Producing fuels from 1,500 degrees of solar heat | Hacker News Reader

Producing fuels from 1,500 degrees of solar heat

thechemicalengineer.com

128 pointsdoener107 comments

> "annually producing several thousand litres of fuel."

That's maybe 10 liters a day. Seems low for the size of the installation. A liter of gasoline is about 10KWh. So if this thing gets 5 hours a day of full sunlight, it's putting out 2KW. That's like 5 standard solar panels.

Either I'm calculating this wrong or this is insanely inefficient.

it is insanely inefficient. whereelse I read you need 3 liters of gasoline to produce 1 liter of synfuel

What is your measure of efficiency? I ask because the cost of sunlight is nominally zero so 10kWh/$0 is a really big number :-).

Note that the article said that as a thermal plant it operates 24/7 by storing heat in a high thermal mass fluid. Another shot at this is Terraform Industries (https://terraformindustries.com/) which is doing something similar with direct PV -> Fuel (rather than using CSP)

The "magic bean", the "trick", the "secret sauce" here is that there are huge swathes of land that are currently both high sunlight receivers and not used (can't farm them, no one wants to live there, think deserts) That could be used to crank out liquid fuel that has no issue being losslessly transported over long distances and for applications that are unlikely to run on batteries any time soon. Making fuel for those by taking CO2 out of the atmosphere is a double win.

The plant is physically quite large; the sunlight might be free but capital and deprecation will influence that ratio.

Although if they can do this economically then good. I have a soft-spot for any solar farm that can be made to look a bit like a flower.

> sunlight is nominally zero

the investment into the panels is not free though. But i do agree - converting electricity into fuels is great, but only if said electricity does not have another use and would've been wasted otherwise.

The capex in something, the opex is neglible, but the opportunity cost doesn't make sense until energy prices go very low or negative is some permanent fashion, or both fuel costs are very high and the switching to alternatives costs too much, or is even infeasible (e.g., air and space transport, shipping, steel production). A premium might make sense for specialist operators who want 'carbon neutral fuel', and have customers willing to pay for it.

A fossil fuel tax would capture the externality that's currently not accounted for, and shift the viability of a project like this.

All that's needed is the desire to avert much larger costs in the future. Or in other words, an absolute impossibility.

That's a very bad idea.

> Making fuel for those by taking CO2 out of the atmosphere is a double win.

It's just a single win because the CO2 is going right back into the atmosphere as the fuel is used (carbon neutral, not carbon negative). Still, a win is better than the loss that extracted fuels amount to.

I agree but you could still argue that any economic incentive to develop better carbon capture is another win. If captured fuel gets lots of use, maybe it will become cheap enough to manufacture that we will do so purely for environmental reasons.

You don't have to use all the fuel produced.

Presumably some could be used to make plastics and everything else we make from oil. You could also bury some.

Carbon-neutral combustion is a win over carbon-positive combustion of extracted coal/oil/gas.

It's not entirely clear, that we can use e.g. deserts "for free". Changing the albedo of significant amounts of surface can have far reaching consequences. And unlike already populated areas (were this already happened), we don't really know what will happen there (probably until we try)

This is a good point. Urban areas are responsible in part for "global warming", just by their albedo. [1]

[1] https://doi.org/10.1038/s41467-022-31558-z

"In part" means 0.00107°C over the next century. That's less than a thousandth of the overall rise. So it's really not the top priority. Even assuming this technology increases it by a factor of 10 it's a massive net gain to displace fossil fuels.

This is a cool project, but it's fair to say that 10 litres of petrol a day isn't sufficient scale to be described as "THE WORLD’S first industrial plant using solar heat to make fuels". It's a research prototype.

My measure of efficiency would at least incorporate capex

It takes a non-zero amount of energy to build all of that energy-capturing equipment – question is, how much?

Sure, if you're doing a TCO analysis you figure out a deprecation schedule for your physical plant and your OpEx for the day to day operations. Of course the CapEx for "one" versus the CapEx per instance for "one thousand" will be quite different. Also how much site prep is needed, how much can be built offsite in a factory setting Etc. So a full economic analysis would incorporate all of that. Then price that against the price of fuel with the carbon and environmental externalities priced in, sure. That would give you a solid set of reasoning to say whether these systems are worse, similar to, or better than existing systems.

If you wait for all of that to be in place (versus risking capital today that might have been used for other things) then you risk dying from those aforementioned externalities of 'business as usual' (aka the do nothing hypothesis).

