we have the rate as a function
of the metal particle diameter.

These are specific rates and

if the rate were current over the whole
surface this would be the dependence.

If it were currying only it being
sites then it would be this dependence

corners whatever point is this
is our data the black line and

the black line fits pretty well with
this come for and surround that.

So again it's sort of pictures that it's
the metal oxide interface metal SCIRI

interface where you're essentially
transferring oxygen to the metal

where reaction occurs and that's true for
all of these metals it turns out.

OK so that's all very nice.

We have this picture where the metal
the oxygen from the Syria transfers

to the metal the problem is should
work OK because if I were to look at

this process play the I'm being oxidised
by Syria one hundred fifty degrees C. if I

just look at the heats of those reactions
I can look up in the C.R.C. Handbook.

What the heat of reaction for
this would be three.

Hundred eighty killer Joules per
mole of one half mile of O two.

I know the heat of reaction for
palladium So this reaction should be.

Plus two hundred five killer
Joules per mole endothermic

now endothermic reactions do occur there's
nothing wrong with endothermic reactions.

But if this were a reaction
that's occurring.

This is per mole of Now then the.

Any reaction involved in that mechanism
has to have an energy barrier that is that

at least as large as the heat of that
reaction and my argument here is

that there is no way you can come up with
a reaction is going to have an activation

energy that is two hundred ninety five
killer Joules per mold No nothing's going

to overcome that barrier at one hundred
fifty C. It just shouldn't happen.

OK.

The other thing I should point
out is that you could argue this

may depend on Spike you know
material Well it doesn't.

If you look at the thermodynamics
that's been published

that reaction is very
independent of oxygen.

So in other words it shouldn't occur.

And the third problem that
I want to talk about today.

I haven't really laid out
the whole picture here.

This is work from Ian and John McCarty.

Where we what they were doing was they
were interested in methane combustion.

And again this can go up to high
temperatures but automotive catalyst from

go up to this too by the way what they're
showing here is that the Palladium alumina

catalyst over a period of one hundred
hours of these temperatures went from

small particles to very large particles
and that is a disaster basically.

You're losing your metal surface area or
you're losing your mental activity and

so it's this problem also
that we like to confront.

So that really leads to my outline for
this talk.

I know I'm half done already.

But this is my outline and

that is first are the Thurmont and
Amec redux properties.

Bulk Syria representative
catalytic forms of Syria.

So catalytic forms are typically
nanoparticles they usually involve

solid solutions with their Konya OK So
is the thermodynamics of that the same

second question that I'm going to address
is what really is storage is it just

simply a capacitance I'm going to
argue it's not show you in a second.

Why I believe that.

And third are there ways we can enhance
performance and prevent centering So

I'd like to talk to you a little bit
about some work we've done in that.

So let me deal first of all with
this question of oxygen storage.

A very common way in which people measure

oxygen storage capacity
is to do pulse testing.

So the idea here is if I were to
take a Platy I'm Siri catalyst.

OK I would hold it at five hundred C.

So there's a fresh catalyst and
now what I do is I post in oxygen.

OK this time I'm pulsing in oxygen.

I feel a bit of C O two here
it's OK don't worry about it.

I then Paulson C.E.O.

and I get C O two coming out and
I can do this over and over again.

Now I can measure the amount of C O two
that comes off and some of that is from

reducing the Palladium but most of it
turns out to be from reducing the Syria.

OK so
that's a way to measure the oxygen stored.

What we were doing and
we weren't necessarily looking for

this at the time we were looking at sulfur
poisoning I should point out that sulfur

is reduced to very low values in
the gasoline that we get today.

Primarily because sulfur is a very
effective poison for oxygen storage.

If it weren't for removing getting rid
of oxygen story if you weren't losing

oxygen storage capacity we could live with
one hundred ppm of sulfur in the gasoline

right now it's reduced below that and
so it's primarily because of this now.

What I'm showing here is this is
a catalyst to spend exposed to air so

to for some period of time so it solvent.

OIS and
I Paulson oxygen OK nothing happens now.

I Paulson C.E.O.
and look at this video to peak.

It is enormous compared to this.

So I have actually increased
the oxygen capacitance by

this measure even though I have
effectively killed the oxygen

storage capacity as far as what
the car catalyst would say.

So the reason that this
oxygen increases is

that the sulfate group
itself can be reduced and so

you can actually get the sulfate coming
off of as decomposed into the oxy sulfide.

Now why is oxygen storage doing that.

Well it's been known for a long time and
if you look in the automotive literature

you would see from work from again
from Ford showing that C O two and

water are critical.

If you want to measure auction sort
you can't measure oxygen storage in

the absence of steam or water.

Well you can start to see why that's
true from these polls studies.

So here's a fresh Palladium Syria
catalyst at four hundred fifty C. and

what I'm doing now is I'm pulsing
in SEO OK second polls of SEO and

now a pulse in water and the point
here is this pulse of C O reduces

I form C o two here so
I'm reducing the catalyst in water.

I'm getting hydrogen.

So I'm oxidize in the Catalyst with water.

