What's the deal with electrolyzers?

In this episode, Raffi Garabedian, CEO of startup Electric
Hydrogen, discusses all things electrolyzer, the current hydrogen
market, and the future risks and opportunities for green
hydrogen.


(PDF
transcript)


(Active
transcript)


Text transcript:


David Roberts


Volts subscribers are likely well aware of the fact that a fully
decarbonized energy system is going to require an enormous amount
of hydrogen to fill in the gaps left by wind and solar. What's
more, they are probably aware that hydrogen comes in a dazzling
variety of colors, from blue to gray to brown, depending on the
carbon intensity of the production.


In the end, though, only one such color matters: green. That is
to say, a fully decarbonized energy system is going to require
lots and lots of hydrogen made with renewable energy, with no
carbon emissions. The way to do that is to run water and
electrical current through an electrolyzer, which splits the
hydrogen off from the oxygen.


Currently, about 95 percent of the world's hydrogen is made using
fossil fuels. Green hydrogen — hydrogen made with renewable
energy and electrolyzers — comprises only a sliver of the
remaining 5 percent. Yet it’s going to have to scale up to 100
percent in the next several decades, even as demand for hydrogen
rises.


This is all a familiar story, at least to energy nerds. But if
you're anything like me, the more you think about it, the more
you realize that, despite the key role they play in that story,
you don't actually know very much about electrolyzers themselves.
What are they, exactly? What do they look like? How can they be
improved? What policy is supporting them?


To talk through these questions, I contacted Raffi Garabedian,
the CEO of Electric Hydrogen, a startup that has set out to
rapidly drive down the cost of green hydrogen. Garabedian, who
was previously chief technology officer at First Solar, believes
that the market for green hydrogen today is roughly where the
solar market was in 2008, with all the attendant risks and
opportunities.


Garabedian (quite patiently) walked me through the basics of
electrolyzers, the current state of the market and the
technology, the kind of cost improvements he believes are
possible within the next five years, the increasingly supportive
policy environment, and the future of green hydrogen.


With no further ado, Raffi Garabedian, welcome to Volts. Thank
you so much for coming.


Raffi Garabedian


It's great to be here, David.


David Roberts


I'm excited to talk today about electrolyzers because I think I
am, and I think probably most of my listeners are already
convinced that hydrogen is going to play an important role in a
decarbonized electricity system. I think we can just assume that.
And I already think, and I think my listeners probably already
know this too, that in a true decarbonized system, it's going to
have to be so called "green hydrogen," hydrogen made without
greenhouse gases. I know there are 50 other colors made from
different other things with varying levels of greenhouse gas
production. But, I think—and obviously you think, since you've
predicated your business on it—that we got to make green hydrogen
work.


And green hydrogen is hydrogen made with renewable electricity
and electrolyzers. So, we know all that. But I find that when I
think about the technologies involved, I have a pretty good
understanding of all the pieces of that puzzle, except for
electrolyzers. They're just kind of this thing that plugs into a
certain spot in the diagram. But I find that when I actually
focus in on it and think about it, I turn out to know very little
about electrolyzers. So I'm very excited to have you on the pod
today because I suspect I'm not the only one who has that sort of
gap in my knowledge. So maybe we can just start with: what is an
electrolyzer?


Raffi Garabedian


Yeah, let's start there. Let's first start by just exploring and
defining the problem that we're trying to solve, right? So we're
trying to make hydrogen, which is both a feedstock and a fuel,
and we're trying to make it using renewable energy. So how does
that work? Well, it all starts with the water molecule. So
everybody knows water is... what's the chemical formula for
water? It's H2O. So think of that as oxidized hydrogen. Hydrogen,
the word, actually is derived from hydro: water, gen: produces.
So when you burn hydrogen yeah, interesting, right? When you burn
hydrogen, you get water as a result.


So burning is oxidation. So what an electrolyzer does is the
opposite of that. It's the reverse of oxidation, which is called
reduction. But it does so electrochemically. Now, what does that
mean? Electrochemistry is a whole field of science and technology
that involves the interaction between chemical reactions and
electricity. And generally the kinds of electrochemical systems
that are used in industry involve things like membranes and
electrodes. But the function of these devices is to drive some
sort of a chemical reaction that requires energy towards a
desired end state. Now, in this case, what we're trying to do is
drive water, which is burned hydrogen, oxidized hydrogen, back to
its original state, right back to hydrogen's original state,
which is H2, which is just the diatom of the element hydrogen.


David Roberts


You're unoxidizing it.


Raffi Garabedian


Yeah, you're unoxidizing it and and to do—you're unburning it.
And to do so requires a tremendous amount of energy. That's
because the reaction of hydrogen plus oxygen equals water,
releases a lot of energy in the first place. It's, as they say,
energetically downhill. So to go the other direction, we have to
pump a lot of energy into it. Now, that's the good news. It's not
bad news in this case. The good news is: we are able to store a
lot of electrical energy as chemical potential energy in the form
of hydrogen. And then we're able to extract that either as heat
by literally burning it like you would any other fuel or through
other mechanisms. For example, you can run hydrogen through a
fuel cell and you can get electricity back out again round trip,
just like a battery.


David Roberts


So all that energy you're using to break up the water is
effectively stored in the hydrogen, and you're getting all that
energy back out when you...


Raffi Garabedian


You got it. I've heard people say, "Oh, well, electrolysis is bad
because it's so energy hungry." Well, that's exactly the point.
It's energy hungry because you're trying to convert, transform
that electrical energy from electrical potential into chemical
potential, right? That's what an electrolyzer does. So an
electrolyzer is the machine that does that, and it's generally
got a bunch of parts to it. The heart of the machine is this
thing called an electrochemical stack, typically. And that's a
bunch of plates, layers of plates, and they're literally stacked
on top of each other, which is why we use that term in the art.


But you flow water through this thing, you pass electricity
through it, and what comes out is oxygen on one side in one pipe
and hydrogen on the other side in another pipe. It's got an anode
and a cathode. And that's where you get the two gases are
produced on either side of that cell. Now there's a bunch more to
it. There's a power converter that delivers the electric power to
the device and controls it. And then there's a bunch of piping,
plumbing, valves, control systems, gas, water, separators, whole
bunch of things wrapped around it, which we usually call like the
balance of plant. Electrolyzers, historically... they've been
around for a long time.


