Importance of Natural Resources

Professor Yi Cui: Nanotechnology for Energy and Environment


[MUSIC PLAYING] YI CUI: Well, I feel
I’m very lucky to be– or to spend 14
years of my career to be in the first hundred
years of the department history. I’m also very excited about
the next number of decades I can go into together with
department for another hundred years. It’s very hard to
follow Jen and Bill. They give very exciting talk. They already consumed
my energy level. So I’m going to speak
about energy, so I need to pump up a little bit. I’d like to share with you about
the nanotechnology for energy and environment. I remember when I applied to
our department as a faculty. The hire was on computational
materials science. But the department didn’t get
any computational material science in that year. The department got Alberto
and me to join in the faculty, so I feel a sense of very lucky. So around the time I
joined in the department, my initial proposal wasn’t
that much into energy at all. But there was a time in 2005– this is a big transition
the society recognized. These are the problem
we are seeing. We need clean energy. We also need to clean up the
environment at the same time. So I was thinking, you
know, how do I do that? How do I contribute? So I was playing now. Even though I haven’t done
anything in these areas yet, I move nearly full speed
into energy and environment. So fast-forward 14 years,
this is program we set up. We work on solar,
work on batteries, work on catalysis, air
filtration and water filtration, and so on. But today, let me just
pick a few examples and, really, in the spirit
Bill just mentioned, it’s an era of engineering
with materials typing now, very exciting opportunity. First example I want
to share with you is reinventing batteries
with nanotechnology. In the morning, when Craig
Barrett was sharing with you the Moore’s law of semiconductor
industry, the chip, and then we look
at our every day, you know, life, the
experience we have. You all remember this. The first of these mobile
phone coming out is so big, and then it’s marching down– smallest, smallest size. Of course, it’s multiple
technology going in there, you know, more powerful computer
chip, a wireless communication. But at the end of the
day, what’s limiting the size of your phone? I know the chip– yeah, the batteries. The first one is using lead
acid batteries– so heavy. So with the nickel
cadmium coming in, you can make smaller
and smaller phone. But we see the opposite
trend in the last 15 years. The phone is getting
bigger and bigger. [LAUGHTER] This reason behind it,
you know, you probably want a bigger screen, but you
wonder why– it’s actually also battery driving the size. If you open a phone– this is iPhone 6 Plus– you see that big
blob right there. That’s a big piece of
batteries, that it’s much bigger than any part of the phone. And you say, well, you know,
looks like from iPhone 6 to iPhone 11 right now, the
size doesn’t change that much. We have more
computational power. What if– from outside,
doesn’t change. Actually, inside change. You see the L shape right there. Now, Apple tried to put in
more batteries into the phone, so we are really power
hunger at this moment, really power hunger. I hope one day, by the time I
retire, maybe the battery size, we’re getting closer to
the computer chip size. That’s the goal. That’s the goal, by
the time I retire. So now you see
nowadays, you know, batteries are used in the mobile
electronics, in the drone, in the electric car. We are talking about this
for the stationary storage– this big industry
driving behind this. Let me search to
ask the question. Oh, yeah, before I
go into that, let me mention a little
bit of history. So you’ve all seen this– Bill showing the Nobel
Prize was finally given to the three
gentlemen for developing the lithium-ion batteries. By the way, John Goodenough
is 97 years old this year. Stanley Whittingham our alumni
from the department in ’78. This study actually is
just to be for you to know, and proceeding on National
Academy of Science to show, and having a little
bit of lithium, and you intake a little bit of
lithium, you can live longer. [LAUGHTER] I think this probably explain– you can go for a long time. You know, between
William Chueh and me, we are working on
lithium batteries. There’s probably a
little bit of lithium spread around the building. I think the whole department
can get much longer, you know, on average, but we’ll see. So the connection right there,
of course, Bob right there, spin out a few, a number of very
talented registered in postdoc. So when I join their
faculty, I actually didn’t know the history
of the department at all. This is all in retrospect. Well, I’ll tell you my
version of the story. So when I join the