OK And then when I pulse in
C O again I see more C O two

hydrogen from decomposition of
the of some hydroxide five K.

but what happens on the sulfur
poison catalyst is I start off here

with an oxidized catalyst
if I reduce it and see.

I see a lot of C O two.

I've made the act itself I again.

But however when I post in water.

I don't see anything or
very little of anything.

And when I Paulson C.-O.
after oxidizing and water.

I don't see much C O two here either.

So in other words what I'm really
showing here is that this Oxy sulfide.

I cannot be oxidized in by by steam or
C O two this essentially the problem and

that is the key point it's not that it
can't be oxygen oxidized by oxygen.

So oxygen storage were a simple
capacitance issue where you were just

taking oxygen in and out this would
not have been observed right.

So action storage is not is really
more a promoter in C O two and

water are really key to that.

That's actually part of the reason
we were interested in in

in the whole reactions with Steam etc So
the point I want to make here.

Oxon story.

Not a simple capacitance it's
much more complicated than that

it's really a promoter who says
hydrogen what I think it really does

is that hydrogen is a much more
effective reduction for Knox.

And so what happens is by having that
promoting capability in there you

essentially widen this
window of opportunity for

removing the knocks in your in your.

Air.

I'm sorry in the automotive system.

I should point out that C.E.O.
is notoriously insensitive our

air oxygen sensors are notoriously
insensitive to C.E.O.

So by having hydrogen present by
having a promoter that makes it.

You're also enabling the oxygen
sensor to work better and so

that's I think also important.

I'm not going to go into the reasons why.

But that's OK.

So let me get back to question one.

What are the Redux properties
of catalytic Syria.

I'm going to assert first of all the the
most common method to use to measure this

is using temperature program reduction
which I think is essentially worthless.

OK What we have argued is that

the technique to use here is
thermodynamic measurements.

So if I consider the reaction
copper plus oxygen goes to.

See you too.

OK I can look at the thermodynamics
that process and I know that

the equilibrium constant for this reaction
is just inversely proportional to one

over the P.O. to the one half right so
the activity of a solid is one.

So therefore the thermodynamics is
simply dependent on the P O two.

If I can get that equilibrium
constant then I have Delta G.

which gives me equilibrium constant
they are right I can also get dealt H.

from temperature dependence
I can get delta S.

If I want that such a and I should keep
in mind that these equilibrium constants

are going to occur where P O
two is a really really low.

So what I told you before is you're going
to have to use equilibrium to get your

P.O. to use these are not
going to be actual P.-O.

twos these are going to be ratios of
water to hydrogen or C O two to C.E.O..

The way we do these measurements is shown.

So essentially takes advantage of.

What is essentially an oxygen sensor.

So an auction sensor is just a cubic
zirconia electrolyte with platinum on both

sides we use over in both sides doesn't
really matter what does actually

not going to that put
a sample in a sealed chamber.

OK And it turns out that this oxygen
sensor slash solid oxide fuel

cell the voltage that builds up across
that is related to the nurse potential and

the P O two across it.

Turns out that this is
also an oxygen pump so

therefore I can apply a current
across this to pump oxygen in or out.

Keep in mind again the way we do this
actually is with a simple two but

that's not that's complicated right.

And I typically back fill it with C O two.

So I'm actually changing
the C O two C O ratio.

This shows you some data that
we took I could show a lot

of other data I'm not going to go.

Thank you.
Board but

what I'm showing here is the oxygen
tube an A.T.M. ratio is.

The tool five right.

As a function of P O two and what you
see here is this is three different

temperature so that your can get dealt H.
from this what you can see is that

if you start with let's start with
you up here so vanadium plus five B.

to A five as you pump oxygen out you get
to a point where the action pressure

doesn't change as you pump oxygen out and
go to an A.T.M. plus four.

OK so this would be V O two and
then you go to a new pressure and

it pumps two oxygen.

Plus three.

All right so
you have these two steps in this process.

There's a transition here from the for

over the two materials because there
are some mix of an A.T.M. oxide so

you don't have pure faces here with some
of the others we get a nice transition

other oxides we get a nice transition from
one face to the other you can also get

Delta H. because we have the temperature
dependence of these curves as well.

And this just points out that the
equilibrium constants based on the C.R.C.

Handbook.

Are Dead on what we're saying they
are here and the heats from C. or C.

versus what we measure
are within twenty kill it.

Jules promote a pretty good.

So what happens with Syria.

Well if you were to look at bulk Syria.

That's shown here looks just
like what I told you before.

It's basically an reducible you do not see

this is three different
temperatures again.

Oxygen to Syrian ratio
as a function P O two.

You don't see any reduction of
Assyria until you get the P.

O. TOS that are ridiculously low.

OK cases where you would never see
it in automotive when you put Syria

in this case thirty percent
lanthanum alumina eight.

What you see is that you get reduction
at much lower P. O. to lose.

OK And if I look now at
the temperature dependence what

you see is that the act
the energies associated with this.

Syria is about seven hundred sixty Killer
jewels with all this what we measure this

is also what the C.R.C. Handbook says
that's promotion of zero two now but

the heat of reaction for Syria is more
like five hundred twenty five at least for

the reduction in the initial phases.