David Roberts


The metal that the plates are made of is something reactive,
right? Such that when you introduce electrical current to it and
water to it, it causes the chemical reaction in question, like
what do we have other than water and electricity?


Raffi Garabedian


Right. There are different technologies for electrolysis, but
they all involve a thing called a catalyst


David Roberts


Right.


Raffi Garabedian


And a catalyst is typically, in these cases, metals or metal
oxides, which, well, like the word implies, catalyze the
reaction. So they sit there, they don't get consumed, but they
play a critical role in facilitating the reaction to occur. You
can make water break into hydrogen and oxygen without a catalyst.
Some of your listeners might remember high school chemistry
class, right? You have like a little beaker of water. You put
some stuff in the water to make it conduct electricity. You put
two wires in it, connect it to a battery, and you see bubbles.
Well, bubbles on one wire, oxygen. The other wire is hydrogen.
And you can collect those gases. But that's a very inefficient
way to electrolyze water. To electrolyze water efficiently, you
need a very special kind of a catalyst. So those are the metals
you're referring to. So they sit there, they're part of the
construction of the device. They don't get consumed in the
process, but they play a critical role in the process.


David Roberts


Right, so the catalyst is the same, whatever, after a year of
production as it was at the beginning. This is not something that
is breaking apart in any way or declining in any way.


Raffi Garabedian


Well, ideally, that's the case. Now, with any practical device,
there are breakdown mechanisms, degradations, all sorts of
practical things we technologists and scientists think about. And
nothing works forever. Everything in the world breaks down. But,
yeah, the first order, what you said is generally true. They're
participants in the reaction, but they're not consumed in the
process.


David Roberts


Okay, so you got these basic parts. What is the scale of this
thing? Like, what is the smallest electrolyzer you could build?
Or the biggest one? Or what is the typical one look like? Is it
bigger than a breadbox? Like, what am I looking at if I'm looking
at an electrolyzer?


Raffi Garabedian


Yeah, let's go back in history a little bit, okay? So I think...
forgive me, my dates are probably off a little bit, but I think
the first electrolyzers were built in Norway, I believe. And they
were built to utilize hydroelectricity to make hydrogen. And
those were built using a technology called alkaline electrolysis,
where caustic soda, or lye, is used as the working fluid, which
is water-based. It's an aqueous solution. And those units are
extremely physically large. Think of an object the size of a
school bus, and they operate at pretty low power levels for their
physical size.


Think about something on the order of hundreds of kilowatts to a
megawatt kind of scale for an object the size of a school bus. On
the other extreme is this more advanced, I would say, technology
called proton exchange membrane electrolysis. And that was
invented... gee, I don't know the exact inventor, but I think it
might have been Westinghouse back in the... fifties? And it was
invented to make oxygen, interestingly, for submarines, for
nuclear power subs. So these were relatively small devices
operating in the kilowatts to 100 kilowatt regime, literally
platinum- and gold-plated, super expensive things. They made it
from submarines to the International Space Station to spaceflight
again for oxygen.


David Roberts


And this is more like breadbox size.


Raffi Garabedian


Yeah, that's right. But they've been adapted and have been scaled
up over the years for the production of hydrogen. They typically
go from, again, from that small scale breadbox scale to objects
the size of a small refrigerator, shall we say, that could be as
big as a megawatt in power capacity. So electrolyzers these days,
the largest electrochemical stacks you can get are on that order,
about a megawatt. They're physically very different in size
depending on the technology.


David Roberts


So it's not crazy to think that electrolyzers could be used in
places where they need to sort of tuck into a relatively small
space. Like, they vary widely in size. They could be made
custom-sized for custom tasks.


Raffi Garabedian


Yeah, they've been used industrially for years. They've been used
in laboratory settings for years and years. They've been used for
on-site production of extremely high purity hydrogen for things
like semiconductor applications, for metal processing
applications, whatnot. There's kind of a history of using these
devices, but generally at relatively small scales. Now, when we
think about the energy transition, which is the topic of your
podcast and certainly my topic of interest, we are starting to
now speak about a much, much different regime of scale.


David Roberts


Right. So this leads to my next question, which is: currently, I
think it's 5%, something like that, of the world's hydrogen is
made with electrolyzers. 95% still comes basically from fossil
fuels. So those of us who are interested in decarbonization are
expecting or asking this technology to scale up super big, super
rapidly, relatively speaking, like this is like we're knocking on
the door like, "Hey, can you like 2000 x in the next decade or
whatever." So my question is just about... is the technology
itself ready to scale up that big? Is it mature in the sense that
efficiencies have been wrung out? Like we have it down as well as
we can? Or do you feel like there's basic tech work to be done
before we're prepared for that sort of explosion?


Raffi Garabedian


That is a great question and a great framing of it. There's all
sorts of really interesting technology in the laboratory today
that promises to ultimately allow electrolysis to expand up in
scale and down in cost. But, our problem is clear and present and
urgent and so waiting for technology to emerge from the lab,
which can take a decade or more because most of these devices
are...first of all, electric chemistry is notoriously difficult
to transition from the laboratory to reality.


David Roberts


When you say in lab, do you mean something fundamentally than
what you described to me, the proton exchange membrane? You mean
like fundamentally different kinds of electronics?


Raffi Garabedian


That's right. There are different chemical cycles that people are
experimenting with, different membranes, new catalysts, all sorts
of really great science.


David Roberts


There was something biological...capillary?


Raffi Garabedian


All sorts of stuff.


David Roberts


We can table that. Maybe we could mention that later.


Raffi Garabedian


Yeah, all sorts of stuff that's out there. Electric Hydrogen, my
company, we of course are aware of all of these kind of more
advanced scientific developments. We've chosen a very different
path because of the time horizon of the industry. So we are
clearly focused on largescale decarbonization and both the market
fundamentals, the secular trends and the policy frameworks are
now in place to facilitate this industry to expand and become
meaningful in the next five years. And that time horizon doesn't
really support basic new technology development. And so we have
opted to take a very well-developed mature technology, proton
exchange membrane electrolysis, and adapted to our new needs.