faculty, up till now, we try to think about, how do we
adjust the grand challenges in the energy storage areas? The question we ask is,
how high energy batteries can go, the energy density? Yeah, that given size, given
weight, how much in more energy can you store? Can you extend a battery
life by three times or more? Instead of 7 or 10 years,
let’s go for 30 years. That’s very important. What about fast charging? Can we do it within 10 minutes? What’s the fundamentals
limiting this process? Can we make the batteries
completely safe? Once in a while,
you’ll see the battery catching fire in an explosion. What’s the reason behind this? Can we come with a
solution to solve that? Can reduce the cost
by three times? Then the whole– the electrical
car will completely fly, you know, low cost enough. How do we sense the battery
inside-health condition? So we know the bad things are
coming, and let’s stop it. And how do we do battery
reuse, recycling? And then how do we do rescale– larger-scale energy storage? I mean these are
the question I think we are sending to
materials scientists with the great opportunities. Of course, they are very,
very tough to solve. So I do not know it now,
so I fast-forward 14 years. So we work on all
type of problem. Let me just pick one. The first one we ask
is how high the energy density of the battery can go. If you look back at history,
the Nobel Prize-winning work is where I find it took
place with lithium, right? There’s lithium cobalt oxide. There’s all this metal oxide
intercalated with lithium. So after about 28 years
of commercial history of lithium-ion batteries, we
are seeing close to limit. And graphite-intercalated
lithium– can you use silicon
to store lithium? Turned out to be in the 1980s– Bill showed that,
Bob Huggins’ paper, the phase diagram indicate
lithium and silicon can alloy together. So when I joined, in fact,
that I didn’t know this at all because I didn’t work
on batteries back in 2005. It just, one day, I
saw some literature and seen Bob’s paper of
lithium and silicon can alloy. That sounds exciting. That could be useful
battery, I told my students. Then I started to work on this. And also, same thing
for the cathode– I mean, this is the
framework of the cathode, through intercalation, lithium
coming in and out reversibly. But could you use
a new material that can hold a lot more lithium,
such as sulfur, very low cost? We don’t know how to do this. We don’t know how
to do this well yet. So it all come back
to this new material, if you can make the
new materials to work. We are really looking into
energy density going from now, about somewhere around 250
watt per kilo, up to 500 or maybe even about 600. We are talking about
double and maybe triple of the energy
density down the road if you could make new
materials to work. But the problem, it’s really
challenging right there. For the past 28
years, these hosts, they will host material used in
your cell phone, in the laptop. The lessons learned is there’s
no chemical bond breaking. But new materials, because so
many lithium ion coming in, they’re going to break
the chemical bonding. Host atoms move
to long distance, complete structural change,
big volume expansion from less than 10% to 100% or
even 400% volume expansion. So we need to design
the materials to take on those challenges,
to be reversible in and out over a thousand
cycles, a few thousand cycles. So speaking of silicon– now come back to silicon– silicon was the first
materials I picked in work with my students. And silicon uptake
11 times more lithium compared to the graphite,
but the volume expansion is about 400%. That’s too much. It caused mechanical breaking. It caused instability of
solid electrolyte interphase, so your battery life
is very, very short. I started this project. This is my first battery with
graduate student Candace Chan. I also see another student. I see David Schoen in
the audience right there. Maybe there is a– the first batch of
student, I had four. I didn’t know Stanford
students cost so much, by the way, Craig, at the time. So I had the four students. I started to see financial
pressure right away. So this is– certainly,
the price show, and publication and
financial all come together. So I didn’t know all of this,
but, anyway, fast-forward, we started to think about,
how do we come out with new solution for that? And my training
in PhD and postdoc is for nanomaterials, nanosize. I know how to make
nanomaterials very well. I told Candace, I said, let’s
make some silicon nanowires. That probably can overcome the
volume expansion and breaking problem. So we did it, and
almost immediately, we saw the effect right away. So you can put lithium in. You can take lithium out,
reversibly, multiple times. That’s really the
landmark paper– get my group going
