Unlocking green hydrogen, and oxygen deprivation as medicine
34:12 Share

This is a science podcast for January 24th, 2025. I'm Sarah Crespi. First this week, while long touted as a green fuel, the traditional approach to hydrogen production is not very sustainable. Staff writer Robert Service joins producer Megan Cantwell to discuss how researchers are improving electrolyzers. These are devices that split water into hydrogen and oxygen. They're looking at more efficient and durable designs.

Next, researcher Robert Rogers talks with me about the idea of hypoxia as medicine. While effective in certain mouse disease models, the big question is whether reduced chronic oxygen could help people with certain serious conditions like mitochondrial defects or brain inflammation. We discuss what's known so far about this potential treatment and the challenges of delivering low levels of oxygen around the clock.

A question that reporters at Science have been following year in and year out is how do we power the world without actually harming it? A couple of weeks ago, Sarah had staff writer Dan Cleary come on to talk about progress and fusion energy. And this week, we're turning to another source of energy, which is hydrogen gas. This week, staff writer Bob Service wrote about new technologies aiming to make green hydrogen a reality. Thank you so much for joining me, Bob. Happy to be here. So I

So I guess when I think of hydrogen as a fuel source, I mainly think of like jet fuel, maybe buses here and there. But when I read your piece, I was surprised to learn it's actually used in a few different industries. What are the main sources that hydrogen fuel are being used for right now? Yeah, it's very widely used to refine gasoline. Also to make synthetic fertilizer is a major use of hydrogen. And then some industries like manufacturing steel, things like that.

For the hydrogen that's being used in these industries right now, what is the production process? What does that look like? So the

They make it by using fossil fuels. They start with methane. They use high pressure and high temperature to break that methane into hydrogen gas, which is just H2. It's like the H2 and H2O and carbon dioxide. And so they just release the carbon dioxide into the air and they capture the hydrogen and use that for whatever they want. Your story is focusing on green hydrogen, which is

an emission-free way to make this hydrogen gas as opposed to this gray hydrogen, which involves a decent amount of emissions. I guess for where we're currently at right now, is it a lot cheaper to produce it using methane than through these other techniques? It is. So the idea behind green hydrogen is to use renewable energy to make electricity, so wind or solar, and then use that electricity to split water.

You capture the hydrogen, release the oxygen. But to your question, it costs right now about $5 or so a kilogram to make hydrogen with green energy. That's about...

the equivalent of $5 per gallon of gasoline, which is probably five times more expensive than what they call gray hydrogen, which is making it from methane. There's a bunch of different parts of the process that have to be kind of made sustainable for it to actually be green hydrogen. And a bulk of your story is focusing on the actual devices that are splitting this water into hydrogen gas and different components. So right now, when it comes to commercial production, which

isn't that large when it comes to green hydrogen. What exactly is the technology that's being used here? It's an old high school chemistry experiment where you can just stick a couple of electrodes in water and out will bubble hydrogen and oxygen. Even back in the turn of the 1900s, the French military, for example, made hydrogen for their airships. Then what

really took off was when synthetic fertilizer was invented, hydroelectric energy was used to create the hydrogen needed for that. But then in the 1930s or so, steam methane reforming came around and that just took over the market completely. But to your question, these devices are called electrolyzers. So they basically just

take electricity as their input and water, and they use that electricity to split the water. There's several different versions of these that use slightly different variations on the chemistry of how you go about doing this. And the oldest version are called alkaline water electrolyzers. These

resemble something like a battery with a, like we were talking about before, with a pair of electrodes dunked in water. You make hydrogen at one of the electrodes and then you make oxygen at the other electrodes. And then there's a separator in between to keep the two gases from mixing because

If you get to certain concentrations in a mixture, it can be explosive. So that's the design is keeps those gases separate. So alkaline water electrolyzers are the most prominent technology that's being used right now. But there are two others that are sort of far along and in the works.

works. How exactly do these work compared to the alkaline water electrolyzers? Yeah, so there are other versions of the technology that have come along. One of them is called proton exchange membrane electrolyzers, and those are somewhat similar, again, to electrodes that split water.

