Bill Gates backed a previously stealth startup called Heliogen which can make hydrogen in a scalable way. They use artificial intelligence and a field of mirrors to reflect so much sunlight that it generates extreme heat above 1,000 degrees Celsius. They can concentrate solar energy to the extreme temperatures needed for industrial production of cement, steel, glass and other industrial processes. Carbon-free sunlight can replace fossil fuels at lower costs.
HelioMax features a breakthrough closed-loop control system that consistently delivers high optical performance and saves months of time in calibration. The HelioMax boasts higher concentration and more flux per dollar than any other heliostat system and is available for both electricity generation and process heat applications.
HelioHeat™ is a carbon-free, ultra-high temperature (up to 1500ºC) heat can be used by customers to replace their use of harmful fossil fuels. Historically dirty processes like calcining, reforming, ore roasting, and sintering, which are used in industries such as cement, refining, and mining can now be made green with HelioHeat.
From there, HelioHeat can be used by customers as an alternative to burning fossil fuels, which would significantly reduce energy costs, dramatically lower carbon emissions, and help make great strides in reaching and exceeding corporate sustainability goals.
The Heliogen technology is also capable of thermo-chemically splitting water to create 100% green hydrogen, which we are working to develop as our first HelioFuel.
Our technology at scale is, for the first time, efficient enough to make our ultra-high temperature heat and renewable fuels competitive with fossil fuels. This is a breakthrough for the industry and a major step toward solving climate change.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
Are you sure?
There is no evidence still that human level intelligence is anything but a bag of tricks and workarounds.
Look at it this way. 50 years ago supersonics were in service, now they are a dream. UNIX (Linux is a UNIX) is 50 years old, but no one cares – it is still everywhere. If people who made those things back then were told this, they would not believe it is even possible. Something broke down with engineering profession. Despite the general mood of this community, things are not going well, and that problem you referred to does not surprise me in the least.
AI is just a workaround for intelligence. Whenever we are working towards a new and better method it was traditionally called AI. Once it was achieved it was called automation. So AI was always in the future.
Now anyone and everyone desiring attention wants automation to be called AI.
When we construct real intelligence (as opposed to workarounds) people will likely, and very confusingly, call it “strong AI.”
A better term would be synthetic intelligence (SI), but language is allowed to hamstring our thinking (as witness the fact the word “cell” has 87 different meanings in English. Or the fact that we refer to the cosmos as space-time, even though space and time are probably attributes of space-time, rather than components.
Technically, syngas is just H2 and CO, which can be readily produced from H2 and CO2 in a separate process. But they seem to produce at least some of the CO by running CO2 over their reduced metal oxide, which does affect the efficiency. But they could just as easily replace the CO2 with more water to produce only hydrogen (and then convert to syngas separately if they want to).
Btw, if they’re using cerium oxides, Wikipedia says their oxygen production stage needs 2000 C – https://en.wikipedia.org/wiki/Cerium(IV)_oxide%E2%80%93cerium(III)_oxide_cycle
(edit: On 2nd look, theirs has some other metals added in. Maybe that reduces the temperature..)
I had a browse through their website and nowhere did I see AI mentioned, not even in their press release. What I did see was “uses advanced computer vision software to hyper-accurately align a large array of mirrors” and “developed and patented an innovative computer vision closed-loop tracking control system”. A little bit of poetic licence on Brian’s part?
‘..plant life explodes in warming trends..’ There’s a lot of plant life exploding in Australia right now, but the Prime Minister assures us it’s got nothing to do with climate change, and it’s indelicate to bring the subject up.
Note that despite Archimedes’ futuristic solar weapon, Syracuse was overrun by the Romans and the inventor was killed. Greek fire – classical napalm – might have worked better. I expect similar results from the current stoush between solar and fossils.
It seems they are claiming 30%+ is possible with the 50KW thermochemical reactor at an input of 2500KW/m2 flux on a 15cm aperture, but they are producing syngas (CO and H2), not just H2. The details are behind a paywall though…
https://pubs.acs.org/doi/abs/10.1021/ef201875v
It may have applications where it works out.
Such applications may not be on Earth.
Oh boy…. Okay, let’s read between the lines: ” Carbon-free sunlight can replace fossil fuels at lower costs… ” “At lower costs” does not mean cheaper. Hey, is everyone up for paying a lot more for energy?!!! Yeah, how about that? Let’s pay a whole lot more for energy because Africa will be under water in 3 years… Yeah, right!
The United States is the Saudi Arabia of Natural Gas. Natural Gas burns very clean and doesn’t evaporate rare birds…
They need need to figure out fusion, and in particular, scaled down Tokomak fusion. Until then, I’m tired of the Greens driving up the cost of everything and their wild claims that the world is doomed…
We have had warming and cooling trends for over a billion years. We are in a Warming Trend right now. And guess what boys and girls? Life explodes in warming trends; plant life explodes in warming trends; abundance explodes in warming trends.