into the battery field. Now let’s look at how
challenging this problem is. Apparently, this
mechanical breaking issue involving a lot
of mechanics thinking. And we need a tool
to study this. At the time, I know very little
about mechanical property. Luckily, we have Bill
Nix in the department. So I went off to talk to Bill. I say, Bill, teach me a little
bit about the mechanics. So that’s how I
started to learn. We used this technique. Matt was the student
developing this technique, who is now faculty in Georgia Tech. I think he’s in
audience somewhere. To really start to put
lithium in and out of silicon, in the same pump,
just watching what’s happening inside
electron microscope. We can put the particles right
there– lithium coming in, electrons coming
in– we could learn. I want to show you one
of the essential video. This silicon nanowire has
diameter about 200 nanometers, with a copper coating
on the surface. Now, lithium coming in, you will
be able to tell– the silicon expansion is a lot. This force is so strong, it
break the surrounding coating. But silicon nanowire
diameter is small enough, it does not break. It does not break. But if you will take a
nanostructure of silicon, like the one in the middle– this big one is about 800
nanometers in diameter. Lithium coming in,
a lot of lithium going in– this
particle is expanding. And then it start to
accumulate a lot of stress. So after a while, this particle
cannot really hold on of its materials any more. Then it’s broken. So this explain–
in the early days, people tried to make silicon
to be usable as battery, but very, very challenging. If you have breaking
happening, materials lose electrical contact,
and then your batteries die the first time
you charge it. That’s the conclusion,
you know, in the past. Now, using nanomaterials,
you could hold on of these materials. They don’t break. And also, using a
technique like this, we can identify what will be
the critical breaking size. Below that size, silicon
will not break any more– roughly about 150 nanometers
for the nanoparticles. So this turned out to
be providing a guiding light for the whole
industry to use, how small particle size they
would like to have in order to avoid breaking. So now this has huge meaning to
the whole batteries industry. People try to put silicon
into the battery electrode. So starting from
generation one, we have been through about 12
years now of a learning curve. So we construct
core-shell nanowires. We do the hollow structure
to relax the stream better. And then we also
know how to utilize double-wall hollow structure to
build stable solid electrolyte interphase, force this
blue-color silicon to expand to what’s inside instead
of going outside. If you have expansion
to what’s outside, even though it doesn’t
break, but it’s facing organic electrolyte. You’d never have a
stable interphase to build the SEI layer,
so the batteries also die. So we learned quite a
bit in the past 12 years. It’s now 12 generation of
design we have been through. We are about to stop right there
because this 12 generation, I think, is really
getting close to solve all these material design
principle right now, and let the industry take over. 2008, I founded
my first company, and not knowing also how
hard that is to do a company. In 2008, I founded Amprius, and
we raised about $150 million by now, but a lot of
money, too, until nowadays. Well, luckily, we still survive. We now have the product on
the market, just a production line 1 and line 2–
particularly in line 2. It’s produced right
now in Fremont. We have the highest
energy-density batteries in the world, you know,
rechargeable, and getting to about 430 watt per kilo, and
1,200 or 1,300 watt per liter, roughly 80% of
the energy density of the current technology. We’ve delivered that product. And this is the Airbus
drone, the Zephyr S, using our battery’s
high-energy density, and fly in the sky 25 days,
and no stopping, just during the day using
solar cells to charge up the batteries–
during the night, keep using the battery
to power this drone. This elevation is very
high, 78,000 feet– very high elevation. I’m glad to see now
silicon finally make it into the real applications. And now the whole
industry is working hard, try to use the silicon
in a larger scale. Of course, it’s still not easy. You know, going to
industry, you need to still solve many,
many challenges I don’t have time to go into. So what’s after silicon? And three years ago,
White House announced a Battery500 Consortium program. It’s tried to make 500 watt
per kilogram of the batteries. Stanford, as a partner
institution in SLAC, and team up with the PNNL– we have a team of
people right there. And the public doesn’t
know, our consortium’s doing better than ours. We have two Nobel prize winners