So it's a very different form of Syria.

This is now just Syria placed
on on the support now that

thirty percent is a lot that's way
more than a monolayer so I don't think

of this is being a monolayer of material
it's particles of Syria support.

Why is this the case.

I'm going to speculate here
no I don't really know but

my argument is that in many
environments Syria is more stable in

the plus three oxidation state case
the Syrian plus three is stable.

If you were to look at a quick solutions.

If you were to look at materials
like Syria and vanity.

This is a impurity that occurs.

Actually it's intentionally
occurs on turbine blades.

You cannot oxidized the Syria plus four
in this particular case even in a plasma.

So Syria in many environments
does not want to be plugged for.

What happens in the fluoride environment
I'm showing here the fluorite structure.

Is that the Syria.

Plus four has the buns
oriented to the oxygens in

just the right configuration the four F.

electrons just match to stabilize
this plus for oxidation state so

it's a very different situation
from other symmetries and what I'm

going to argue happens in low temperature
forms of Syria or small particles of Syria

is that you have a lot of defects that
help stabilize the plus three state.

So it's a very different type of material.

OK very different from bulk Syria now
I should point out that thirty percent

Syrian land alumina if you read ox cycle
it enough it will move towards the ball.

OK And so you can't use pure Syria
in an automotive environment.

What's done.

Is they use serious or
Konya solid solutions and

I'm not going to go through
the data you'd be bored to tears.

Right.
What I'm showing here though is we looked

at a whole series of materials solid
solutions with different zirconia Syria

ratios and what I what we have I'm
showing here is the heat of reaction

of oxidation as a function of
the oxygen stock Yama tree.

OK And for bulk Syria.

Again seven hundred sixty kilograms from
all our measurements are pretty close.

I mean there's a little bigger error
bars but rough order of magnitude but

all of these series are Kone as.

With difference like Yama trees as
a function of different oxygen contents

are all about two hundred two hundred
fifty killer jewels promote lower.

So it's a much lower heat of reaction
belt ages five twenty five or

seven sixty pages independent
of the extent of reduction and

surprising independent of composition so

independent of the series of Konya ratio
at least within a certain range and

interestingly interest similar to
a pirate floor structure of reaction.

I recall or
that was published in literature so

what we really think is going on here.

If you look at the fluorite series or
Kone is structure and

I'm looking here with two zirconium
atoms those little atoms there are Konya

what we're arguing is that
when you remove an oxygen

you're taking a vacancy and
you're pulling that vacancy out

if you will in a certain position
with respect to those two zirconium.

And so
now you taint to Syria two plus three and

this locally looks like
a pirate floor structure.

Even though it's only within the bulk
it's not the whole structure.

We're arguing that locally
that we're forming this

pirate chlor like material and
that's what.

It's the heat.

So in other words you don't have
some kind of long range order here.

It's really the local structure that
gives you those kinds of energetics.

Now getting back to a possible answer
leaving a question mark there isn't

a going to argue that this is
necessarily the total situation but

a partial answer to question one.

If I now want to ask the question can
I played in with series or Konya.

Now if I use the new numbers
that I have instead of taking.

Three hundred eighty half of
three seven sixty Killer jewels.

I take to fifty five hundred and I add
that to the heat of reaction of P.T.O.

now I come up with plus one sixty
five killer jewels from all

that's still in the thermic I'm not going
to argue that the oxidation in play.

B.M. isn't into thermic but it is at least
has a barrier hype that's more reasonable.

OK So we believe that that's really
what what's going on here and

why is the Syrian issue only active a
series or Coney A Well again there's a lot

of defects in that structures of catalytic
materials similar has a similar the fear

and so it's really that that's causing
giving us the catalytic activity.

So let me go now into the third question.

That I talked about in
the very beginning and

that is are there any is there anything
we can do to enhance reactivity.

Of these Syria based catalysts and
potentially prevent centering And

so what we've been
looking at are these core

shell structures the idea is
that you have a metal particle.

In the middle protected by
some kind of oxide shell and

the idea here is that you'd have
optimal contact between the metal and

the Syria we know it's at that
interface where reaction occurs and

we also have the possibility to
prevent the metal from moving around.

Now there's a lot of ways to
make core shell particles.

I'm not going to go through.

All the different methods.

We've developed a method in collaboration
with polyphenols hero and me and

his student Mikael where
what we do is the following.

This is all done solution.

We start by taking making
a play the I'm nanoparticle So

you just reduce a Platy I'm solved in
organic solutions with tiles around.

So those red squiggles
there are all files.

Sulpher endurance they're at
the end of those tiles though

over this was material this idea is
you put a car box a lender upon.

OK so it's a file on one
end car box on the other.

That car box all group now allows
you to react with an Al Cox site so

we can bring in a Syrian male cock
side and form an asteroid linkage.

So now you have a serial
cock side bonded to that

through the through the IF
by all the leg and.

The heart.

If you're.

Problems here.

Way.

So

this is why

this.

I'm very

very.

Trying

very

very while

yet.

So let's try.

Let

me

go.

Right

right

right.