So… get the fusion going Bill, and if you want to pay a hell of a lot more for energy – you do it! You can afford it bill!! Most the rest of us are not billionaires!!!
And yet modern solar thermal still suffers from the lighthouse tower problem, AFAIK.
I think Gates is OK, he scatters seed money on a number of technologies and projects. Another one is the perpetually next year pumped storage battery. I presume he hopes that one or the other idea may actually take off, and expects that most of them will come to nothing. And it is not just his money he is just the front man of a group.
Maybe when he feels that he has a winner he will actually throw real money at the project
The idea is to provide high-density process heat. A solar updraft tower can’t do that.
Awwww. You mean Gates’ pwaanny waanny wasunt stwong enuff to stand up to big bad owange man?
Give me a break. Gates should have connected with real industry players ten years ago. His money was wasted. He got had by PNNL posers and academics he probably met at a ball. All TP produced was a decade of salaries, and some renderings – oh and some papers. I still laugh when I think about the original concept of TWR that was supposed to literally burn like a cigar, end to end. Ha! Clowns. Even I siphoned 2 months salary out of TerraPower through BWXT (engineering services) in 2014. Tasty charity. I laugh at NDAs like that! What IP? Your renderings? LOL.
Someone blew his plans away with new restrictions on technology deals with China.
He got cut off at the knee the last time he tried to self fund with his nuclear pet proj, live and learn.
Can’t the pilots just close their eyes while over a solar field?
They would have to add a salt battery to operate 24/7. More cost.
So this is a 2% improvement on a dead end solution.
So is multi-billionaire Bill-G going to go cap in hand to the federal government to beg for money to build these or is he going to self fund and put some of his money in to his own R&D?
Excuse you, self-driving blockchain.
I think it is trivial to reduce scattered reflection without any magical materials. It is literally a textbook solution implemented at least over a century ago. If Mr Gates somehow missed it, well, it is his problem. In any case, I do not see how that can be cheaper that coal, unless some Greta tax is imposed on it. But that cannot be done, as it automatically murders any meaningful economy.
Google says solar thermal power plants have ~15% efficiency, assuming ~30% for steam to electricity. That gives ~50% for the solar heat capture, which should be about the same here.
I don’t know which thermal process Heliogen plan to use for their water splitting, but S-I cycle is ~50% efficient. High-temperature electrolysis at 850 C is 64% efficient according to Wikipedia, and may be more efficient at higher temperatures. But would still take electricity, which isn’t 100% efficient to make. So assuming 50%, total efficiency for making hydrogen would be 50% * 50% = 25%.
What’s the efficiency of Sun-to-Liquid’s hydrogen production?
For comparison, with H2Pro’s process, if they can scale it (sounding like a broken record here, but don’t know whey they’re not getting much attention):
Solar to electricity ~30% (photovoltaic) * H2Pro water splitting 98% = ~29%.
Or combined cycle gas to electricity ~60% * H2Pro water splitting 98% = ~59%. (But then you have to deal with the emissions.)
Have you looked at this variant of the solar updraft tower?
http://vortexengine.ca/index.shtml
It won’t really be commercially viable until it also uses blockchain. Everyone knows this.
Which is partly why I’m not buying the economics angle.
If they would make electricity, the system would be no better than all the solar power heliostats so far. The only advantage would be the lack of foundations for the mirror towers, and how much would that help? 10% cheaper from a starting point of being a multiple times more expensive than coal/gas electricity..?
Depends on the insulation of the pipes. But sure, there are always losses.
For cement making, you need about 800 C, as I recall. Steel making is closer to 1200-1500 C.
And how much would they loose in thermal losses using a heat carrier fluid? The fluid would have to be 1500 deg Centigrade, which is quite hot. If the fluid is just a tad colder – say 1300 degrees – when coming down from the tower, would it still be sufficient to make cement?
Good point. Another option is some sort of metamaterial structure, if made of sufficiently high-temperature materials. They managed to simulate a black hole response with metamaterials a while back. But either of these likely isn’t cheap.
Maybe just a series of high-temp ribs would be enough to reduce reflection. Or shine the concentrated beam into a hollow sphere, so most of the reflection bounces inside for a while, where it can be absorbed and collected. That could increase the efficiency if you collect reflected energy in addition to the primary beam.
No. Vanta Black is carbon nanotubes – they will burn above 500C in air, as any carbon.
You wouldn’t pump the cement. You’d pump a heat carrier fluid. Or use the heat to make hydrogen or electricity, and pump that.