as PI and in the consortium. So try to adjust the challenge
of 500 watt per kilo. But in order to
get that to work, we have to look at
lithium-metal anode. If you look all the way
back to the Bob Huggins, then everything have its time. In the late 1960s to
beginning of 1970s, that was the time people tried
to make lithium-metal anode to work. There’s no host materials. By 1980s, there’s
a company coming up with lithium-metal
batteries sold commercially. Soon after that, the
batteries catch a fire, explosion took place, so
the company was closed. And nobody dared to touch
lithium metal since then in the commercial space. But we are running
out of materials. We need to look
at lithium metal. So now your states start
to have these lithium platings stripping
during the charging, in this charging process. How do we handle that,
make it reversible with high efficiency, without
growing all these dendrites? Let’s revisit that. In this perspective
paper, we wrote down, we said, well, lithium metal,
what’s the root causes? Number one is high
chemical reactivity. Number two is still
lithium plating. You’re creating this
volume change in infinite– in the reality of sense,
it’s infinite volume change right there, going to a
plating state or empty state. So every 1 milliamp
output centimeters squared capacity unify
micron on lithium, realistically, you need 15
micron per layer of lithium, you know, plating, stripping. How do you handle
that volume change? These two couple
together creating all the surrounding problem. So we need to solve
the root causes. Then we come up with an idea. In 2013, Steve Chu
come back to Stanford after stepping down from
Secretary of Energy position. He and I talk about this. We say, we need to solve
lithium-metal problem. There’s a few breakthrough
coming in, particularly in 2016. We say, well, what did
we learn from graphite? Graphite is the host material
to intercalate a lithium ion. We do want to have a
host material which you can embed in metallic lithium– in there. So we construct seeded
hollow-carbon host, graphene-oxide host. So if our host structure right,
then you put lithium metal in, the volume is holding there. You take lithium out. It will not collapse back down. You’re really holding that
structure right there. You overcome the
volume expansion issue, really holding there. And then you can build
stable interface. So we showed that. It’s in these two publication. And also, we also
think now, you know, the interface to control
chemical reactivity, it’s very important. We come up with a number
of different ideas to enable stable interface. Let me highlight one. This is the collaboration
with Zhenan Bao in Chemical Engineering. Turned out to be, we can
apply self-healing polymer, very dynamic polymer,
in the interface. So this dynamic polymer, if
you want to break this polymer, it has hydrogen bond. It can reform. So what it means is you want
to shut out lithium dendrite. To break that polymer,
it’s actually very hard. This polymer is
flowable, keep covering the lithium-metal
surface without expose the hot spot to the electrolyte. So we saw quite a
bit of improvement along this direction. Well, we are not in
the prime time yet to say we solve all
the problem now yet. We still have the
next few decades– we probably need to work very
hard to solve this problem. So along with all this
material development, and then my group keep wondering. I wondered myself– you
know, lithium batteries has been commercialized,
but nobody have really seen atomic-scale resolution
on metallic lithium. We know battery materials, a
lot of them are very fragile. You put it into a
TEM, they’re gone. It will being
damaged right away. So about three years ago,
apparently, you know, we noticed the
biology community, structural biologists
developed a cryo-EM to dissolve the protein
crystal structure. That caught my attention. I talked to two of my student,
and I say, why don’t we borrow this technology
from biologists and develop that towards
the battery materials. So we start to work on this. Yuzhang and Yanbin are
two graduate student. Yuzhang is about to join in
UCLA as an assistant professor. We poured the lithium, and
then plunge in liquid nitrogen. You know, liquid
nitrogen is really cold. All the side chemical reactions
stop at that temperature. Without exposed to air, we
developed a cryo chest for it and transfer into the TEM. So that is not trivial to
develop this technique, and it took us a while. I always put this paper– if you’ll notice, we
submitted 2016, December 15, and get accepted
2017, September 14– nine months reveal, nine months. The first time I show
atomic-scale resolution imaging to people, people
wouldn’t believe it. They’d say, you can get
atomic-scale resolution, metallic lithium? So it took us three rounds of
