it's and thank you for the invitation

enjoy meeting with great work that's

being done by three students and

postdocs here it's always very exciting

how does that work can you hear me

alright it's always great yeah energetic

energized and exciting to come on these

visits you get to see all the really

neat research and the highlights of the

work that's being done in the different

groups and I always go back really

energized and motivated about all the

great science that our weakness being

done by friends and collaborators at

other universities show thanks thanks

again for the invitation so today I

wanted to talk to you a little bit about

one of the things that we'd worked on

over the last couple of years we work on

all kinds of different things we do

computational material science or

computational chemistry using quantum

mechanical simulation methods and so

it's really easy for us on a computer to

just switch systems as easy as is

sometimes to just retype the names of

some atoms now it takes a lot of

expertise to actually model those

systems correctly but it gives us a lot

of flexibility where we can go in and

use some of these fundamental modeling

tools that can describe material systems

or molecular systems and to be able to

do design and engineering of those

systems from a very fundamental aspect

I'm gonna tell you about this one story

today that was that we've done over the

last couple of years involving

computational design of organic public

habit catalysts in photocatalyst and

this was done by my former student

turned away limb and our collaborators

in the chemistry department student

named Jordan thereö and a chemistry

faculty member named here at Miyake so

as you probably all know we make a lot

of plastics we make about 600 billion

pounds of plastics a year most of that

is done by chemical engineers and so

that's a lot of material so we better be

very good at doing it and they're making

quantities on that scale

that's about 300 pounds per person in

the US and about 85 pounds per person

around the world so a lot of plastic

most of this is done by free radical

polymerization some free radical

polymerization you chained together

monomers to make a polymer and you do

that by having some initial radical that

attacks a monomer and transfers a

radical to that monomer which then that

radical attacks the next monomer and

grow the chain and the radicals transfer

that radical attacks the next monomer to

transfer the radical to the growth a

chain longer and so you end up getting

these oligomers a poem or chains and

they can be very long and of course but

the length of these polymers for example

determines the properties of materials

that you make now free radicals are

quite reactive so you get all kinds of

different structures here's just a few

cartoon examples of the kinds of

structures that you might make with free

radical polymerizations again you have

the initiating radical they start with

it caps one end of the polymer and then

you can get structures of different

lengths and they can be brand structures

etc but you get and then when they

terminate you'll have eventually your

growing polymer that's a radical we'll

find another radical and those radical

to react with each other and Clint and

that will terminate your polymerization

so the problem with this again is you

get this whole variety of different

structures and if those different

structures you get this broad range of

molecular weights so this broad

distribution gives you oftentimes low

quality materials so things like

low-density polyethylene you can sell

them for basically the price of whatever

your city decides to charge you for

passive bags at the grocery store

Boulder it's ten cents a bag mainly

because they don't want you using

plastic bags but they're basically

disposable materials and so if you

really want to do better than this

you're gonna have to somehow gain

control over this process the problem is

is that these radicals are very reactive

most of them are made by free radical

generation either thermally or using

light activation in thermal activation

you take molecules for example like this

one heat them up and it fragments into

three species

- and these two radicals those radicals

go off and react with monomer to grow

polymer in light activated

polymerization you'll have photo

absorbing molecules like this one that

absorb the light and then photo

dissociate to create these two fragments

those two fragments or radicals that can

you then go off in reactive monomer can

grow your polymer and again these

radicals are very reactive they're hard

to control and poorly controlled

radicals lead to quality control

structures large dispersity dispersity

is one of our metrics that the breath of

that molecular weight distribution when

that's broad it means basically that our

reactions are not very selective and as

Temuco generators usually we're looking

for very selective reactions and a free

radical polymerization you really don't

get much selectivity in terms of the

types of polymers and you grow okay so

how do we avoid this problem with free

radical generation and get an more

controlled polymerization well one way

to do it is by using what's called

reversible deactivation and in

reversible deactivation you form this

equilibrium between the active radical

form of the growing polymer and the

dormant form so you're drawing this

polymer and if you can cap it with some

terminating group or some capping group

that is no longer radical and it's

dormant and it doesn't grow and if the

rate of deactivation is much faster than

the rate of activation you will you can

gain control over the polymerization and

what that does is it lowers the the

concentration of radicals so fast

deactivation lowers the concentration

radicals and when your radical

concentration is low the probability for

biomolecular termination is low so and

by molecular termination and you have

these two rolling polymers coming

together and and bonding to each other

that's a large addition molecular way

all at once and it really adds to your

molecular weight distribution

significantly to give you these nice bra

distributions

so yeah we can have kinetics where we

can design react your activity so that

we have this dormant state that's

preferred over this active state and

just a low concentration of our active

radical we can pain control over and

clamors ation and so instead of just

having our normal radical chain growth

polymerization we have these dormant

States and you rapidly go into these

dormant States and then slowly going

back into active form that grows for a

while Dorman active road-going active

grow at that time so we really want to

control those kinds of kinetics and

design systems that do that so if you do

that this is just a cartoon but again

get much narrower molecular weight

distributions and there's much narrower

molecular weight distributions if we can