Now, that is done through some very core technical innovations.
But the fundamental outcome of those innovations is to
dramatically increase the productivity of that physical object.
So that refrigerator-sized electrochemical stack we're talking
about? Think about a five times increase in both the amount of
power that we can run through it and the amount of hydrogen we
can get out of it. The same physical-sized object. Now we can
talk a little bit about how that's done. I'll be cagey about how
we do it.


David Roberts


Yes, this is exactly my set of questions right here. I figured
you would be a little bit cagey about it, but maybe you can tell
us some general things because we're talking about both kind of
the electrolyzer itself, the core, the electrodes, the membrane.
And then, as you said, there's all this "balance of plant" stuff.
It's a big complex process which suggests that there are lots of
places within that process to tweak things and tighten things up
and make things more efficient, et cetera, et cetera.


Can you tell us a little bit about in that whole big messy
process where you are going in and tightening the screws and
finding these efficiencies?


Raffi Garabedian


The pat answer to your question is "everywhere," but let's peel
that onion apart a little bit and explore it. So, let's start
with the product we're building. The biggest electrolyzer that
one can buy today is roughly, I think, a 20-megawatt plant.


David Roberts


What does that look like? What does a 20-megawatt electrolyzer
plant look like? Is that a factory?


Raffi Garabedian


It's a huge building that looks like a small chemical plant. And
it has, you know, maybe 15 or 20 electrochemical stacks in it...


David Roberts


Got it.


Raffi Garabedian


...with a lot of plumbing and pipes and pumps and all sorts of
things. And it's extremely expensive. That's the key, right? The
key isn't actually how big it is. The key is the cost. Now, those
things are related because the cost of plumbing, piping, valves,
pumps, all of those things scales kind of with size. So to make
it smaller is a good thing for cost, but also to keep it large in
terms of its production capacity is extremely important.


Let's talk about the application for a second before we go into
the guts of the plant, because I think it's useful to have a
frame of reference for scale. So you mentioned the current use
cases for hydrogen, of which 95% is supplied using steam methane,
reformation of natural gas. So those use cases are roughly 50-50
split between refineries. So hydrogen is a chemical input to the
petrochemical refining process. And the other half goes to the
production of ammonia, which is fertilizer.


David Roberts


That's the current hydrogen market?


Raffi Garabedian


That's the current hydrogen market, yeah. Now let's just talk for
a second about an ammonia plant. There are dozens and dozens of
these around the world. A typical ammonia plant, if you wanted to
run it off of renewable green electrolyzed hydrogen would require
rough numbers on the order of a gigawatt of renewable capacity
and electrolysis. So that's a big number. And remember, we're
talking about these electrochemical units which are at their
fundamental building block level. They're one megawatt in scale.
So it would be a thousand of these things. That's a problem.


David Roberts


A thousand of these stacks?


Raffi Garabedian


Yeah, that's right.


David Roberts


Per ammonia plant?


Raffi Garabedian


Per single ammonia plant. World scale ammonia plant. Of which
there are many, many. And that's just scratching the surface of
what we want to do with hydrogen. Because, look, cleaning up
ammonia production for fertilizer. That's great. That's some
single digit percentage of global CO2 emissions right there. But
the world is not interested in green hydrogen just for that
reason. The world is interested in green hydrogen as an energy
vector for moving energy from where renewables are abundant to
where people are abundant and need the energy, for example,
right? But also as an input to numerous industrial processes
which cannot be electrified directly with renewables.


For example, the DRI primary steel production process, which
converts iron ore to metallic iron.


David Roberts


Right. Industrial fuel, airplane fuel...


Raffi Garabedian


Ship fuel. So synthetic fuels is generally a broad category of
applications so.


David Roberts


Seasonal energy storage is another...


Raffi Garabedian


You got it.


David Roberts


...big one. Really, hydrogen can do literally anything. So you
could go down the list.


Raffi Garabedian


Yeah. And when you add up all those things that are very hard to
decarbonize directly, you end up with about 50% of global
emissions. I think the IEA's official number is something like
30%. But that skips over a very important factor you mentioned,
which is the long duration storage of energy for the electric
system. And when you add that back in, you get to about 50% of
global emissions. So it's a massive, massive problem, requiring
ultimately terawatts of capacity to be built and installed.


David Roberts


Because I know some people have lots of weird hangups about
hydrogen. I know there are some people out there who are
skeptical about some of these larger uses of hydrogen. But I
think it's worth saying that even if that estimate is off by
several percent in either side, it's still many, many, many,
many, many times the current production. Like, at a certain
point, it hardly matters. The target is so far distant.


Raffi Garabedian


That's right. You could be ten times over ambitious about what
hydrogen could do for the world and still have an immense scaling
opportunity and challenge ahead of you. Because again, we're
talking about terawatts of ultimate available or necessary
capacity to decarbonize all of these sectors. A tenth of that is
100 gigawatts. So big numbers here that we're talking about. So
how do we address the scale of the opportunities and the scale of
the plants that have to be built with technology that's at a
fundamentally different scale today?


David Roberts


Right. We cannot meet that scale we need with current technology.


Raffi Garabedian


Exactly. So we can just try to make more of them. But the issue
with that is that they're too expensive when they're that small.
So the other thing we can try to do is make them more productive.
And the word I use is: "higher throughput." So again, take those
same physical-sized objects and get a lot more value out of them.
The name of the game in all of this is to make green hydrogen
cheap. Let's not lose sight of that. Some of the applications you
mentioned for example, shipping fuel. The economic parity point
for shipping fuel is almost impossible to reach with anything
other than what's currently being burned, which is bunker fuel.


