Cracks in the off-grid AI data center boom; a court admission that the Trump administration cancelled $7.6 billion in clean energy grants along political lines; BP's talks to sell Lightsource to a Kuwait-backed group; the FCC's new ban on foreign-made power inverters; and Heineken's pilot of a Rondo heat battery to decarbonize brewing in Portugal. WSJ's Clara Hudson joins hosts Paul Gerke and Mike Casey to discuss the heat battery story, plus this week's Cleantecher of the Week: Tim Shepherd and Sylvain Marseille.
Hello and welcome to This Week in Cleantech, your favorite 15-minute-ish roundup of some of the biggest stories in climate and clean energy each week.
Today is Friday, July 31st. We've got guest Clara Hudson from the Wall Street Journal — she'll be joining us shortly.
If you don't know me, I'm Factor This content director Paul Gerke, joined as always by cleantech commentator Mike Casey of Tigercomm.
All right, Mike, we got five stories to get to. As always, we do appreciate hearing from you guys, the listeners and viewers who make this show possible. If you've got a story you want to see covered, or something to share with us, please send us an email: twic@tigercomm.us.
Mike, story number one this week.
Katie Fehrenbacher from Axios has a piece: "Cracks appear in the vision of off-grid AI data centers."
A new Cleanview report counts 59 data centers. They're planning to build roughly 90 gigawatts of behind-the-meter power. That's so much — gas turbines, generators, fuel cells, the works. A smaller group of about a dozen projects tracked by Occam Edge is trying to go almost fully off-grid, representing more than 10 gigawatts.
Recently, a few of those projects have run into trouble. Earlier this month, New Mexico's top land official rejected a gas pipeline that was supposed to fuel fuel cells at Oracle's 2.5-gigawatt project, Jupiter. You may have heard about that one — it's part of Oracle's OpenAI Stargate buildout. It's all very Star Trek-y to me. That could push the project back years.
A week later, a much smaller off-grid data center out in Virginia lost its on-site gas turbines for 24 hours and had to run on diesel backup right in the middle of a really bad air quality day, thanks to that Canadian wildfire smoke that plagued all of us a while back. The neighbors have been complaining around a bunch of these data centers, Mike, about burning lungs and loud noise because of the generators. The Crusoe Stargate site in Abilene, Texas has reportedly gone offline for days at a stretch because of power and cooling issues. We also had that big power drop-off recently in PJM — gigawatts coming off the grid at once.
Mike, what do you think about all this?
Well, I'll tell you what — Jigar Shah thinks this is, quote, "a flimsy way to deploy AI." And he doesn't think the bullish off-grid numbers everyone's throwing around are going to hold up.
I agree.
S&P just downgraded Oracle's credit rating to one notch above junk, citing its data center spending, including all that on-site power infrastructure. Trillions of dollars are riding on whether large off-grid data centers can work as advertised and justify their added cost to win the support of local communities.
Paul, story number two.
Story number two — it's from a former This Week in Cleantech guest. Shout out to Matthew Daly from the Associated Press. The title of this one: "Trump Administration Admits Grants for Clean Energy Were Cancelled Based on Politics."
I saw this one in the evening, post-work hours, when it was published. I immediately shared it with my wife. It's one of those things that transcends what we do for a living — it's kind of a striking article.
What's in it, Mike?
Okay, so the current administration admitted in court that it canceled $7.6 billion in grants across 223 clean energy projects based solely on, quote, "the political identity of the grant recipient state," end quote — since all 16 states had voted for Harris in 2024.
Energy Secretary Chris Wright had spent months saying the cuts came down to economics, calling them, quote, "business decisions based on whether projects were a good use of taxpayer money." But Chris Wright was wrong again. The admission surfaced in a case called Thakur v. Trump. Federal lawyers said they used keywords tied to diversity, gender, vaccine hesitancy, and COVID-19 to screen out projects. It sounds like a merit-based approach to me.
Paul, the money had been funding battery plants, hydrogen projects, grid upgrades, and carbon capture across states like California, New York, and Colorado. Quite a way to run a railroad, wouldn't you say?
Yeah, I would say that, Mike. This isn't something that came entirely out of left field — I mean, obviously we had suspected things like this — but just to see it in bold print, in a headline in the Associated Press, kind of brings us all back down to earth and the reality, the gravity of the situation. And this isn't even the first admission from the Trump administration.
I did notice a lot of editorializing on the Mike Casey part of the docket, but I get it — and government lawyers have already conceded, in an earlier filing, that grant selection was influenced by whether a state, quote, "tends to elect Democratic candidates," end quote. So we kind of already knew this — this is just a little more front and center. The administration has pledged about $3 billion so far to cancel offshore wind projects — we've talked about that on the show a few times — literally giving these companies money to go away and pursue gas and coal ventures instead. Democrats want accountability. Republicans hold both chambers of Congress. So, I'll hold my breath.