make it much narrower it gives us higher

control over the structure more valuable

materials so the metric again that's

used typically for describing this

molecular weight distribution is

dispersity so it's basically how broad

is the molecular weight distribution

you're trying to get the dispersed ities

as close to one as possible so in some

radical polymerization processes some

processes can get down very close to one

some of them don't get down very close

to one so this is our metric versus how

well we're doing in terms of controlling

the radical polymerization and

controlling the molecular weight

distribution the typical type of

controlled radical polymerization that's

the news recently and that we started

studying I'm going to talk about the day

called atom transfer radical

polymerization and an atom transfer

radical polymerization you have this

growing polymer it has a leaving group

on it this is the dormant form you

activate it somehow into its radical

form radical reacts with monomer do

polymerization and then at some point

you have deactivation which you hope is

fast and that takes you back to this

dormant form and going from active form

to dormant form mostly involves taking

this leaving group off to activate and

putting the leaving group back on the

activated so that's called atom transfer

radical polymerization and it's

typically catalyzed by catalysts that

contain different types of mostly

transition metals and copper so and and

this this has gained a lot of interest

over the years and one of the beautiful

so much interest actually shown you this

subplot here this is the cumulative

number of publications between 1995 and

2012 on atom transfer radical

polymerization just one type of

controlled radical polymerization so

Green is atom transfer Reauthorization a

PRP and between 1995 when it was

discovered by Manish SP in 2012 it had

about 13,000 publications on that on

that process it's a lot of interest in

this and the reason there's so much

interested in this is again because this

is one process for growing gaining

control over these structures over the

polymer structure so using a TRP can

gain control over the composition the

topology the functionality and control

over architecture and by gaining control

over these structural properties of the

system you can then open up different

applications for where to apply those

materials that you make with a TRP so

that sounds great that's the way a lot

of people sell a TRP but then the next

question might be well why don't we see

a lot more of these applications

actually happen there are some

applications but it's not as big as what

you'd expect

especially considering there's been

13,000 about six years ago it's probably

up to 20 something thousand now

publications on this topic well one of

the main objections to this has been

that when you catalyze a TRP you well

this is what matters chessy says the

primary objection to commercial

introduction materials prepared by a PRP

was the presence of a significant amount

of the transition metal catalyst complex

in the final product so that could be a

problem for a number of reasons if

you're doing biomaterials oftentimes you

want we don't want transition metals in

your biomaterials that end up being

implanted into your patients for example

if you're doing electronic materials or

opto electronic materials you

don't want metals in there because

they'll kill things like your your

electronic carriers and makes the

materials that you're doing like a a

let's say organic transistor or organic

LED it kills those devices so this is

really limited the application just this

metal looks in there of course you could

separate out the metal chemical gears

are good at doing separations but it you

still need some metal in there and it

adds another unit operation another step

to the process and it's really not ideal

it'd be better not to have the metal in

there in the first place so HRP mainly

takes place by two different mechanisms

the first mechanism is called reductive

quenching and in reductive quenching I

start with a photo catalyst that photo

catalyst I'm gonna make the person on

the video camera and the reductive

crunchy cycle we start with our ground

state pole catalyst we shine light on it

and we go to an excited state in that

excited state it reacts with the

sacrificial donor typically in a mean

and it transfers an electron to this

excited state of the public catalyst to

make the anion of the photo catalyst

that anion then transfers an electron to

my initiator this initiator may be my

initial initiating radical or it may be

the polymer that's already added a

certain number of monomer units to it

when I transfer an electron to the

initiator this leaving group comes off

creates the radical radical reacts with

monomer new polymerization so if I start

my find ground state of my photo

catalyst and I'm just showing the top to

highest energy electrons sitting in me

though the highest occupied molecular

orbital I shine light on this and that

photo excites the photo catalyst so I

promote an electron to a higher energy

orbital and then in this excited state

this sacrificial donor these electrons

have enough energy to reduce the excited

state basically the back fill that hole

that's left behind by the photo

excitation so it does that there's an

electron transfer that gives me the

negatively-charged photocatalyst now

this electron is high-energy

it has a lot of reducing potential and I

can transfer it to things like

initiators that I want to activate and

so this is a pretty interesting system

one example of this was reported by Troy

in 2011 it has the diversity it's

between 1.7 and 3.8 say 1.7 and 3 point

that's actually not great you want to

get as close to 1 as possible even 1.7

is not that good I'd consider that just

ok so if you look at this there's a

couple of problems with it and the first

problem you might notice is that let me

transfer an electron from the

sacrificial donor to the excited state

we produce a radical now this radical is

in addition to the radical over here of

my growing polymer that I want to

activate so now I have two radicals I'm

trying to control the radicals and now

I'm introducing additional radicals as

byproducts in my system that's probably

not a good idea as an engineer I'm

trying to I'm trying to increase

selectivity and usually you do that by

not introducing additional side

reactions so this has some inherent

problems and because of those inherent

problem it's really hard to get those

like this versity is lower than then

somewhere in this particular range so

another mechanism an alternative

mechanism is called oxidative quenching

an oxidative clenching I take my

photocatalyst excited with light and

then this excited state I transfer an

electron directly to my initiating

molecule either the growing polymer or

my original initiator and then that