That, and it would be impractical to build such a solar farm next to every industrial plant that needs process heat. Much easier to hook it up to a gas line or electric grid. I’m not buying the “cheaper than fossil fuels” angle just yet.
It reflects off the tower. The tower becomes an ultra bright lighthouse. For which the solution is Vanta Black coating.
Agreed, I could code the algrithm for the control of such a mirror field in a few days. There’s no need for any AI in there, and the article doesn’t tell about any actual breakthrough novelty compared to existing focused solar thermal powerplants.
Personally, I like the solar updraft towers concept a lot better because it’s technically a lot simpler, works with diffused light and can keep producing some electriciy past sunset (from heat stored in the ground).
To compensate for a biological shortage.
Bird kills! They complain about that in Chernobyl on Thanksgiving.
Sooooo easy in Space!
And who can forget the Ringworld sunflowers?
https://larryniven.fandom.com/wiki/Slaver_Sunflower
The industrial issue is that high temperature systems don’t like to thermally cycle a lot. That, an many industrial process right now are optimized for continuous operation. Many engineers would balk at returning to batch oriented systems that can operate in daily solar cycles.
The available heat fit nicely with hydrogen production, see the EU’s SOLAR-JET project and the follow-on
https://www.sun-to-liquid.eu/
which is dealing with full syngas production plus associated production chain to FT production of synthetic hydrocarbons aka synfuels.
but that requires direct lighting of the thermochemical reactor it seems. If you could use a pumpable solid like SiC coated tungsten grains as your thermal receiver, you could store the heat nicely, other thermal transfer fluids like molten salts may be suitable (but normal solar salts have a lower acceptable thermal range so maybe not so great a fit?)
Well, there is one good thing about this system, and that is that they forego the step of producing electricity and use it directly for industrial processes. By doing so, they avoid the efficiency loss when going from heat to electricity, transporting the electricity and to heat again.
This way, they have 100% of the thermal power instead of 40% * 90%*100% = 45% (at best case), so their energy should be at least twice as cheap as electricity generating heliostats. But is that sufficient to compete with coal? And is it practical to transport cement to the top of a tower and down again, or would it increase the capital costs too much?
But, but, they use artificial intelligence!
I think Billyboy should go all medieval with it and use it to burn witches at the stake…
What’s new about this? They have had Arizona dessert mirror roasting tech since the 80s…. a clean source of power …Yadda…Yadda… Yadda….
The basic idea has been around since Archimedes. An advance probably involves better automatic control of the mirror pointing and/or absorbing and using the concentrated light at the hot spot.
I’m not clear on how a focused solar system can blind pilots or other people.
I would have guessed that they spend a fair amount of effort to have the reflected sunbeams go into their heat exchanger. Shining them at random up into the sky would be just wasteful. So unless the pilot was inside the high temp furnace how does the sunlight get focused on them?
Not to mention that a converging beam designed to be focused at a target in a tower 100m away is going to be pretty unfocussed by the time it’s at 10 000 m, even if the stupid computers managed to miss the tower.
Awwww. Whattabout Terrapower?
I hooked my BS meter up to a 24 hour news station and use it to generate free electricity to keep my beer cold.
Items 1, 4 and 5 are more useful than silly stories of “valley of heliostats will be cheaper than coal or gas”. In addition to a useful 1000C+ temperature, this is bonus:
And the power generation makes a perfect cover story for the really useful functions. Definitely worth attention, unless it gets into “dual use” category.
This is old stuff. The French have a similar setup in the Pyrenees. Since decades. Surely Gates know that. There must be something else.
Basically this.
They achieve low costs by skipping foundations. The “AI” is there to automatically re-focus the system even as it settles.
PV to drive DC resistors is probably cheaper if anyone has use for intermittent high temperatures.
It’s not AI, though. It’s just a few hundred servo loops. It’s not even particularly compute-intensive, by modern standards.
We’ve reached the point where marketing people insist on shoehorning “AI” into everything they say. However, like most of the stuff that marketing people say, it doesn’t need to be accurate to sound good.
Ok but that’s just a regular computer program, not AI.
Look up Ivanpah.
Cost is one of the problems. It isn’t the primary problem. This technology doesn’t solve any of the other problems.
1) bird kills
2) only works in arid areas because diffuse light doesn’t concentrate
3) arid areas tend to not have water. Most high temperature processes consume a fair bit of water.
4) blinds airplane pilots
5) glare complaints from everyone else
Etc.
My BS meter is tingling
Because it needs to calculate the angles of hundreds of mirrors to shine light on one specific point that’ll heat up. It would need to move throughout the day and calculate when to move and the best ways to move, just a lot of calculations that would be impossible for people to do within a short amount of time. I work with a firm that built most of India’s solar fields, this is the same principle.
I am totally missing how artificial intelligence is supposed to help with such a trivially simple task as pointing mirrors.