with a membrane separator. But instead of using a liquid electrolyte, you use a solid. That makes certain things more efficient in that case, but they require more expensive catalysts. In this case, a material called iridium, which is very rare and expensive, and that makes the devices more expensive. And then there's a newer technology called anion exchange membrane electrolyzers. And these are trying to sort of

combine the two to be able to use the cheap catalysts that the alkaline versions use, but the solid membranes that the proton versions use. And then within each of these areas, there are advances going on. With these different sort of approaches to this technology, what are they all sort of coalescing towards to try to improve to make green hydrogen more of a possibility? The limitations of the electrolyzers are

boil down to cost and efficiency. So you're really just trying to get as much hydrogen out at the end of the day for as little money as you're putting into. Some of the technologies tend to be lower cost, like the alkaline water technology, and some tend to be more efficient, like the proton exchange membranes. There's a group in Australia that has redesigned the proton exchange membrane electrolyzer to...

use what they call our capillary forces. So just to dig into the science about how these work a little bit is you have your electrodes submerged in water. And one of the problems is that when the gases form at the electrodes, they create little bubbles, of course, right, that bubble up around the electrodes. Well, those bubbles block more water from getting in. That turns

That turns out to be an efficiency hit because when those bubbles form all around the electrode, then they block more water from being turned into hydrogen gas. So what this group has done is, so instead of most of the electrodes sitting in a liquid water, just the tip of it sits in liquid water and then it wicks that water up through the electrode. Now those electrodes are essentially dry.

or drier. And so when the gas comes off of them, it just comes right off. It turns out their efficiency improvement on that technology is so great that in an instant, they go to beating some of the benchmarks for that technology, which is really impressive. There's just a lot of room for creativity for how you design these things and play around with different catalyst mixtures. Into

In terms of the actual sustainable power that's needed to go through this process with the electrolyzers, if you're using wind power versus solar power, does that kind of play into what might be the best sort of fit for which electrolyzer to use? It does. The challenge with all renewable sources of energy, or I guess solar and wind, and maybe not hydro quite so much, but is that they're intermittent, right? So the sun shines during the day, not at night. If you build, for example, an

an electrolyzer farm that is able to take advantage of solar energy, well, you're getting it for, I don't know, 12 hours a day, right? And so you've spent all this capital to build this electrolyzer plant that you're only using half the time.

So either you need to build a lot of battery infrastructure or build in some redundancy so you're getting extra wind power at night or something like that. I guess what I'm trying to say is that the challenge is in overcoming that intermittency. Is there kind of a world where all three of these sort of different approaches have a place depending on like the situation or do you think one is going to come out on top and be the most dominantly used? I think it's

Probably likely to be where each technology will find its own niche, where it will be the most beneficial. But the race is on right now, actually, among companies and among governments actually pushing different versions of the technology. One example that we can look to is the photovoltaic industry. So in different stages of the development of technology,

PV technology, it looked like different technologies might come and rise up and take over from silicon. Well, silicon has continued to dominate that industry for decades, and it's because the technology development doesn't stop. It keeps relentlessly going forward, even if it's in small steps. So if one of these electrolyzer technologies gets on that same train, well, it might very well become the dominant technology just because it

It's being used. So it's really kind of hard to foresee where this is all going to go right now. In the immediate term, it looks like the initial technology, the alkaline water electrolyzers, have the strongest foothold for sure.

The PEM, the proton exchange membrane, have their own niche. And the anion exchange membrane technologies still need to prove themselves to be durable over the very long term that companies need the assurances to invest in the technology. Before this interview, I was looking to see

What's like one of the oldest stories you've maybe written about this topic? I found one from 2002. There's probably one maybe that's even from earlier than that. Are we kind of at a turning point where it looks like some of these technologies actually can be commercialized and expanded? I would say yes, but with a caveat, it's a little complicated. So I think one of the most hopeful signs is that there is an emerging large scale market. So you're starting to see companies make large devices and