It's the sludge that comes out of the bottom of a distillation
column and a petrol... super gross. Bunker fuel, because it's so
gross, is super cheap. Now we know we can't compete economically
against bunker fuel. Nothing can. However, the alternative of
continuing to burn bunker fuel is less and less tractable. And
so, both by legal restriction and through other economic means,
the cost of the conventional approach, the bunker fuel approach,
is gradually rising and we see it rising at an accelerated pace
as time goes forward. So what we have to do if we want to change
that industry, just as an example, right? We have to drive down
the cost of green hydrogen-derived fuels very rapidly to
intersect as soon as possible with the rising cost of burning
bunker in ships.


David Roberts


And we're all presuming and we'll maybe discuss this in a little
bit more detail in a minute. But the assumption behind all this
is that policy is going to help do this. Like policy is going to
be pushing up the price of fossil fuels—one hopes—even as
technology and scale are pushing down the cost of hydrogen.


Raffi Garabedian


It's happening as we speak. Okay, let's go back to the machine.
So the machine has got to be big.


David Roberts


Yes.


Raffi Garabedian


And let me just say the product that we are building at Electric
Hydrogen, it's about an acre in size. It's funny to think of that
as a product, but that's our product.


David Roberts


And that's multiple... stacks within that?


Raffi Garabedian


It's multiple stacks, but not nearly as many as you might think.
And it's comprised of modular process units. So think about kind
of tractor trailer-sized frames like you might see in an oil
field or a gas field. These are fluid processing units, heat
exchangers, tanks, pumping skids, water treatment units and power
conversion equipment, right? So we modularize all this equipment
so that it can be easily put in place at a project site
interconnected to produce a large-scale hydrogen plant. When I
say large-scale, our product, that one acre size thing, that is
100 megawatt plant.


David Roberts


What is 100 megawatt referring to? 100 megawatts worth of
hydrogen coming out or energy going in?


Raffi Garabedian


That's the energy going in to the electrolyzer. And if you want
to get geeky, that's about 50 tons per day of hydrogen output at
full nameplate capacity. Now, a green hydrogen electrolyzer
should almost never run at full capacity all day long.


David Roberts


Oh, really?


Raffi Garabedian


The only scenario in which you should be able to do that is if
you're connected to a hydroelectric power plant.


David Roberts


Are you talking about because of renewables coming and going?


Raffi Garabedian


Exactly.


David Roberts


Being variable.


Raffi Garabedian


Which is kind of the point of the architecture of our product as
well. Is to be able to follow and track those renewables without
firming the energy. And it has significant bearing on the choice
of technology and the design of both the electrolyzer itself and
also the whole plant that's put around it.


David Roberts


And to what extent are you improving the electrolyzer itself?
From what I understand, the electrodes are made from fairly
specialized metals? Like, are you looking for cheaper electrodes?
Are you trying to improve the actual physical electrolyzing
process itself?


Raffi Garabedian


The latter. So there are really two approaches to making
electrolysis more cost effective to making the equipment cheaper,
which goes directly to making the gas cheaper. The two
approaches, broadly speaking are: make the existing hardware for
less money, find cheaper materials to make it out of, reduce the
labor content of the manufacturing, et cetera, et cetera. The
other approach is: get more out of the same hardware, the same
kind of hardware. Now, when we analyzed these two approaches when
we were starting the company, we kind of looked at the scenarios
and what we thought could be done.


What's the entitlement for the two approaches? Our conclusion was
there might be 30-40% cost to find in the existing technologies
if one can thrift the materials, find cheaper materials, lower
labor, all that, right? And frankly, we're not all that good at
that. I mean, we're good at it, but it's not our forte. What
we're actually really good at is driving a technology roadmap
around performance. And when we thought that through and really
analyzed the entitlement, we found numerous opportunities to get
multiples higher performance out of what is essentially existing
technology, materials and components that are well developed.


David Roberts


And this is the balance of plant stuff you're talking about,
like, stuff outside the elec...?


Raffi Garabedian


No, this is in the very, very guts of the electrolyzer cell
itself. So, we take a device physics approach to the problem. For
those of your listeners who don't know my background, I was Chief
Technology Officer at First Solar for many years. My co-founder
at Electric Hydrogen was also Chief Technology Officer at First
Solar. Before that, he was with Bell Labs running their device
physics department. Long career in electrical and electrochemical
devices.


David Roberts


Yeah, I guess I'm a little baffled how you get more out of the
same materials, so please.


Raffi Garabedian


Yeah. And the way you do it is by understanding the physics of
the device, deconvolving the various contributors to both
performance losses and efficiency losses, and designing solutions
to those material science and interfacial and transport problems.
So it's all around interfaces, material choices, and the physics
of how the device operates. And so with that kind of device
physics approach, we've been able to quintuple essentially, like,
dramatically improve the performance of the electrolyzer cell
itself. Now, that gives us the ability to, without changing the
size or materially changing the cost of that refrigerator thing,
it allows us to get a lot more power into it and a lot more
hydrogen out of it, and that's the secret trick. Right, so our
hydrogen production is extremely physically dense.


Now, when you sell an electrolyzer, what does the customer care
about? The customer is typically a project developer or
industrial owner. They care about the hydrogen production cost.
Ideally, about half of that cost is the cost of capital that's
involved with the purchase of the plant.


David Roberts


Right.


Raffi Garabedian


So when you buy a plant like this, you price it in terms of
dollars per watt or dollars per ton of hydrogen per day produced.
So the more you can get into and out of that thing, the lower its
price proportionately on a dollar per watt basis. So if I make
that refrigerator size box and I can get a megawatt out of it,
its price is a dollar, right? If it costs a million dollars to
make or to sell, it's a dollar. If I can get just make up a
number 10 megawatts out of it, wow. That same thing costs ten
cents a watt.


David Roberts


So those are your biggest gains you're making. Or is it in the
electrolyzer?


Raffi Garabedian


Absolutely. And the electrolyzer then has knock on effects and
ramifications for the rest of the balance of the plant that's
around it. If you look at our plant architecture, at first flush,
it looks like others. But on more detailed inspection, one finds
a lot of very critical differences. Some of them actually add
cost to the plant relative to a conventional approach in order to
support this really fancy, super high energy electrolyzer stack.
But on average, in some total, the cost is greatly reduced even
at the plant scale.