John was, in my reading, defining "efficiency" to be turning the solar power available as electricity in the surface area of the plant into liquid fuel. And my response to that is always that the solar energy was going to hit that patch of ground anyway, and if you don't have the infrastructure to move it to where it is needed "right now" or store it, then its wasted. California is, today, having days where Solar and/or Wind generation is discarded because there are no customers demanding it. At some point (hopefully soon) we'll get better at dealing with this situation, and converting that "extra" power into syngas is a good use for it.

> That's maybe 10 liters a day. Seems low for the size of the installation. A liter of gasoline is about 10KWh. So if this thing gets 5 hours a day of full sunlight, it's putting out 2KW. That's like 5 standard solar panels.

> Either I'm calculating this wrong or this is insanely inefficient.

With your numbers: 10 l/day * 10 kWh/l / 5 h/day = 20 kW

Right. That's better. But there's more mirror area there than even 50 standard solar panels.

It's a pilot plant. Its purpose is to develop, test, and demonstrate the technology. If successful then they can follow through with plans to build bigger plants that run at full capacity. They won't get the funding for the latter step without the former step first.

It's a small-scale prototype of a much larger plant:

> Synhelion already has plans to build a much larger plant in Spain in 2025 that will ramp production up to around 1,000 t/y. And beyond that, its ambition is to be manufacturing 1m t/y within a decade

> It uses an AI-based method involving drones to calibrate the mirrors 200 times faster compared to traditional techniques using cameras, Synhelion says. Precision is key to ensure the mirrors track the sun and efficiently reflect its light into a solar receiver at the top of a 20 m tall tower.

Naively I would assume we know where the sun is and how to calculate the angles. What are drones and AI doing?

Creating investor hype. What else?

Calibrating out mechanical backlash and mis-alignment in the mountings I assume.

As in fly a scanning pattern around the focus point and look for mirrors that blind the drone instead of frying the tower.

Should be crazy efficient compared to the terrible alternative which would require something like deliberately de-focusing all other mirrors only to be able to verify that the mirror under test actually does project the sun where expected.

Tempted to keep my initial upvote on the "investor hype" sibling, out of spite for all projects that do, but the ability to check arbitrary points in the nonilluminated volume for accidental illumination must really make do much of a difference I can't. Thanks for pointing me in the right direction!

> fly a scanning pattern around the focus point and look for mirrors that blind the drone instead of frying the tower.

Or just put up a big black board with reference points on it, and point a camera at that?

So then you have a live image of a big black board?

The beauty of the misalignment patrol drone is that it does not leave you guessing wich one exactly of your large number of inexpensive mirrors is out of line. The culprit clearly lights up in the video feed.

You normally keep all mirrors pointed at the real target. Then take one at a time and point it at the black board off to the side.

You can then use the camera image of the position of the sun spot on the black board to figure out your new calibration settings for this particular mirror.

Ah, separate target for testing assumptions in isolation.

AI can be just a linear regression these days

> Naively I would assume we know where the sun is and how to calculate the angles. What are drones and AI doing?

Yes, you know exactly where the sun is in a global coordinate system.

The problem is that your mirrors are not fixed in that global coordinate system.

You need to continuously re-calibrate the local coordinate system of the mirrors against the global system.

That is a simple math + sensor problem, not an AI and drone problem.

https://www.topconpositioning.com/us/en/solutions/technology...

I think it’s more that it’s been shown that a motor at each panel is more expensive than manually rotatable mirrors that are maintained by a smaller set of mirrors. For example, there was a startup a few years back that had robots on a rail system go to each panel and adjust it.

The reason should be obvious in that the motors are higher maintenance systems so you don’t want to scale with the size of your plant. I imagine drones are being used for a similar purpose: the drones further reduce how many adjusters you need vs rails and the AI is so that the drones can actually adjust the panels.

For concentrated solar polar, you need almost constant movement of the mirrors to keep tracking the sun well enough.

So you need independent azimuth+elevation for each mirror.

https://www.technologyreview.com/2012/09/04/184005/a-robot-t...

There’s always a performance/cost target you’re trying to hit. Sometimes it’s not about performance at all costs and it’s ok to sacrifice some optimality to get a much cheaper solution.

So while the mirrors are independently adjusted, the there’s a single motor robot on a rail that comes by and custom adjusts each mirror instead of having the mirrors self adjust locally at all times.