electron transfer activates this where

the leaving group comes off goes here

and I have a radical reacts with monomer

to grow polymer and then there's our

deactivation step that takes me back to

my ground state of the photo catalysts

and the dormant state of the growing

polymer so an example of oxidative

quenching is catalyzed by this iridium

so this Iridium catalyst can catalyze a

TRP hawker discovered it and reported in

2012 it's dispersed these are naturally

lower than molecules that catalyze

through the reductive Crunchie cycle

dispersity is about 1.2 is is about as

far down as they've been able to get for

these types of catalysts pretty good 1.2

is pretty good but still not fantastic

so we like the new better than that okay

so just a little bit more of the photo

physics it turns most you know this is

something that a lot of my colleagues

and a lot of my students don't spend a

lot of time with but there's a there's a

lot of opportunity in photocatalysis and

for chemical engineers to go in and

start working more in photocatalysis and

some of us do already but not very many

of us and if you're going to be a good

engineer in a picker area it's good to

understand fundamentals that govern how

these processes work so I'm gonna tell

you a little bit about about some of

these photo physical processes so the

first one is going to be activation

where iphoto excite my photocatalyst and

then do a direct electron transfer so

here's my ground state of my photo

catalysts and in this ground state of

the photo catalyst I shine light on it

that activates it

oops that activates it promotes an

electron into an excited state now Bo

tongs carry no spin so notice when I did

that photo excitation the two electrons

I allowed the one electrons absorb the

energy from light that I didn't allow it

to flip it spin and so that conserved a

momentum in the excited state of the

photo catalyst this is pretty simple

there's enough reducing potential of

this electron to transfer directly to a

lower energy orbital of my initiator

molecule and so we'll do that and when

it does that my initiator gets activated

this extra electron density goes off

with the leaving group so so that's just

that's the simplest bull excitation

going from say the homo to the LUMO and

a direct electron transfer from the

excited state o slightly more

complicated complex process slightly

more is where my photo excitation isn't

into the lowest excited orbital

it's in the sub-orbital it's higher in

energy than the lowest higher than the

LUMO the lowest unoccupied molecular

orbital and this particular photo

excitation I start with my ground state

with my photocatalyst I shine light on

it the phone excite it it promotes the

electron now into a higher energy

orbital not the lowest one now this

might happen because it has those two

states are most strongly coupled by the

electric field of the light that's

called dipole couple and so when these

are strongly dipole couple I have strong

absorption and I can do that excitation

well from that excited state I can have

what's called an internal conversion so

an internal conversion just takes that

excited electron and it just relaxes

that electron into a lower energy

orbital but still into an excited state

so still an excited orbital so this is

still excited I gave up some energy and

then from there I can do an electron

transfer so this can happen sometimes

it's best if this is strongly absorbing

and then and this may be the next higher

orbital it can be orbitals even higher

in energy that can happen in cases where

there's strong oscillator strength or

strong electronic dipole coupling

between these two states so that happens

sometimes and then there's a third

possibility this is just three of

actually many and this is where the

electron transfers from a triplet state

so I start with my ground state of my

photocatalyst and I shine light on it as

I already mentioned I excite into a

singlet state that means like the spins

are all paired my spin doesn't change

during that photo excitation it stays

spin down as it gets promoted to the

higher energy level from there I can

flip the spin so this is called an inter

system crossing that means you're

crossing from the set of singlet States

to the set of triplet States when I have

a net spin here of 1/2 plus 1/2 that's a

spin of 1/2 this one out that's 1 that's

a net spin of one that those are called

triplet States

so in this triplet state I'll easily

bit of energy but if they can transfer

that electron over to my initiator nice

thing about about me stripping the

triplet state and the previous states

and they could have really long lifetime

some of these have really long lifetime

so designing molecules going in and

designing my molecules specifically so

they have long lifetimes for example if

I could design the state knock should go

backwards like it if I could go through

a triplet that has a really long

lifetime it has enough time to stay in

this excited state and wander around and

collide with an initiator that it can

then transfer electrons you otherwise if

a lifetime short it will just go back to

the ground state before it has time to

transfer the electron and the energy

from the photon that originally gone

okay so that's just a little bit of the

some of these photo physical processes

in general kind of language that we use

to describe them okay so now it's back

to the mechanism we really want to

understand the mechanism and we want to

understand the specific steps in the

mechanism so we can can design molecules

to control these reactions so here's my

initiating molecule with that leaving

group after I've already added and

monomer units to it I activate it by

transferring electron to it to create my

radical and live evening group it react

the radical reacts with monomer

nuclearization and now I have this

branched mechanism where I can either do

biomolecular termination or I can do

deactivation and the faster I can make

deactivation relative to by molecular

termination the more control I'm going

to have over structure the better my

diversity is going to be so I'm really

trying to control the relative rates of

deactivation versus by molecular

termination I need fast deactivation so

that's kind of a theme in the research

that we've conducted was designing

molecules that would have been undergo

very fast the activation and what were

the principles that we could use and

exploit in the design of molecules he

spoke at Allah

to make them deactivate quickly okay so

there's our mechanism again this photo

catalysts that I'd mentioned that Hocker

discovered back in 2012 it is a metal

containing a PRP catalyst as a metal in

it so that's not ideal we'd rather not

have a metal in there we don't want to

separate it out later and in many

applications you don't let that metal

there it has a reduction potential of

negative one point seven volts so for my