David Roberts


Oh, that's... sort of counter my intuitions. I sort of figured
that with nobody having really tried to scale these things up to
the point you're scaling up, I would think that all the stuff
outside the electrolyzer all that plumbing and structure of the
plan itself and.... I figured there's lots of efficiencies in
that stuff that just nobody's thought to look at yet or squeeze
out of yet.


Raffi Garabedian


Yeah, I think that's also true. Maybe there's about 500 megawatts
of electrolysis installed in the world today. I think the vast
majority of those plants are custom engineered and designed and
built—stick built by EPCs—for a site, right. So there's not a
whole lot of economy of scale or learning yet in this industry.


David Roberts


And your acre-size plant is a set thing. It's the same every time
you build it?


Raffi Garabedian


Yeah, that's why I call it a product. It's not a project, it's
not engineered for a site. You can buy any size and shape
electrolyzer you want from us as long as it's this one. It's like
the old black Model T.


David Roberts


Well, theoretically you could build two if you had two acres.
Right?


Raffi Garabedian


Yeah.


David Roberts


Modular in that sense.


Raffi Garabedian


That's right. And that modular approach is actually really
interesting in the market as well because we're at the early
stages of this industry's growth and so project finance is a
major constraint. Of course, people who are building a
gigawatt-scale plant. They don't want to take the risk on
building that whole gigawatt all at once.


David Roberts


Right.


Raffi Garabedian


They want to build it in small modules. So it does also serve a
market need in that regard.


David Roberts


A couple of other questions about cost. How big a factor is the
cost of the energy you're buying, the cost of the renewable
energy. Like, if renewable energy continues to get cheaper or for
whatever reason, you're able to just find cheaper energy, does
that make a major difference or is that marginal?


Raffi Garabedian


Yeah, it's a huge difference. So let's talk about some numbers
here, okay? So hydrogen produced from steam methane reforming of
natural gas in Texas and Louisiana today, which is kind of the
lowest cost region of production in the world, save maybe a few
places in the Middle East—which are comparable. That hydrogen
costs about a $1.50 to $2 a kilogram to produce and buy. And it's
directly proportional almost to the price of natural gas. Okay,
so that's the bogey. If you can beat a $1.50 a kilo or lower, the
world is your oyster because you're at so called fossil parity
hydrogen.


David Roberts


Right.


Raffi Garabedian


That's—and by the way, that's the dirtiest hydrogen. That...well,
not quite. Hydrogen from coal is dirtier, but that's dirty
hydrogen, that's so called gray hydrogen, right. Where the CO2 is
emitted. If you try to capture the CO2, that price only goes up.
Now, when you talk about electrolysis, the energy input to the
production of hydrogen for most electrolyzers equates to about,
well, every $10 a megawatt hour of electricity price contributes
about $0.60 roughly per kilogram of hydrogen production cost.


David Roberts


So really makes the difference whether you're competing with that
cheap stuff or not.


Raffi Garabedian


That's right. So the best in class solar energy power purchase
agreement that I've seen is just under $10 a megawatt hour in the
Middle East. We also see in the US, new build wind in the wind
belt that's on that order, below $10 a megawatt hour, is
possible. But if your energy is $30 a megawatt hour, you're
already going to be north of a $1.50 a kilogram hydrogen
production cost just with the energy input, not even counting the
cost of capital to build the electrolyzer plant.


David Roberts


Right. And you have limited control over that. I mean, in some
sense, you're subject to what energy is available.


Raffi Garabedian


Yes and no. So here's what we know. We know that renewable power
all around the world where the resource is rich is extremely
inexpensive as long as you don't try to firm it, as long as you
can take it when it's produced. Anything you do to try to firm
that power adds substantial cost, right? Because batteries are
expensive.


David Roberts


Yes.


Raffi Garabedian


So the key for making low cost hydrogen is to take the renewable
energy intermittently is to take it when the wind's blowing, when
the sun's shining.


David Roberts


I mean, I'm sure you get this question all the time. One of the,
you know, I sort of threw this out on Twitter, and I got many
versions of this question, which is: how is it economic to run a
hydrogen-making plant where the capacity factor is, whatever, 40%
or 50% or lower? How do you pencil out the economics when your
energy supply is variable?


Raffi Garabedian


Well, your Twitter followers are brilliant.


David Roberts


I could not agree more.


Raffi Garabedian


That is exactly why we have been working so hard towards this
singular goal of making a large-scale electrolyzer plant that's
really cheap. Not cheap on a dollars per unit basis, but cheap on
a dollars per watt or dollars per ton of hydrogen produced per
day basis, right? That's the key. If your capital plant, your
electrolyzer, is too expensive, you can't afford to run it at a
low capacity factor.


David Roberts


Right.


Raffi Garabedian


If it's cheap enough, now you can afford to use really cheap
energy and run your electrolyzer intermittently. That is the
secret. That is the way you get to low cost hydrogen production.
That's also completely green hydrogen production. The other thing
we should note here is that if you try to firm the energy input
to an electrolyzer using the grid, what you're literally doing is
in the hours when the wind isn't blowing or the sun isn't
shining, you are ramping up a fossil generator somewhere to power
that electrolyzer. And that is a terrible outcome!


David Roberts


And you no longer have green hydrogen!


Raffi Garabedian


You have the worst possible thing. You're burning a fossil fuel,
which you could have converted directly into hydrogen to make
electricity and then convert back into hydrogen right through an
electrolyzer. That's a terrible thing to do. By the way, there's
a policy...I don't know, maybe I'll call it a food fight going on
in the US right now around the rulemaking that results from the
IRA. I go back to policy.


David Roberts


We'll get there. So accommodating intermittent energy input is
not so much a specific technological thing as just a price thing.
If you can get your plant cheap enough, then you can make it work
with intermittent energy.


Raffi Garabedian


Yeah, it's a bunch of things. We have to make the plant cheap
enough so that it can work with intermittent energy. That's the
only scalable kind of clean energy in the first place. There's a
lot of hydro in the world, but not enough to solve a
terawatt-scale problem.