And as you scale up these devices in their size and then make more of them, over time, the costs will come down in the same way that the world's governments have committed to reducing climate gases. And so as part of that, research funding agencies such as the Department of Energy have been making a concerted push

to try to advance all of these technologies to get them to be cheaper and more efficient. The challenge is kind of the everything else piece of it, which is we don't have an infrastructure widely distributed for the storage and distribution of hydrogen. So we do have pipelines for hydrogen in the U.S. and other countries. But if hydrogen were to be a major fuel source worldwide, you would need a lot more storage and a lot more

And that costs money. That's going to be a major challenge going forward. And so the work on the electrolyzers doesn't really solve those problems. The other piece of that is we are developing as a world global society, you know, a pretty sizable infrastructure of solar and wind power. In order for green hydrogen to work, you have to make solar and wind turbocharged. You got to basically double it.

because you would need to power a whole nother industrial sector to replace all the fossil fuels we'd be wanting to cut out. And so not very many people have access to a hydrogen fueling station. So what does that mean in terms of the future of that technology? In terms of the commercial production, how much green hydrogen is actually being produced around the world right now? And how does that compare to the actual need that we have?

Yeah, that's a good question. So today there's about a million tons worldwide of green hydrogen being produced. And while that might sound like a lot, the International Energy Agency estimates that the world will need about 300 million tons by 2050 if it's going to be the sort of foreseen major player needed to minimize climate change. So long, long, long way to go.

And all of that green hydrogen today still comes with subsidies. So in order for the market to really balloon and take off as would be needed to meet those IEA projections,

It would have to get cheap enough that governments would no longer need to subsidize it. Thank you so much, Bob. Well, thank you for doing this. I appreciate it. Bob Service is a staff writer for Science. You can find a link to his story at science.org slash podcasts. Stay tuned for my chat with researcher Robert Rogers about the idea of using chronic reduced oxygen as a treatment for disease.

Hypoxia is a condition in which tissues in the body aren't getting enough oxygen. You may remember those little clip-on pulse oximeters that measure the amount of oxygen in the blood, very popular during the COVID pandemic. Back in early 2020, some researchers noticed that they have these so-called happy hypoxia patients. These are people with COVID.

They had low oxygen in their blood, according to the oximeters, but they didn't really seem to have a lot of negative effects. It turns out hypoxia isn't always harmful, and it can even be helpful in certain cases. This week in Science Translational Medicine, Robert Rogers and colleagues wrote a review on what we know about hypoxia as treatment or as medicine and the outlook for this intervention. Hi, Rob. Welcome to the Science Podcast. Hi, Sarah. Great to be with you.

Yeah, it was really interesting to look this up on the science site to see how much we've written about it because it is kind of a newer idea. And so I found that one on happy hypoxia patients that, you know, they seem to not need interventions just because they have low blood oxygen. But then I also found an upsetting story on the chronic altitude sickness in gold miners living in the highest city in the world. This is in Peru. So I've learned from my scanty research that hypoxia can be neutral or bad, but

But as you write in your review, there is some evidence today that hypoxia can be healing. What are some of the examples of that? Yeah, well, Sarah, I think that's a really great summary of the state of knowledge. And that was really one of the reasons that we wanted to write this piece in Science Translational Medicine. I'm a lung doctor by training, actually. And so this idea that hypoxia could be good for you is, of course, very counterintuitive. And so what we wanted to do is really recap

this past decade of research, which actually was kicked off with an article by my mentor and senior author on this article, Vamsi Muthu, in Science in 2016. And so you're absolutely right. I think in most cases that one could think of, hypoxia is probably not good for you or at best neutral. But based on research in preclinical models, I want to be very clear, all of this right now is in preclinical models, in animal models. There might be certain scenarios in which hypoxia is actually beneficial.

And this work kicked off with a focus on genetic mitochondrial diseases. But then folks have also come at it from several other angles, including diseases that involve inflammation in the brain and in the nervous system, and also after ischemic injury and recovery from that. Right. I kind of see a connection here. You know, mitochondrial diseases might have to do with not being able to turn oxygen and food into energy using your mitochondria. And then ischemic injury has to do with

you know, blood supply to the heart and the brain. Exactly. The initial injury to the heart and the brain comes from a lack of blood flow and therefore deprivation of oxygen and nutrients for some period of time. But then there are, again, in these preclinical models, some evidence that if after that insult, the animal is taken to a lower oxygen setting, they might actually be able to improve their recovery. Right.