friends to do electrochemistry they'd

say that has plenty of juice to do

electron transfers to many different

systems that you might want to reduce so

for example the types of initiators that

we could have that would be at the end

groups of these drilling polymers or the

initial initiators have reduction

potentials of about negative 0.7 volts

these these four cases here and these

are these are typical initiators

negative point seven versus negative one

point seven I have about an extra volt

of reduction potential so there's plenty

of thermodynamic driving force to get

the electron transfers to go with this

photo catalysts but again that photo

catalyst has iridium in it and it's not

only iridium which is a met and it's not

only has a metal which we'd like to get

rid of the metal but it's iridium it's

the rarest of the rare metals and I see

publications all the time where people

are talking about using iridium and if

I'm thinking about scaling up the

process I'm thinking I'm gonna scale up

across US and make 600 billion pounds of

something annually the thing is a

catalyst but still I'd rather not use a

metal that's as rare as iridium is so

there's other other choices maybe we can

do something and that's use a catalyst

that's not only more environmentally

friendly more friendly and biomaterials

etc but also that is something so non

toxic more available less expensive etc

may be something that you might even buy

at the grocery store so I buy these

kinds of things that we're storing eat

them occasionally they

you should be eating these because

things are nice and colorful that means

they absorb an indivisible and they're

relatively inexpensive we can light

pounds of this stuff so and again I just

need these Provo fours these molecules

that absorb light that are naturally

occurring systems I just needed them to

have reduction potentials in their

excited States that's more negative than

0.7 volts so here's just three examples

of many many naturally occurring organic

chromophores

so Laura Steen yes and why and alizarin

red they all absorb in the visible for

fifty five forty 420 nanometers they're

all organics they have reduction

potentials that are all more negative

than 0.7 negative one point three into

one point one eighty to one point six so

at least in terms of being organic

absorbing light absorbing light in the

right range not having and having

reduction potentials that are

sufficiently negative these all look

pretty good now I'm going to tell you

about other design criteria in addition

to those that we need but you can you

can start to see how sure organic

molecules might not be a bad class of

systems that we could use for doing

different types of catalysis for example

a TRP okay

so here's some motivating questions that

we had back it's been three years now

when our collaborators came to us and

told us about their idea about using

organic catalysts new photo

polymerization by a TRP and there it

Miyake when he came to talk to me he

basically had a design in mind a

molecule mind and we're computational so

he said to me which of whether this

molecule you think if this molecule

would work for a TRP that's basically

the question will this work for a TRP

and the reason he asked that was because

they figured it was going to take them

anywhere from six to nine months to

figure out how to synthesize it and they

didn't want to waste a year for example

to synthesize it when it wasn't a very

good choice and so they were hoping that

we can give them advice about what

molecules might be the best ones to

synthesize if this wasn't a good choice

so some of the questions that we came up

initially with were we're gonna do

design of a TLT catalyst can Organo

based catalyst replace transition metal

catalyst rate ERP pan organic catalyst

achieve low dispersity high conversions

and high molecular weight control of

your polymer materials an organic

catalyst she can write an account must

be sufficiently strong reducing agents

they probably already answered that

question and high-performance organic

coder kind of must be activated by

visible light it'd be nice to have

visible light instead of ultraviolet

ultraviolets takes a lot of energy to

produce ultraviolet light and with that

much energy and your photons we can

imagine you now have enough energy to

activate other reactions and again we're

trying to avoid side reactions we we

want selectivity we learn and control

structure so visible light better than

UV light and then finally can we design

high-performance photocatalyst in

general using types of computational

chemistry methods that have been around

for a while but engineers chemical

engineers like myself have been using

over the last couple decades now to try

to design catalysts and other systems to

make them better options for different

engineering applications so here's the

here's three existing catalysts when we

started our work three years ago

so here's hawkers Iridium catalyst or he

told you it contains a metal even worse

means iridium it absorbs invisible

that's good

and it's dispersity is just okay one

point to getting out about one way to

that's that's reasonable but it's not

fantastic Walker also published on a

different organic catalyst so this is

better in the sense it's an organic it's

a fantasizing and it unfortunately

absorbs in the UV so that wasn't so good

it's dispersed --'tis again we're pretty

good but and reasonable but didn't quite

get down to the range where we really

wanted again this was the first organic

molecule that my collaborator had worked

on it works ok it's all organic that was

good

it absorbs an invisible that's good but

it's really hard to get the dispersant

he's low enough down to one point three

again reasonable but not fantastic so

the kinds of designs that we came up

with or collaborators to go off in and

work on we were hoping can perform

better than the existing catalyst

otherwise there'd be no reason to view

this research and to come talk to you

about it today

so one of the first designs that I'm

going to tell you about that we came up

with is molecules called phenazines they

look like this and they are all organic

they absorb in the visible and we can

get the dispersed to these below 1.1

even without much often without any

optimization really just some conditions

that we try occasionally we get this

person below 1.1 they work really well

I'll tell you a little bit about the

thought and the thought process that

went into designing and selecting these

molecules okay a little bit more of the

photo physics just to give you another

step or detail so in total excitation of

my full catalyst from its brown state

that can go into the LUMO or the photo

excitation can excite an electron and

there's some higher energy level we want

that to be fast and efficient so we want

to design molecules that have high

absorbance we don't want to waste

photons so so we we basically want this