reveal to convince and reveal that we have seen
the real thing. Now, let me share with you. First of all is
those tests, right? The left, or top left,
is the cryo-EM image. The top right– those
were on the black hollow. It’s whole during right there. That’s because, on the
regular TEM, if you zoom in, you damage your
sample right away. It’s absolutely true. We do the dose testing. And in the lithium
metal, it’s very stable in the cryo condition. You can do 10,000 electrons
per angstrom square of imaging before you see the
obvious damaging. So it’s about, you know,
1,000 times more robust than the protein molecules. And we confirm it’s metallic
lithium using [INAUDIBLE].. And this is
atomic-scale resolution, thanks to Bob’s leadership,
bringing Titan to Stanford. We have Titan. We purchase a cryo holder. We were able to see atomic-scale
resolution on lithium. So not only that,
using cryo-EM, you know, we found Jen’s
presentation you have seen, cryo-EM, it’s very powerful
because you can stabilize very fragile materials. Using cryo-EM, we could solve– but one of the long-outstanding
problem in the battery field is about a solid
electrolyte interphase. What’s the
interfacial structure? This propose it’s a mosaicking
of the layer structure, but nobody really
have seen that. So we were able to– you know, on the one
electrolyte condition, you can see a lithium metal,
and this, that 20 nanometers, below the white-colored line. That’s the SEI. So we see inorganic
particle embedded into amorphous matrix– sounds like a matrix model,
mosaic type of model. But as soon as you
put an electrolyte, you add in
fluorinated carbonate. We see the multi-layer. You see bilayer one. It’s an inorganic
coating in the outside. Using a TEM like this, we
were able to also correlate this structure with
electrochemical performance. So this is cryo-EM. In the past, about,
14 years, there’s a lot going on at
Stanford campus. We have Will Chueh
joining the department, and we successfully launch
a very exciting initiative last week. I could see that this is
also a gift to the department for our 100 years
birthday, as a gift, is launching this major
initiative called StorageX. And Will and I serve
as the co-directors. And this is the program– tried to bring in significant
industrial support to support each research group,
support graduate student, having a very visible platform
to put Stanford globally on the map of energy storage. So these– another
reason to start this is because, for the
battery research, it’s opportunity from both
the landscape and timescale of doing research for
materials scientists, as well as for other
engineering department to work together to solve a
big problem for materials all the way to the system level. We also have already
Critical Mass, and crossing this multiple landscape. So we feel like this is
a great timing to do so. I probably have a
few minutes left. Let me just mention
one small topic to you, but a lot of fun, that’s also– in a sense, it’s also
engineering these materials. It’s a new type of textile. I would never imagine
when I joined the faculty I would work on textile. I thought this is
very old technology, really nothing there. But that not to be. That wasn’t true. This is in collaboration
with Shanhui Fan in double E department. About four or five years ago,
ARPA-E and the Department of Energy asked
us this question, and it’s saying, how do we save
building energy consumption? Every year, we spend 13%
of energy, total energy consumption, just for heating
and cooling the building. Our air conditioning is
set at 22 degrees Celsius. In the summertime, if we allow
this air conditioning sampling to go up, wintertime to go
down, we save a lot of energy. Every 1 degree Celsius change
on sampling, we save 10%– 2 degrees Celsius, we save 20%. The DOE is asking for, can
we do 2.5 degrees Celsius? Then we save 25% of the energy. So now the question becomes,
instead of cooling and heating the building, can we cool
and heat individual person? That would be very nice. Everybody’s carrying an air
conditioning on your body, but make sure it’s
very lightweight. So then we say, well,
how do we do that? Each person is about 100
watt of a light bulb. That’s the power we consume. So we have a few hundred
light bulb right here, you know, giving out
energy at this moment. But let’s look at how
we dissipate the heat below 30 degrees Celsius. This gray color curve right
there, it’s radiation. About 40% to 60% how we lose
heat depends on radiation. And then there’s
convection coming in, and then there’s evaporation. If we are doing exercise,
having sweat coming out, evaporation will pick up. So in the office
environment, we really need to work on radiation. Well, it turned out to
be in the textile field. There’s not many idea