Right. So we're starting to see, I think most of this is mice, like genetic models of diseases in mice that treatment with hypoxia can kind of help their recovery or help combat a genetic disease, which is super interesting because, you know, you think a genetic disease, you got to fix the genes, right? Or you got to go in there with a molecule and replace the protein or whatever. But this is like, no, what if we just serve you less oxygen? Maybe

maybe you'll be better off. So how much oxygen do we normally have and what are some dangerous levels and what are people considering for treatment levels? As they say in toxicology, the dose makes the poison, right? And so anything can be dangerous in excess. Certainly you can go too low, right? In this piece, we try to set some guidelines for the field. And I'll say the vast majority of this work in mice deals with oxygen concentration around 11%.

at sea level, as opposed to normal atmospheric oxygen of about 21%.

But as far as general ranges, I like to kind of bucket it into three big categories. One is what we'll call minimal hypoxia. And that's 17% or the equivalent of like when you're in Denver, Colorado, right? Because lots of people, they fly to Denver and they can immediately ski. And then all the way down to say like 11%, we'll call that moderate hypoxia. And then below 11%, I think that's pretty extreme. There are actually people who do live their whole lives at an elevation where the oxygen levels

concentration is equivalent to 11%, but there are some significant health effects there. And it's actually quite hard to, let's say, deliver a baby at that level. If we were translating this into humans for the first time, we think that could be kind of, quote, the exposure cap as low as one would want to go. Giving people less oxygen for just short periods of time is not a totally new idea.

What's different here is that this is chronic around-the-clock exposure to low oxygen. Can you talk a little bit about the differences between these approaches or these paradigms? There are a few different paradigms of low oxygen hypoxia therapy, and many of them have been around for many decades. Those have names like hypoxic preconditioning or acute intermittent hypoxia. The key there is that

In the models and in the human trials where those regimens have been tested is that people are exposed to hypoxia for a short amount of time, like minutes to hours, and then alternating with normal oxygen. What's new about the paradigm is that this really is chronic and it's continuous. If you need to chronically administer this, how do you do that? How do you give...

chronic low oxygen to humans? In the studies that have been done most recently, so at Mass General Hospital, led by our colleague Lorenzo Berra, there was a study where healthy volunteers were brought into a low oxygen tent, basically, or low oxygen chamber for about five days. They started at 16% oxygen and they lowered it by 1% every day, stayed at 11% and then took it back up

It was safe and well tolerated. And then there was actually another study that was done at the German Aerospace Center. And this was in a few patients who were very stable, but shortly beforehand, they had had some sort of heart attack. And then so they brought them to about 12% oxygen for four or five days. It was also safe in them. And they're looking kind of at what happens to their heart's recovery after they had that heart attack. Okay.

Very interesting. Yeah. And then you did you mention also you were saying how is it delivered or? Yeah. Let's talk about that. So you talk about a tent. But like if this were to be translated in such a way that we were treating very different kinds of people, maybe with genetic conditions where you don't turn off the genetic conditions, you keep the oxygen levels low for really long periods of time.

How would you do that? So there's hypoxia-inspired therapy, and then there's chronic continuous hypoxia itself. What I mean by hypoxia-inspired therapy is in all of this fundamental basic science research that's being done, you could imagine that if you really could understand, hey, what is the hypoxia doing in that model? What genes and proteins and metabolites is it affecting? Could you understand that and then translate that into a pill or

or an antibody, or some traditional therapeutic modality that goes ahead and recapitulates the same thing that the hypoxia is doing. But we know that it might be actually really hard for several reasons to do that. And so, you know, we're interested in parallel, really exploring hypoxia itself. And so how to deliver that. So-

One big category is what we call inhalational hypoxia, lowering the amount of oxygen you breathe. And there's really two ways to do that. There's the natural way, which is moving to altitude. And we provide some really interesting epidemiology in the piece just about the number of people worldwide who live at very high altitudes.