to be a efficient process we would

really like to avoid fluorescence

fluorescence just means that it

reinvents the photon and you just waste

it and so we rather avoid fluorescence

and if the undergoes internal conversion

from this high-energy excited state to a

lower energy excited state we'd also

like this state to not undergo

fluorescence a big much rather have it

go off and do an electron transfer to

activate the initiator so in order to

have fast photo excitation I want to

have the ground state of the

photocatalyst have a nuclear wave

function the open that's a lot with the

new nuclear wave function of the excited

state so back if you can remember your

physical chemistry course that you took

I took many many years ago this is

sometimes called a franck-condon factor

or it is called a franck-condon factor

and the idea is if you're gonna absorb a

photon absorb light then this is like

the symptoms like a simple harmonic

oscillator you need the atoms in the

molecule to basically be in the same

positions in the ground state and in the

excited state when you absorb that

Photon that's the basic idea and so

sometimes the lowest energy excited

state doesn't have the atoms in the same

position as the as the ground state and

so you won't have you won't have

excitation into this state here it will

be into some higher state where there's

a lot of overlap of the nuclear

coordinates here and here

if you then do internal conversion down

in this lowest energy excited state its

nuclear wave function doesn't overlap

very well with wave functions on the

ground state nuclear wave functions the

atomic positions so this can we need it

we need the nuclear overlap to have

excitations but it's good to avoid

nuclear overlap when we want to avoid

fluorescence and so when I mentioned

that we can have photo excitation and

then an internal conversion to lower

energy excited state that can get me out

of a state that has high nuclear overlap

and now this singlet state can have long

lifetimes and wander around and find an

initiator and then transfer an electron

through the activation okay so again we

could this this is helpful to make photo

excitation fast and then fluorescence

this fluorescence from the states slow

another way I can slow down fluorescence

and make my lifetime so my excited state

long is by doing these inter system

crossing so photo excitation and then

flip the spin so if I'm gonna flip the

spin inter system crossing I'm going to

change the angular momentum okay when

you change the angular momentum when we

know in physics angler mention is

conserved in processes right supposed to

be conserved

well spin is angular momentum so

something has to carry away that angular

momentum how do you do that well one way

to do that is

basically it requires an animation that

ice skaters are really good at this ice

skaters when they go into a spin they

are learning to spin they can control

the rate and spinning by coupling

vibration with with the spin so when

they push their arms out they increase

the rotational inertia and slow down the

spin when they pull into a tight spin

they speed up the spin so you can couple

vibrations with spin molecules are

pretty good at this to some molecules so

we pick molecules specifically that can

do this

so that they can have fast intercept and

crossings so if we can go into this

triplet if now what we call spin

forbidden quantum Kathleen can go back

to the ground state a canceled rest it

can go back to nothing but it can

fluoresce

so that's a no that's a second way we

can increase the lifetime of these two

excited state to allow them to go off

and do electron transfers any reduction

okay again we went fast an activation

step and we want to especially a fast

deactivation step all right so let's get

a little bit more into this and this is

a tool if you're gonna do photo photo

photo physical processes photocatalysis

you learn to use tools like this

Jablonski diagram that shows you all the

different processes that you can have

photo excitations fluorescence internal

conversions inter system crossing set

cetera each one of these conversions of

one state of molecule to another and

that's one thing we do as chemical

engineers right that's a big thing that

temperatures do we convert molecules

from one form to another to add value in

this case we're converting one type of

molecule into another are basically by

the same it's a different state of the

molecule so how can we control these

conversions well we need to know the

rules for those conversions and the

rules come from quantum mechanics so for

fluorescence sorry for photo absorption

photoexcitation the symmetries have to

be have to allow the excitation it has

to be spin allowed so the spin doesn't

change

there has to be high nuclear overlap

like we talked about and asked me dipole

couple that is me what the ground state

and the excited States didn't couple

through the electric field of the light

the fluorescence is just the reverse of

excitation so the same thing rules for

internal conversions for example like

these ones we talked about we just went

large nuclear overlaps or if you want to

avoid them we had like so a low E

nuclear overlaps again for inter system

crossing we want vibronic coupling like

our ice skater okay so how do we do this

so we basically need to design molecules

that can satisfy all those kinds of

properties so sufficiently negative

electron reduction potentials you want

molecules that have high molar

absorptivity they absorb light we want

absorb visible light we went slow

fluorescence we want so inter system

crossing Zoar internal conversions fast

electron transfers or fast deactivations

so so we want to design molecules they

can do all that that's a lot to ask but

this doesn't always work alone i won't i

won't go into any of this in detail but

we've worked on a lot of other organic

systems that that we learned a lot of

the photo physics from that we can then

transfer over to these systems and

that's something that we learn a lot in

research right is that when you learn

the physical principles from one system

you can then oftentimes transfer those

principles over to designing or

understanding other systems so this was

like a system that absorbs light and

down converts photons using pentacene

and another system we looked at was

photo i sorry electro chemical reduction

of co2 the fuels again using aromatic

organics like pyridine and it again used

a lot of the same kinds of principles

that we're going to exploit today and

then another system i'm not going to go

into detail but we looked at a different

type of photo polymerization these

molecules that look a lot like our

phenazines

where you absorb light really quickly

and then very slowly over a period of

time let that energy out through

polymerization which is a

in or analogous to the dark reactions of

photosynthesis so a lot of other systems