to work on radiation– maybe one or two
brand right there, not really working that way. After I test their textile,
no, it’s not working well. They are just doing
this marketing only. But let’s look at,
how do we do that? Let’s look at cooling. In the summertime, we
want to do cooling. You want your infrared
radiation to go out. Human body radiation is
between 4 microns or 20 micron. And our body temperature
peak around 10 micron. If you look at all the
clothing, everybody here, you are wearing. They are not transparent
with the infrared radiation. They are not good for
this radiation cooling. But can we find the right
polymer that could do it? We do have that polymer
right at the bottom right. Guess what? It’s very low cost. In everybody’s kitchen is your
kitchen wrap, polyethylene and polypropylene. If you look at
this fraction right there, that green
curve right there, you only have a very
small absorption peak. It’s mostly transparent to
human-body infrared radiation. But these polymer,
they’re not wearable. What’s the biggest problem,
the first problem, you see? They are also transparent
with the visible light. [LAUGHTER] That’s the biggest problem
we need to solve first. I mean, don’t think
this is trivial. It’s not trivial because
polyethylene is so inert. Well, I learned all my photonics
knowledge from Mark Brongersma and Michael McGehee
when, at early days, I joined their faculty, and I
didn’t know much about photons. So learning from these two, I
have a little bit of background after that, and then
later with Shanhui. Then we said, well,
let’s engineer nanopores into the polyethylene. Well, turn out to be,
engineering nanopores, certainly the idea
is the pore size is a similar size
as the visible wave, like, you can scatter
all this we supply. Infrared can still go through. But polyethylene is
very hard to process. It’s been now– it took
us a year to figure out, how do we do that? Actually, within a year, we
couldn’t figure that out. Just one day, I talked
to my student, Po-Chun. I said, what? This material exists in my lab. This is just a
battery separator. This is the polyethylene
used in the batteries to separate cathode and
anode to prevent shorting. I asked Po-Chun, I
said, why don’t you go measure the battery separator? It’s probably the coolest
materials in the world. So we did that, and then you
look at the infrared light, still now about
90% transmittance, but visible range, 0.4
to 0.7 micron is 0. Now, it’s opaque. It’s white color, but still
transparent to the IR. And we measure how cool that is. In the summertime, in
order to be the coolest, you wear nothing. That’s the coolest way. That’s the bare skin. Nothing can beat bare skin. But as soon as you put a
cotton thin layer right there, your temperature go up to
37 degrees Celsius, the skin temperature. But our NanoPE go
up to only 34.3– 34.3. We have about 2.7 degrees
cooler than the cotton. Then we save close to 30% of
energy and building-energy consumption. So this is exciting. Then we ask, well, can we invent
something that’s even better? We can wear it in the summertime
with this, and also wintertime. And in California, the
temperature– the outside is so big between day and night. Let’s do bifunctional. And we actually invented
a dual-mode bifunctional clothing. That is, you changed in a
series this bilayer coating right there– one side emit– have
high emissivity. The other side has
low emissivity. So if you want to have
cooling effect happening, you let the high-emissivity
facing outside. You emit the IR a lot faster. And then, you want
to do heating– goes to the restroom
and wear inside-out, so you will go to heating. We showed this is possible. You can go out to
a bar, you know, 6 degree, 7 degrees Celsius,
in that type of range of delta. I mean, the best we can do,
probably go to 9 and 10 degrees Celsius, perfect
for California– one clothing during the
day, during the night. Oh yeah, it can all work. So let me end my talk, just
with the summary slide. It’s now the lessons
I’d learn is, if you discover a problem
you want to solve, and you can sometimes
come back to all the way to the materials. You change the materials. You tune it. You can adjust that problem. This one was carried out by a
big group of graduate student and postdocs, and many
of them have gone now to become very successful. And we’re looking at the past
100 years, the alumni we have, I think this is this
department’s tradition, very collegial, very supportive
to each other, to students, to colleagues. I believe this will continue
for the next 100 years. At portal, you will have a– you don’t need to do much
to maintain this, right? We will just keep going. Yeah, hopefully, that’s easy. Thank you very much. [APPLAUSE]


Leave a Reply

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