Mike, your third story.
Ivan Levingston and Malcolm Moore from the Financial Times. "BP Nears Deal to Sell Solar Business Lightsource to Kuwait-Backed Group."
BP is in advanced talks to sell its solar subsidiary, if you've heard of it, Lightsource, to a consortium backed by Kuwait's sovereign wealth fund. Qualitas Energy and Wren House have emerged as the final bidders. Although the deal isn't confirmed yet, it could still fall through.
Analysts say the real draw for BP isn't the sale price — it's offloading billions of dollars of Lightsource debt. The sale price itself is expected to be pretty insignificant for a company as big as BP.
Lightsource was Europe's largest solar developer, for what it's worth, when BP first invested in it back in 2017. It runs about 4 gigawatts of solar, wind, and battery capacity across 15 countries now, powering about 4 million homes. Two years ago, BP paid over $500 million to buy out Lightsource's stake, taking on more than $2.8 billion in debt in that deal alone.
BP already wrote off more than $4 billion tied to Lightsource and its biogas company, Archaea, last year, and it's flagging another billion-dollar impairment for next month's Q2 results.
So the sale was supposed to be a win for new CEO Meg O'Neill, but her start hasn't gone too smoothly. BP chair Albert Manifold was just ousted over, quote, "serious concerns," end quote, about his behavior, and he's publicly pushed back, denying misconduct.
So BP's net debt sits around $23 billion, and O'Neill wants that down to $18 billion by the end of next year. Buybacks are paused, and divestment proceeds are going straight to debt paydown. Worth noting: rising oil prices from the Iran war are actually helping BP's debt position too, which probably takes a little pressure off deals like this one to close fast.
I do want to point out — as we in this country try to go backwards into the '90s and the '80s — even a country in the middle of a war zone is patiently buying clean energy capacity. So the Kuwaitis are, I think, buying on the dip, and I predict that time will prove them smart.
Story number four. Jeff St. John from Canary Media, someone we talk about a lot on this show. "Trump Admin Bans New Foreign-Made Inverters."
The FCC added, quote, "connected power inverters produced in foreign countries," end quote, to its Covered List — the government's official list of tech deemed a national security threat. Landing on that list generally bars a product from getting FCC authorization to be imported, marketed, or sold in the US.
Inverters convert DC power to AC and are essential for utility-grade solar, battery, and wind projects, plus home solar systems and EV chargers. What a surprise.
The ban only applies to new device models going forward. Anything already approved and on the market — which is most of what's installed today — is untouched for now.
Paul, your thoughts?
Let's hope it remains untouched — that's the big takeaway. We know that solar, wind, and battery projects make up a huge percentage of new generation capacity in the US — more than 90% last year. Whatever happens to inverters is going to happen to most of what's getting built right now in this country.
Currently approved products will eventually become obsolete, but the moment a manufacturer needs a new FCC approval for a hardware update or a cybersecurity requirement — anything like that — foreign-made models won't get approval unless they land a waiver.
Domestic manufacturing can cover some of the gap on storage, but companies like EPC Power and Tesla are already pretty scaled up. Chinese makers still hold about 60% of the US market for utility-scale solar inverters.
All right, Mike, time to bring in a fresh face. Our last story of the day, and our guest on this program: Clara Hudson, Wall Street Journal.
She's got a piece: "Heineken Is Trying a Heat Battery to Generate Steam for Its Brewing." Clara, what's brewing with you? Glad to have you on. So, if people haven't read this story, which we want them to, what's the big takeaway?
So I think what made this story really interesting, and of course fun as always, was the idea that there is this new way to lower emissions for what, as we all know, is a very old, ancient process — brewing beer — which relates to these heat batteries that were the focus of the piece. So essentially the recipe of the beer is going to stay the same. The amount of heat needed to make the beer is going to be the same. But what is making the difference here is switching up how a company is going to store heat.
And I also think that people got into the story because I was talking about beer, and then, you know, they got to learn a little bit about emissions and industrial heat also as a bonus.
That's how you lure them in. You catch more flies with honey than you do with vinegar, I believe is how the expression goes. That makes a lot of sense. Clara, we talk a lot about battery energy storage, and there is a really exciting, emerging part of that space where we're talking about novel technologies — it's not just the lithium-ion stuff that we're used to talking about, and this is one of those. I'm curious, when you were poking around for the story, did you see any of Heineken's contemporaries considering something like this? Are they the only big beverage player that's interested in this kind of new-age energy storage, or is there like a coven of these sort of people?