use similar kind of principles and we we

exploited these okay so so again so

let's let's look at our hypothetical

mechanism and start looking at these

different molecules Co start with our

ground state and excite excited with

light in this excited state and either

do an electron transfer to activate a

PRP and then deactivation this is the

deactivation part or my singlet excited

state can go to the triplet and then I

can do electron transfer to activate

through the triplet a trp mechanism and

then I can deactivate send my catalyst

back to its ground state so that was our

hypothetical mechanism let's see if we

can start working on molecules that

actually do this I'm not going to talk

about this in the interest of time but

there's all kinds of checks that you can

do and examining the photo

polymerization to see if it really is a

TRP and whether the elector weight grows

the time and the diversity comes down

the conversion whether you can turn the

reaction on and off by light and whether

you can do things like block

polymerization so I won't get into that

I want to talk about is starting music

calculations we did lots of calculations

we calculated the reduction potentials

of the excited States we calculated the

dipole coupling of the ground state to

the excited state see if it'll absorb

light we calculated the reduction

potential in the triplet States we

calculated the geometries of the

different states relative to each other

to see how much the geometry of the

molecule changed as I can't transfer the

electrons so this is just one set of

data in this case it's the reduction

potentials for about 200 different

designs that we look at and behind

reduction potentials is negative is 3

volts and then these ones down here

we're not viable I want to be more

negative than negative 0.8 so up here

somewhere

and here's the these phenazine type

molecules they were quite negative

reduction potentials these other designs

that my collaborator is going to look at

looked okay but they weren't very

negative

turns out that the actual candidate he

was looking at was this one which

actually looked didn't look good at all

so we told him don't try that one try

these guys here's that phenazine here's

hawkers beneth isay has a very similar

reduction potential but aren't catalyst

works significantly better the question

was why and then here's that Iridium

catalyst that Hawker discovered as a

really negative reduction potential

doesn't work nearly as well so that was

just one set of calculations we did all

these we looked for all these other

properties and show you some of the

calculations some of them is just what

does the these what are the reduction

potentials but also what do the excited

states look like what is the nature of

these excited States okay so so one

thing here's our this is phenazine and

this is the triplet excited state so

like I have this triplet excited state

if you look at the triplet excited state

and then after I transfer an electron to

do the activation that's negative two

point two four volts a lot of reduction

potential the geometry is almost the

same that's good is we gonna do electron

transfers you don't want the molecules

geometry to relax very much that would

be what's called a high reorganization

energy and it really slows down electron

transfers or it slow down your

activation now the other part of this is

we're going to transfer that electron to

our initiator take the end of our

polymer if I put an electron on that and

don't let the molecule relax it takes

negative 2.6 volts more negative than

this and this this is not permanent

reliable this this electron transfer

will happen but I fixed the geometry I

can do that on a computer in the real

system then this molecule relaxes as it

accepts the electron and it when it

relaxes the bromine comes off as a

bromide leaving group and that's called

an adiabatic electron transfer or an

associative electron has

and it only takes negative point seven

eight volts which is less negative than

negative two point four so this should

happen so activation should be pretty

fast the more important step is

deactivation

so in deactivation here's my oxidized

photocatalyst so after I've photo

excited my catalyst and then transfer of

the electron this is the oxidized state

of the photo catalyst it takes point 1 3

volts to put an electron back on the

photo catalyst to take it back to its

ground state point 1 3 that should be

pretty easy if I try to put an electron

take an electron off of my growing

polymer and don't let the polymer take

the bromide back at the same time that's

point nine seven volts that's not more

negative than point 1 3 so this

shouldn't happen but again the molecule

isn't fixed it relaxes and so if it

relaxes the bromide comes on as the

electron transfers off onto my photo

catalyst then that's only negative point

seven eight volts which is more negative

than point 1 3 so that should happen but

catalysis you can see a little bit of a

problem here we teach this in

undergraduate kinetics usually right

what's the problem

it's a three body event I need to have

my oxides photocatalyst my bromide

leaving group that's gonna cap the

polymer growth and my radical polymer

all in one place at one time

we know three body events are rare so in

kinetics they'll tell you that's a

really slow step that's probably not

gonna happen we want the activation to

be fast right this is not what you would

think you'd do to get a fast

deactivation read by the event that's

not gonna work but what turns out it

doesn't work why is that well it's

because this bromide and the oxide photo

callus this is negatively charged the

oxidized floating health is a cation

this forms an ion pair complex their

Coulomb eclis attracted to each other

and when they do that the bromide is

attracted to the cation photocatalyst

that's bound by about eight Bakehouse

per mole doesn't sound like a lot but

the equilibrium constant of the bound

complex is about 10 to the fifth or so

at room temperature so that keeps them

together and now this can collide with

the end of my growing polymer like a

pseudo to body event and set to be

pretty fast and then the bromide

transfers back to cap off the growing

polymer simultaneous with the electron

transferring back to the oxides floater

panels so no reorganization energies so

what does the what does experiment look

like for these we handed this molecule

and others off to my collaborator we

handed four of them over they were

commercially certain they were

synthesizable in two steps from

commercially available and inexpensive

reagents we kind of lucked out that

wasn't part of our design criteria that

we I'm I'm a beautician but it kind of

lucked out that they just happen designs

that were easy for them to to make

molecular weight control as a function

of conversion was very linear dispersity

was intrinsically low got down to about