David Roberts


Right. Well, people talk about building nuclear energy plants
specifically to make hydrogen. People talk about using offshore
wind energy specifically to make hydrogen, that way you wouldn't
have to string a power wire out to the offshore wind. Is any of
that stuff on your radar or you think that's mostly distraction?


Raffi Garabedian


Well, the energy problem is extremely localized. It's regional.
So if you're in a place where natural gas is super expensive and
in short supply for geopolitical reasons, whatever, your fossil
parity price for hydrogen might be a lot higher than the numbers
I threw out, right? So if you're in Northern Europe and you're
concerned about Russian gas supply, you might be willing to spend
a lot more for your hydrogen production. And in those scenarios,
things like offshore wind could make a lot of sense.


Nuclear is a tough one for me to understand, quite frankly,
because, look, I mean, the best use of a flexible nuclear plant,
I think, is to continue to clean up the electric system first and
foremost. So if we're able to figure out how to build and scale
more nuclear, wow, let's go do that. I'm not sure making hydrogen
out of it is the right answer. Also, the price of that power is
quite high.


David Roberts


Yes. I can just tell you, Raffi, out there in the world, out
there on Twitter world, people really just want nuclear. They
want it to be useful for something, and so they propose it for
everything. So one more question about the physical thing here,
which is: where are these things currently manufactured? Because
one of the big arguments going on around all the rest of clean
energy is who's making it and who should make it, and is it worth
trying to onshore manufacturing? Who makes electrolyzers today?


Raffi Garabedian


Today... I mentioned I spent over a decade in the solar industry.
When I started in 2008, the solar industry felt a lot like the
electrolyzer industry is today. So we had really strong
industrials in the electrolyzers today, just like in solar back
then. We had really strong industrial players in the US and
Europe, who kind of had the core technology and sold the bulk of
the equipment. But we're seeing China emerging. They're doubling
down on electrolysis, and they're coming up in capacity. I think
there's probably more electrolyzer manufacturing capacity in
China today than anywhere else in the world. Though, it's a
different kind of business than solar. And I think exporting
these very, very large units from China to the rest of the world
is going to be a different kind of challenge. I'm not sure it
rolls the same way.


David Roberts


Right, so the plants you're building in the US are manufactured
in the US? All your parts and pieces?


Raffi Garabedian


Yeah, we'll be manufacturing in the US


David Roberts


One of the things we've been talking a lot about on the pod
recently are learning curves and what kinds of technologies do
and don't get on learning curves. And this work out of Oxford
last year made such a big splash—claimed that electrolyzers are
on a learning curve. So what's your take on that? Is there an
answer about the percentage that the cost falls per doubling of
deployment? Or is it still, do you feel, like, too nascent to
have an answer to that question?


Raffi Garabedian


Yes, I think we are too nascent. Look, I mean, learning curves
are an analysts' way of explaining the trajectory of a whole
industry's work in an area. The goal of any given company,
technology company, is to be a nonlinear force in going down that
curve. Right, so, I don't want to be on a learning curve at
Electric Hydrogen. I want to disrupt that learning curve and put
it in a new direction. And, you know, the analyst learning curve
is simply the aggregate, the average, of a whole bunch of
companies trying to do the same thing, which has come out on top
with the best solution to the problem. So I don't put a lot of
stock in learning curves. I understand why they exist. They're
useful, particularly on the buy side, to kind of try to
understand and predict the future.


David Roberts


Well, they're good descriptive. The question is whether they're
predictive at all, right?


Raffi Garabedian


Yeah. And predicting the future is notoriously difficult, right,
so.


David Roberts


But do you think, based on your experience, that costs are headed
down? You're confident that costs are headed toward that target
they need to hit, and that target is reachable?


Raffi Garabedian


I am. And our explicit goal at our company, at Electric Hydrogen,
is to accelerate that cost down curve. So it's not that
electrolysis isn't going to be scaled without us. It's not that
it's not going to get to the price point it needs to without us.
We think our role in this industry, at least with the role we're
going to try to play, is to be an accelerant, a catalyst.


David Roberts


Pardon the pun. But it's odd, given what a central role
electrolyzers play in this sort of vision of a decarbonized
future. I find it odd, I guess, that I just haven't heard of more
people doing what you're doing, trying to squeeze costs out of
this juncture of the whole system. Do you have a lot of
competitors? Do people coming to this...like who's solar? You
could name five zillion companies, five zillion research labs. Is
there comparable brain power going toward this right now?


Raffi Garabedian


There's a lot of research. So if you look at companies involved
in electrolysis, it's kind of bimodal. On the one extreme, you
have a group, a mode of companies who are large established
players. I'm talking about thyssenkrupp, Siemens, Cummins, folks
like that, right? And then on the other extreme, you have a large
number of very small companies who are at that low technology
readiness level stage. So kind of in the lab playing with new
membrane materials or new catalyst chemistries or whatnot. Now,
the large industrial players, they tend to be very conservative
and slow moving in their technology, road mapping. They tend to
be risk averse because they have massive businesses and their
reputation is contingent on every piece of that business
performing as advertised. So they don't like to take risks. And
then on the other hand, you've got the small companies, the
material science-y companies, who might be a decade from the
market. There are relatively few companies, you could name a
small handful today in kind of the middle ground, which is where
we are.


David Roberts


Right, right.


Raffi Garabedian


Where we're rapidly developing and commercializing technology
that has relatively low risk profile.


David Roberts


Right. And part of what you could do if you succeed, and tell me
if I'm wrong about this, is de-risk this a little bit and lure
some of those bigger players into devoting more resources to
this.


Raffi Garabedian


Absolutely. Now, I don't know, history is sometimes a good
teacher. And again, if we go back to solar, the big industrials
all got out.


David Roberts


Yeah, I remember BP bailing.


Raffi Garabedian


Oh, yeah, everybody bailed. And whether... you could list the
litany of names who no longer exist in the solar industry, who
really wilted under very, very rapid technology innovation
cycles.


David Roberts


Yeah.