And I don't know, I don't recall the exact number, but it's well over 10 million people who reside. I was super surprised. Who permanently reside at an altitude where the oxygen is equivalent to like 13% or so at sea level. A lot of those are maybe small settlements, but some of them are quite major cities. For example, like La Paz and El Alto, you know, these are places that have hundreds of thousands, millions of people. You could imagine that for certain conditions,

You have sophisticated medical infrastructure in large cities like that. I really love that you guys invoke the TB clinics from the 1800s. This has put people up in the mountains and changed the amount of oxygen they're getting. And that actually does arrest the development of TB. So this is something that's kind of founded in...

Very old science at this point. Absolutely. It's such a fascinating story how that goes back in the history there. But the other thing is, you know, it would be great, of course, if we could supply hypoxia to people in their native home environment. Right. If you don't want to move to the most upper regions of Peru. Yeah. Right.

Yeah, and so what I would say is that the technologies that would enable this, right, to create a so-called oxygen-conditioned environment at home, these technologies exist. They'd have to be tweaked a little and miniaturized so that they'd be more practical. But for example, there are high-level endurance athletes who will sleep in an altitude tent. And basically what's happening there is you have a machine that dilutes out or scrubs out the oxygen in ambient air and delivers more nitrogen. You could imagine scaling that up.

Similarly, there are athletic training paradigms where people will breathe a hypoxic mask. Now, you don't want to walk around with some huge tank or anything like that, but you could imagine in the same way that people have portable oxygen concentrators, if they're on oxygen for lung disease, they have perfectly active and normal lives that they carry that around. If that technology could be miniaturized, you could imagine supplying this at home. But I think it would also be cool if you could do hypoxia in a pill, right? What if at any given time,

oxygen level that you're living at or even sea level, you could safely decrease the amount of oxygen that actually gets to the tissues.

And recently in the Muthal Laboratory, my colleagues there had a really exciting preclinical proof of concept of this idea. The key thing to understand is that the oxygen in our blood that's carried around by our bloodstream delivered to our tissues, the vast majority of it is bound to hemoglobin. And so there are things, some of which have been known for a very long time, like low doses of carbon monoxide or a condition called methemoglobinemia, where the

hemoglobin will hold onto the oxygen tighter. No matter how much air you're getting, your body is not translating that into oxygenated tissues. There's something going on in the blood cell that's changing that proportion of oxygen that's delivered. Exactly. For whatever oxygen you're at, it's holding onto it tighter. So there's less oxygen

that's delivered to the tissues. There's actually been a lot of pharmacological research in the last few years about ways to do that. Theoretically, one could think of designing a pill that helps hemoglobin hold onto that oxygen more tightly. And so in the same mouse model in which this benefit of chronically breathing, 11% oxygen was first described, my colleagues have recently described a beneficial effect of just such an approach.

That's super fascinating. But let's talk a little bit about, you know, some of the trade-offs here. So I've been up high. I get tired very easily when that happens. There are some side effects of hypoxia that could be pretty serious. What do people have to worry about if they're going to, you know...

if this gets into humans as a potential treatment and their body is going to be getting lower levels of oxygen? Absolutely. Safety first. And, you know, I just want to be very clear that we don't know if this whole paradigm will be safe or effective if translated to patients. And so what we're talking about is really laying out a framework with very rigorous checking for safety. I think we kind of know, like in an early clinical trial, what

what exactly we would have to look out for. So it's the things like acute altitude sickness and the headache and the nausea that can go on. And that can be on a spectrum from relatively mild disease that people might have experienced if they took a trip to somewhere in the world, high altitude, all the way to very serious conditions, high altitude pulmonary edema, high altitude cerebral edema,

And so any clinical trials, especially in a patient population that might be sick at baseline, has to be done with extreme caution, very slow, I think, initial lowering of the oxygen exposure. What are some of the big questions about translating this from mice? They are very different than us. They're very tiny, for example.