1.2 business without any optimization of

process relations in fact when they did

change some of the conditions again not

my optimization but it's trying

different ratios of monomer to initiator

to photocatalyst they got different

molecular weights and different

conversions and they got different

diversities and several of these

dispersants were very promising again

without any optimization 1.1 1.1 eight

one point one eight one point one two

though that's in a very good range again

we haven't even done we've done the

design part of the catalyst but not the

design part of the process yet my

collaborators then took those molecules

and they did block copolymer block block

memorization with them a green polymer

for a while with one monomer when they

ran out of that monomer they introduced

another monomer and then used that big

polymer

to initiate with a TRP the next

polymerization to add another block of

polymer then when that monomer is done

they added another monomer you to add

another block and so they can make these

block copolymers again that was one of

the things that makes a TRP a TRP so

Lecter weight grew with conversion this

version was went down with conversion to

turn on and off of light and you get

this block copolymers ation or you

retain the polymer chain and groups so

it really look like a TRP so while this

paper was under review he noticed

something this is the lower lying

electron of the excited state this is

the higher energy electron of the

excited state these are the orbitals

these two have electron donating groups

out here on the shtetl hoop and both of

the electrons sit on the core of the

catalyst they don't work so well these

two the lower lying electron is sitting

on the core the higher lying electron

sits out here on the phenyl group and

these worked better so he said wow those

working better we should design

molecules that automatically do that so

we picked a couple additional designs

where we put natural groups out here and

these natural groups give us a lower

energy excited electron sitting on the

core and a higher energy excited

electron sitting out here on the natural

group turns out they absorb in the

visible they have nice negative

reduction potentials the dispersity is

1.03 so again without any optimization

we're getting really close to one

somehow these are better I can talk

about that more detail if anyone's

interested but these are better full of

pellets and so what happens is for these

we're both the electron sit on the core

I we call that a localized excitation

when the electron gets excited it stays

in the same general region localized

excitation when I put electron

withdrawing groups out here it lowers

the energy of the orbitals sitting out

here and and so they come down

they come down in energy and my full

excitation is still into a localized

excitation that looks like this but then

I have an internal conversion right

transfer an electron from this part of

the molecule into that orbital and then

this and we call that a charge transfer

state these two electrons are sitting on

different parts of the molecule and that

makes for better photo catalyst so when

we design molecules where you have local

excitations that don't work as well when

we change the polarity of the solvent we

don't see any shift in the fluorescence

but for these charge transfer excited

state that we design when my

experimental collaborators go off and

synthesize them when they change the

political solvent that more polar

solvent stabilizes this charge transfer

excited state and redshifts it and you

can see this without a chromatic effect

that occurs in this register so

mechanism looks like sight to this first

excite to singlet and then you enter

conversion to this charge transfer

excited state and then you can activate

either singlet or triplet HPRP okay just

a couple of other things if we're gonna

design these photo catalysts that do

this charge transfer what about just

having these these are fucked or not -

these these designs have only one group

out here the transfer the electron to

instead of having groups on both sides

of molecule it's only on one side and

when we pick these ions like this we've

got really good nature people look at

how in this particular product this is

called a phenoxy beam letter weight

control is great this person again is

really low without optimization it looks

very similar to hawkers fantasising case

where there's no sulfur here instead of

an oxygen they don't get as good look

and weight control dispersity is kind of

all over the place and just doesn't work

as well so and the reason is in the case

of the finnex Izzie sorry Senate

denizens and kanata's Eanes

they look at the excited state the

oxidized state and the ground state that

are involved in activation and

deactivation for the phenazines

structures all the various

each other same thing for citizens they

look very similar not much not much

geometry change but for this venoth I

see when you have a sulfur there excited

state the oxidized catalyst and the

ground state have very significantly

different molecular structures so

there's a lot of reorganization energy

and that really slows down the electron

transfer for deactivation which means

you lose control over the deactivation

and you get this larger this person and

then one last thing you can do other

things to make these even better

for example if these pendants if they're

not two things you could put by phenyl

groups out here really increase the

coupling to the electric field and get

very strong adsorption and tails that

then go well into the visible and absorb

light very well in the visible and these

work great for protocol

so with that the answers are yes we can

have organic catalysts that request

transition no catalyst we think we've

answered that Organic catalysts can

achieve low disperse these high

conversions and molecular weight control

especially if you design knowing the

mechanism these catalysts can be very

strong reducing agents that's been

careful to not just look at reducing

potential but also these other

properties for deactivation especially

they can be activated by visible light

you can design to absorb invisible we

figured out what this mechanism is the

different steps involved that mechanism

and it exploited that and yes see these

photo catalysts perform really well they

perform so well that even without

optimization they're already competing

with other ATP catalysts that had been

developed over a period of five ten

fifteen years or so so if that I'd like

to thank you and thank our funding

sources and then let you know that you

can come and visit us in Boulder

we have several faculty positions open

probably for the next several years for

those postdocs and graduate students

will be graduating next few years you

hope you apply to our open positions I'd

like to thank you for your attention and

taking questions

[Applause]