Raffi Garabedian


And resulting steep cost reduction curves. So the learning curve
in solar was just brutal and fast and hard. And if you weren't
willing to run with it, you weren't going to survive in it.
Electrolyzers very well could go that way. They also could go the
way of wind, right? So wind power has really been a game that's
been dominated by large industrials because wind has scaled the
other direction. It hasn't scaled in the volume axis, it's scaled
on the size axis. So frankly, the physical size of those units
makes them very hard to innovate rapidly.


David Roberts


Right. Fairly curious where EVs fall and that sort of... maybe
it's too early to tell.


Raffi Garabedian


I think it might be.


David Roberts


Whether the big players will be able to pivot fast enough.


Raffi Garabedian


Yeah, I'll hold my opinions on that.


David Roberts


Alright, so at long last, let's talk about policy. Because one of
the questions I got about this, which I think is a very good
question, is the danger, it seems to me, of being in the green
hydrogen business is that the danger is getting out ahead of
policy such that you start producing on a greater scale than
there is demand. Generally, the market will opt for the cheapest
hydrogen until forced not to by some sort of policy. So is there
enough policy support for green hydrogen now that you're
confident demand will exist for whatever amount you can produce?


Raffi Garabedian


Yeah, I think in the long term, absolutely. Yes both...


David Roberts


Long term is, we're all dead in the long term, or whatever the
phrase is.


Raffi Garabedian


Yeah. Well, let's frame it more carefully. So five plus years
out.


David Roberts


Right.


Raffi Garabedian


I think the answer is yes. There is still a big question mark
around the two to five year time frame. There are gigawatts and
gigawatts of announcements globally of companies who are
building—air quotes building—fossil-free hydrogen production
plants for industrial uses, for energy uses, for grid firm, all
sorts of applications that we've talked about and mentioned. But
what does it take for those to get to a financial investment
decision and for ground to be broken and for electrolyzers to be
installed? That's the question, right.


And I think there is a lot of risk in that. It comes down to
understanding, from our point of view as a company in the
business of making and selling electrolyzers, we do as much
diligence on our customers as they do on us. So our customers are
worried about our technology. "It's a new technology. We haven't
seen it before, and you guys haven't really built one before. How
do we know it's going to work? Okay, great." That's our customers
diligence on us. The other side of that coin is, "Hey, we want to
understand... who's your offtaker for hydrogen? Why do they want
the gas?"


David Roberts


Alright.


Raffi Garabedian


"Why is it economical today? What policies are supporting it?
What's the end use segment and application? And how does all that
work? Why does your project actually make sense?" Because if we
believe it actually makes sense, then we can have much, much
higher confidence that it will go through, that it will get
built.


David Roberts


A little wariness on both sides then, at this point, like the
supply side and the demand side.


Raffi Garabedian


Absolutely. I would say the reality-to-hype ratio is about one to
ten right now.


David Roberts


Yes, it is right there in that cycle. But, presumably your
business thesis, the way you're raising money, is by saying
demand is on the rise.


Raffi Garabedian


Demand is on the horizon. And we look really carefully and
thoughtfully, at least we try to, at the leading indicators that
predict demand. Because, look, I mean, this is an industry that
essentially doesn't exist and has never existed before, so we
can't use past performance to predict the future, right? So we've
got to look at leading indicators. We've got to look a layer or
two underneath what's being built today to understand what's
driving that behavior. And we think the fundamentals are there.
So a number of things come together. Certainly the European
policy framework has firmed up, continues to firm up, and is
driving bona fide, like, verifiable activity on the European
subcontinent.


David Roberts


Is that just the cap and trade? Is that just the general
squeezing of carbon? Or there's the hydrogen-specific stuff
you're talking about.


Raffi Garabedian


There's hydrogen-specific stuff in Europe as well, but there's
also a lot of secondary, "Hey, it can only be solved. We can only
meet these requirements with hydrogen we don't know how else to
do it" kind of things. And then when we look at the US, again the
IRA, which I mentioned, that makes fossil-free hydrogen an
economic viability, like with the snap of a finger.


David Roberts


And that's just a big tax credit. That's like a per production
tax credit. What exactly is the structure of the...?


Raffi Garabedian


That's right, it's framed as a production tax credit. So for each
kilogram of hydrogen you produce, you get a certain number of
dollars in tax credit back, which goes to the bottom line of a
project.


David Roberts


And does it have any stuff about the other colors of hydrogen, or
is this a black and white, sort of like, "We'll give you money if
you do green."?


Raffi Garabedian


Well, it's thoughtfully framed, actually, in terms of the
greenhouse gas intensity of the hydrogen that's produced. It's
technology agnostic, it's greenhouse gas indexed, so you can get
anywhere from $0 to $3 a kilogram tax credit, depending on your
carbon intensity. I mentioned a little while ago there's a bit of
a food fight going on around the rules for that because the
quality of the electricity going into an electrolyzer is what's
being fought over.


David Roberts


Oh, interesting.


Raffi Garabedian


Yeah. Some of us in the industry kind of want to take a long
view, and the long view says, "Gee, that electricity really needs
to be directly fed from a renewal plant." Not on one wire, but at
least time-synchronized and locationally matched.


David Roberts


As opposed to just sort of using grid electricity and then buying
wrecks or whatever.


Raffi Garabedian


You got it. That's exactly the fight that's going on.


David Roberts


This is, again, Volts listeners will find this whole discussion
familiar from the distinction between going 100% renewable and
going 24/7 renewable, matching on an hour-to-hour level.


Raffi Garabedian


It is the same exact thing being fought out right now against
this $3 potential tax credit.


David Roberts


So you're advocating for, "we need to be doing this hour-by-hour
so we know...not just that we're offsetting, but that we're using
clean energy."


Raffi Garabedian


I will unabashedly say we're advocating to do it right, for God's
sake.


David Roberts


And this is ambiguous in the language of the law. So this will
be...who's making this decision in the end?


Raffi Garabedian


Yeah, like any law, right. There's a of lot rules associated with
it. So yeah, the decision is being effectively litigated at this
point.


David Roberts


And does the IRA tax credit sort of add a stroke, make your
current product viable? Is it enough in and of itself?