So I think that Heineken is the first major beverage company, that at least I'm aware of, that has been investing in this. But this is not the only heat battery startup that exists — the company they're working with is not the only one. So another startup that I mention in the piece is called Antora. They use carbon blocks for their heat batteries. So it's sort of a similar premise to Rondo, which was the focus of the Heineken piece. But I think it's also really important to mention that there's really a significant amount of financial might and backing in this space. So the piece I wrote mentioned companies like Microsoft and Aramco and BlackRock that have been putting money into these heat batteries to try and scale them up. So obviously this technology isn't super widespread at the moment, but I do think that having a really influential company like Heineken giving it a shot is a pretty big deal.
Clara, one question for you. It seems to me that about 40% of Europe's carbon pollution comes from burning gas to produce industrial heat and steam. So industrial heat and steam is a big deal. But the next step up for thermal storage, which is by design cheaper than electricity storage, is to be able to have enough heat retained that you can produce power, electricity. Did you get a sense that that's on the horizon for some of these companies, in the course of the reporting, potentially?
I mean, the big takeaway that I see about what is happening between Europe and the US is that European companies are obviously going to be paying far more attention to carbon emissions than those in the US. But then again, it's not always going to be clear-cut, because you've got US multinationals that are operating in Europe. So it's a little bit of a mess — it's a little bit tricky to think of it in such a black-and-white way, I suppose.
But what I think is really the broader context here is that we're seeing so much energy volatility right now with the oil and gas crisis. So the idea that you could have another solution that can give you more control — I think that is what seems so appealing. So just having a new option that you can try, and then maybe scaling that up, is, you know, a promising start.
You kind of led me where I wanted to take you, Clara, and that's about the drivers behind adoption of technologies like this. I feel like in the United States it was very in vogue to lean into the carbon emissions and the going-green and the 100%-by-X-date thing, and that has since very much fallen out of favor in the United States, and it's more about speed to power — how much is this going to cost us, and are there any tax credits available. I feel like in Europe the drivers are still largely based around carbon emissions and, like you mentioned, trying to survive and navigate this fuel crisis. Is that your sentiment as well — that the drivers are distinct across the pond — or do the US folks and those in Europe have more in common than we realize?
I think they have more in common than we realize. So I suppose what is important to remember is that companies everywhere need to lower their emissions and find different ways of securing energy, and that effort isn't going away. You know, despite the political tides, at the end of the day they need to be thinking about that. So they have to think about new energy solutions, and that means they're going to have to try new things and invest in new technology, and maybe try something that's a little bit more out of the box.
All right. Well, Paul, we need to go to our Cleantecher of the Week, but I need to not skip a very important question for Clara. Is it data centers, or data-tuh centers? Because you have an international accent. Yeah. And Paul is always wrong saying "data centers," and I just need you to help him realize that. So Clara, where are we at — data, or data-tuh?
I'm afraid, with the British background, I'm going to say data centers.
What? Let's go, you traitor. And we all know that when somebody says something with a British accent, that means it's right, because it sounds smarter. No, accents just show up in Roman movies — movies about ancient Rome. That's all. You should have thrown him a curveball and said something like "datter centers," and then we could have had a third entrant into this debate. All right, traitor — one more, buddy. Michigander or Michiganian? Which one? We had Alex St. John on last time, she was from Michigan. So Michigander or Michiganian?
It's Michigander. You're a Michigan goose.
Let's go to the Cleantechers of the Week, everybody. We got two of them this week — I've got to get to them, there's two names, not just one, Mike.
Thankfully, neither one of them is that difficult to pronounce. This week's Cleantechers of the Week — drumroll, please — Tim Shepherd and Sylvain Marseille from Pelton Shepherd Industries. They make Nutri Ice. If you haven't heard of this, they're ice packs you can use to feed plants. I guess that makes sense — ice becomes water, right?
It's the only truly compostable gel pack. It offers the same performance as a regular gel pack, and it's obviously safe. It's not — that was a joke. Congratulations to Tim and Sylvain, our Cleantechers of the Week.
All right. Well, we want to thank our story researchers, Alex Petersen and Clare Quirin.
Thank you, the listener, for sticking around to the end of this episode of This Week in Cleantech and appreciating the tomfoolery on this show every week. Thank you to Clara Hudson for sticking around as well and joining this week's episode and lending her expertise. Go check out that Heineken story if you haven't read it yet. Crack yourself a cold one, enjoy the weekend when it hits you. Subscribe. Leave a little feedback. Share a story suggestion. Do what you want — I'm not your dad. Read all the articles we talk about each week; they're in the links in the episode description and in the post on factorthis.com.
Be good, people.