Much easier to put them in a tent. But yeah, what are some of the big question marks when it comes to, you know, thinking about this moving from animal models into humans? There's specific differences in oxygen metabolism and just metabolism between mice and humans that really makes us think about that translational leap. One of them is just that, you know, mice are like a lot smaller than humans, as you said. And so per cell, like per amount of tissue, they're much more metabolically active. They're consuming a lot more oxygen. They pump less blood to their brains than humans do. We

We think that the baseline approach

amount of oxygen in the human brain at the tissue level is a little bit higher in humans. Our hemoglobins are different. So there's just so many differences. I'll say that, you know, when you add up all of those differences that I just mentioned, I'm not sure a priori you could predict, does that mean that you're going to need like a higher or lower level in humans than you do in mice? Because they're kind of conflicting effects that go in all sorts of different directions. And so like many issues in translational medicine, your mice can only get you so far. At some point you have to take a leap into humans.

So we talked a little bit earlier about some of the main diseases that were in mind when you started writing this review. What other areas might this be beneficial for? And how would you find them? Like, how would you figure out where else hypoxia could help people? Yeah, I think there's at least kind of three main areas.

buckets of ways that I would think about finding perhaps new diseases and new disease models. And the first one is just kind of extrapolation from the disease models in which it's already been showing some benefit. So you can imagine if you're working on a disease that's kind of closely related, maybe take a look.

The second way is, of course, to do high throughput screens and screen cells in low oxygen conditions and see when you knock out genes, which ones might be helpful. And of course, that really has to be then followed up. You got to then do the hard work of following up in larger animal models for confirmation.

The third bucket is kind of my favorite as someone who's clinically trained first and foremost, and that is looking for inspiration from the human epidemiology of altitude, right? There's huge variation in the oxygen tension that people are breathing in around the world, and there's probably something to be learned from that. And there is a literature on epidemiology of conditions that are more or less severe or prevalent at various altitudes.

Of course, you know, you got to be careful there because correlation is not causation. And there are so many confounders when you're talking about altitude. Many things differ besides the oxygen concentration, the UV light, the temperature. Yeah. And you can't forget that there are people who have genetic changes from living at altitude for so many generations. Exactly. There's many Highlander populations or at least some Highlander populations in which specific genetic changes have been described. And you can't recapitulate that by just moving a lowlander to Highland's.

So, you know, there's some interesting clues there. I think some of what you would want to do, right, if you could think, how could I generate a list of diseases from epidemiology that are like really exciting? I think I know where we're going with this. This is this kind of natural experiment that you mentioned in your review where a group of people were divided at random and half were basically moved up a mountain and the other stayed at sea level.

That's just a really fascinating story. It comes from a study published by physicians in the Indian Army in the late 1970s. And they were looking back on a group of soldiers in the Indian Army. I believe in the late 60s, early 70s, there were particular border tensions between India and China. And so

Several thousand were stationed at high altitude and the remainder stayed at low altitude. And you could see when the authors who wrote this paper, they were expecting there to be so many health detriments in the soldiers who lived at high altitude. But in fact, a whole bunch of things related to metabolic health were much better. Things like their blood pressure, incidents of ischemic heart disease, incidents of diabetes and their overall glucose control. So there were many metabolic benefits associated.

to the soldiers who lived at high altitude for about two to three years. And then the other thing that was also fascinating was related to our conversation earlier about tuberculosis. They had a much lower incidence of developing tuberculosis. So there's definitely some really interesting clues from human epidemiology. That is very cool. Okay, Rob, we're going to have to stop there. Thanks so much for talking with me today. Thanks.

Robert Rogers was a postdoctoral fellow in molecular biology at Massachusetts General Hospital when this work was conducted. He's now medical director at Tectonic Therapeutic, a biotechnology company. You can find a link to the science translational medicine review we discussed at science.org slash podcast. And that concludes this edition of the Science Podcast. If you have any comments or suggestions, write to us at sciencepodcast at aas.org.

To find us on podcast apps, search for Science magazine or listen on our website, science.org/podcast.

This show was edited by me, Sarah Crespi, Megan Cantwell, and Kevin McLean. We had production help from Megan Tuck at Podigy. Our show music is by Jeffrey Cook and Wenkui Wen. Finally, we're thankful for a generous donation in support of the Science Podcast from the Jacobs Family Foundation. The Jacobs Family Foundation promotes education, health, and scientific endeavors that make a difference in the lives of others. On behalf of Science and its publisher, AAAS, thanks for joining us.