Raffi Garabedian


Yeah, frankly it does. But I will also tell you that, again,
based on experience in solar and the resulting scar tissue, I'm
extremely wary of subsidies. I value them highly. They're
necessary to get a nascent industry like this off the ground in
the face of a much cheaper but dirty alternative, which is fossil
fuels.


David Roberts


It's the iconic case for subsidies.


Raffi Garabedian


Absolutely. But having said all that, our goal as a business is
to enable subsidy-free fossil parity hydrogen production as soon
as possible because the subsidies are always at risk. They're
expensive, right.


David Roberts


But the ones in IRA are not capped or time as an expiration date
or?


Raffi Garabedian


I believe it's a ten-year.


David Roberts


So that's a pretty good runway. It's a great runway assuming it
stays in place, right.


Raffi Garabedian


Assuming it stays in place. And of course, these things are
political at the end of the day.


David Roberts


Yeah.


Raffi Garabedian


So for a lot of reasons, both fundamental reasons and also
political reasons, our goal is to be subsidy-free to enable
subsidy-free fossil parity pricing as soon as possible. And we
think we can do that in under five years.


David Roberts


Really? That's pretty tight. Like, have you built a plant yet?
Where where are you at in deployment? Have you got a
demonstration plant? Where...what's going on?


Raffi Garabedian


We have a small scale prototype here in California. We'll be
building a pilot towards the end of this year. So the answer to
your question is no.


David Roberts


The pilot is the full-acre plant?


Raffi Garabedian


It's the indivisible unit of that full-acre thing. It's not the
full-acre thing, but that'll be coming in 2024.


David Roberts


And background policy. I know that the IRA is a big, huge deal. I
know there are supports in Europe. What about procurement rules?
I think, like, the federal government now has some sort of like,
rules about the cleanliness of the hydrogen can buy, or big
institutions basically saying, "We'll be a market for this, we'll
be guaranteed offtakers." Is there much of that, or is it mostly
the IRA you're banking on?


Raffi Garabedian


Well, right now in the US, it's mostly the IRA that's driving
adoption. Well, it's the IRA coupled to corporate procurement and
decarbonisation strategies. And just like you see in the world of
renewable energy procurement, the same is going on now in
renewable fuels, clean fuels, and hydrogen writ large as an
element of various industries.


David Roberts


Right, well, like steel and stuff like that, too. If you want
green steel, you're basically saying you want hydrogen. Green
hydrogen.


Raffi Garabedian


That's right. And if you survey the steel industry, you'll find a
spectrum of opinions from company to company as to how seriously
these producers are approaching decarbonization. Some are
extremely committed to decarbonizing rapidly, and others, I guess
I would say, are making moderate moves, grudgingly in that
direction. So you see that in every sector that we work in. You
see it in ammonia production, you see it in fuels, you see it in
steel. There's a spectrum of opinions.


David Roberts


Well, as you said, specifically in the hydrogen market, there's
this sort of like, "You go first. No, you go first." between the
buyers and the sellers. It's a very specific moment in the
market.


Raffi Garabedian


It is. You could say it's a high risk moment. It is. It's also a
high opportunity moment.


David Roberts


Exactly.


Raffi Garabedian


One of the things about this industry that I think will track in
a similar direction as both wind and solar did, there's going to
be a large wave of adoption. And if you're not a participant in
that first wave of adoption as a technology provider, I think it
could become very, very difficult to get down the learning curve
and to scale at a future date.


David Roberts


So you think early movers have a big advantage here?


Raffi Garabedian


I do think so. I do think so.


David Roberts


But it's worth saying, you seem confident that these efficiencies
exist, that the technological possibility exists. So even if, God
forbid, your company doesn't make it, you think this is going to
happen, these electrolyzers are going to get cheaper and cheaper
until green hydrogen is cost competitive? You think that's more
or less inevitable?


Raffi Garabedian


I do. I think that's inevitable. I also think the continued
reduction in renewable power costs is inevitable. Despite there
are short-term disruptions in that market kind of supply, demand,
balance thing. But in the longer term, again, it only gets
cheaper. It doesn't get more expensive.


David Roberts


Awesome. I've kept you a long time, but I guess just by way of a
final question, is you're, as you say, specifically not messing
with the stuff on the lab right now. You are trying to economize
and bring down the cost of existing technology. Is your sort of
like business plan open to the idea that if one of these
capillary things come along or one of these sort of fundamentally
new...because this is a question about storage...that I often ask
people in the storage industry. Like, lithium ion is so
established and so far ahead that if you want to compete with
lithium ion, at the very least, you need to be able to slip
stream in to basically the same manufacturing process because
otherwise you're just starting from nothing and you'll never
catch up.


Is this the sort of thing where if a fundamentally new kind of
electrolyzer comes along that you could just slot it in? Or just
how modular is this and how open is your business plan to sort of
big advances in the technology like that?


Raffi Garabedian


Oh, we're not only open to it, we're eager for it. And we expend
some of our R&D effort on just those kinds of directions as
well as talking to other companies who are more in the lab than
we are. So, yeah, we're absolutely open to it. It's early days in
this industry and there are very few examples in the world of
technology where the solution today is the solution 20 years from
now, right? So this is a long game. We're going to be doing this
for a while, and the technology will shift and will adapt with
it, but not at the expense of losing focus. Again, I think we
have an opportunity that's three to five years in front of us to
scale this industry from a glimmer in our eyes to something that
actually matters at the scale of the global energy system. That's
what we're laser focused on.


David Roberts


Well, thank you so much for coming. This is hugely clarifying for
me now. I feel like I have a little bit something in that
electrolyzer bucket in my head now, and I know our listeners will
appreciate that too. Thanks so much for coming on.


Raffi Garabedian


This has been a great conversation. Thanks for all the awesome
questions and thanks for having me on the show.


David Roberts


Thank you for listening to the Volts podcast. It is ad-free,
powered entirely by listeners like you. If you value
conversations like this, please consider becoming a paid Volts
subscriber at volts.wtf. Yes, that's volts.wtf, so that I can
continue doing this work. Thank you so much and I'll see you next
time.


This is a public episode. If you'd like to discuss this with other
subscribers or get access to bonus episodes, visit
www.volts.wtf/subscribe