yeah so there's a so it's kind of the

LEDs it's kind of the reverse process a

lot of it's the reverse process for the

chromophore it's a lot of the quantum

mechanical rules that govern whether it

works as a as an LED vitamin er are

similar to what you need for Toma for

now some of them are what's the right

way to put this some of them are the

opposite of what you need but very

similar photo physics goes into it

because a lot not all of it there's a

lot of it's the reverse process so yes

that's a good observation and act some

of the some of my students and other

people we've talked to it basically says

like well why don't we propose not only

due to these organic photo redox

catalysts

but let's also propose to do the reverse

process where we can do very similar

engineering and design for those systems

yeah

we have it so the vibronic coupling hey

you know those are a kind of pain in the

neck to calculate and my student was

getting ready to send his thesis and I

pushed and pushed and pushed and I try

and I and I pulled and pulled and pull

and I he you is like no I need to

graduate and those are really hard to

calculate so unfortunately those are

left undone so at this point we're

assuming that that's what's happening we

pick designs to explore in the first

place that we knew would likely have

that my bonnet coupling so in some ways

he was telling me is like well these

these kinds of heterocyclic aromatics

naturally do this vibronic coupling

you're gonna make me do all these really

hard calculations and learn a lot I was

annoyed bad part learn a lot and and

really not get to any better design we

already know they're gonna work and they

and they the other thing was experiment

or any shoulder work to so he was not

very motivated and he he went off and

did his post talking it's it's they're

undone

ah yeah that's an interesting question I

mean some of these systems do you know

pan breakdown they're exposed to light

and oxygen at the same time for example

is one thing that's a degradation

process for some of them we don't really

look at degradation processes in general

but they're stable enough to work with

and these molecules I mean they're

stable enough to sit in grocery store

it's very similar molecules and and the

edible for some time and we see similar

things we don't have to my collaborators

don't do too much in terms of special

treatment of these systems they do

remove oxygen in the experiments they

remove the oxygen because the oxygen can

quench these excited States and you just

end up with very low conversions because

they're fighting against oxidative

reductions that quench these guys but in

terms of degradation they I think they

do pretty well I'm but again I'm a

theory so I don't spend a whole lot of

time examining those more practical

aspects I shouldn't have an engineer

yeah so they do they oftentimes do this

and meet some of them sometimes they've

done it meet monitor but they've also

done it in solution they do it both ways

it could be I mean there are parts where

you end up getting into transport

limited regimes where either by

molecular termination or the excited

States wandering around when you get to

high conversion so it's like wandering

around trying to find a monomer that

hasn't reacted yet it could be pretty

slow when you get to very few monomers

left they've done both my collaborators

have done both meat monomer solutions

but also diluted so so for example

electron transfers to roots on the

growing polymer

yeah yeah that's yeah we've wondered

about some of those we haven't really

seen anything specific for those kinds

of side reactions you don't see evidence

of products of weird side reactions but

you're right there's a there's there's

quite a bit of reduction potential and

there's a lot of reduction potential who

knows which reactions can be now active

but it might be that some of those

processes are kind of dead ends they you

transfer an electron and they don't

actually go anywhere one of the side

reactions that my my collaborators found

later was that some of these radicals

will come back and attack the rings of

the photocatalyst and so you can

sometimes get your monomer oh sorry oh

yeah you're growing polymer to attack

the the ring of your photocatalyst now

the photo catalysts tend to be pre low

concentration but the same concept

activated form the concentrations are

similar to the growing polymer and you

know that that's one of the minority

products you can get yes my my clap it

has relatively low quantum efficiency

overall in terms of how many photons

actually end up going into making

polymer but it's but it's relatively

high in terms of other photocatalysis

systems or photo polymerization

so so yeah so in these two cases here if

you're you know you have ring systems

and so in these polycyclic aromatic s--

those rings are relatively rigid that

can help with limiting the amount of

reorganization but this is the ring

system as well but with that sulfur

there that adds a lot of strain in the

molecule and that strain that strain a

big sulfur sitting in that in that ring

basically lowers the energy of some of

these distorted states relative to the

planar state and so one is to try to

stick with and heterocyclic aromatics

that it cuts to only things like from

the first row or boron oxygen nitrogen

etc so first something you want to avoid

to kick house per mole reorganization

energy is is pretty good we love it if

it were lower than that but that's but

these particular designs that's about as

low as you can get you can imagine other

types of reorganization but we don't see

it these rings you can imagine they

might rotate some afternoo an electron

transfer to create the charge transfer

excited state but we actually we

actually didn't see that this this is

still kind of orthogonal to the core of

the molecule but but generally you want

relatively rigid molecules and that

seems to work mostly but it works so

well in identify z8k cows so 8k customer

will take a small for this women this is

deactivation going from two point three

to four point one doesn't sound like

that much right that's only one point

eight may cost a little more

reorganization energy so a trick to you

learning kinetics is one point three

seven cake house per mole is a factor of

ten and a rate constant at room

temperature some 1.37 this is one point

eight cake a little different the one

point eight divided by one point three

seven is like one point three one and so

ten to the one point three one is like

twenty point six so that's twenty point

six times slower if that's the

difference in energy about twenty point

six times slower because I have a one

point eight kcal per mole so

so it's really hard to control things

like realization energy when you're

looking at differences that small so in

some cases you get get kind of a little

bit lucky we we would like to say that

we designed purposely for that but we

kind of lucked out and those designs

just happen to have both your

organization