It's an interesting idea, but I think you'll be surprised at how fast the sun moves. You can try it yourself. Put up a small mirror that reflects a sun spot on a surface. Clock how fast it moves a centimeter. Multiply that distance with how much further away a central tower would be.

Interesting, do you have any links about this startup ?

You say it's more expensive, do you have a rough estimate of the cost of such a panel (with or without motors) ?

I'm asking because I'm currently working on a small scale automated solar concentrator and I haven't managed to get an idea of how much such a panel costs.

I have only found some sun trackers for photovoltaic panels but they have different angular accuracy requirements than CSP.

Not the startup I was thinking of as I can’t find links for it but same concept:

https://www.technologyreview.com/2012/09/04/184005/a-robot-t...

Wasting investor money and using a disproportionate amount of the worlds energy to build chat bots.

This is just using solar heat to reform methane from biogas. I was hoping it would be using solar heat for production of the reduced chemicals from water or CO2. This latter problem is much harder, but there have been proposals, for example using heat to reduce transition metal oxides with evolution of oxygen, then reacting the oxides with steam to make hydrogen.

M2O3 --> 2 MO + 1/2 O2

2 MO + H2O --> M2O3 + H2

The main issue with CO2 is that the concentrations in air are pretty low. A couple of hundreds of parts per million. So, most synthetic fuel generation is bottle necked on getting enough CO2.

If you are processing a kg of air, you'd be getting only a fraction of a gram of CO2. So if you want to produce say a ton of fuel, you are looking at processing many millions of tons of air.

That's why a lot of synthetic fuel generation is often paired with carbon capture schemes or bio mass (like this one). But of course a lot of those carbon sources aren't actually that clean. Biomass sounds nice until you realize that farming is a big emitter of CO2. And generating fuel from carbon capture of course defeats the purpose. You capture it at great cost and then you create a fuel. Which you then burn and dump in the atmosphere.

This particular scheme in Germany exists because of the influence of car manufacturers there. Companies like BMW are dragging their heels getting rid of their ICE car manufacturing operations and pretending that there are unicorn solutions like hydrogen, synthetic fuels, etc. are great excuses to keep their factories going for a few years longer. Of course EVs are eating their lunch at this point and they are now actually producing lots of those as well.

> You capture it at great cost and then you create a fuel. Which you then burn and dump in the atmosphere.

But if the need for carbon based fuels is not able to be eliminated, this is the next best thing. It is better than digging more of it out of the ground, and then burn and dump it into the air!

Mainly jet fuel. Cars and trucks will do just fine on batteries.

The point with carbon capture is that the captured carbon still comes from fossil fuels. You burn it, you capture it, make fuel, and then you burn it again. So, it's slightly less worse than burning it only once but 100% of it still ends up in the atmosphere.

Synthetic fuels made with biomass are more sustainable but only if the biomass is produced sustainably. Which it often isn't. Dumping a lot of CO2 in the air to produce corn and then turn it into ethanol would be the classic example here of something a lot of countries do at scale that isn't really all that green.

> The point with carbon capture is that the captured carbon still comes from fossil fuels

That's not a property of carbon capture, it's a property of the fuel coming in on the other side of the (stationary) process. We certainly won't be seeing carbon capture of jet fuel any time soon, so the fuel preceeding the capture will definitely not be the one produced. Think waste incineration, think bioplastics (don't waste that precious energy on some pointless biodegradation!), think single use paper, think moving the part of the natural carbon cycle that happens in bushfires and the like into a controlled environment.

Actually jet fuel is already being synthesized from a mix of biomass and captured carbon.

A lot of the carbon capture schemes propose capturing carbon from coal plants or similar sources where massive amounts of fossil fuels are being burned.

But you are right that there are clean sources as well. But mostly with carbon capture, the source is fossil fuel.

Fossil burning plants that capture carbon do it for the greenwashing, not because there might be some use for the concentrated CO2. Renewable burning plants have little incentive to greenwash. Renewable burning does exist, and I suppose that group is much bigger than the tiny subset of fossil burning plants that actually do capture. If we include waste incineration plants in the renewable group (the true multifuel specialists) then the conclusion is that capacity is already abundant, and some of those might actually be considering capture (but their greenwashing pressure isn't remotely as big as in e.g. the coal business, where the greenwashing is nothing less than an existential last ditch hail mary attempt)

> Renewable burning plants have little incentive to greenwash.

That just reflects lack of sufficient CO2 price. Make that high enough and it would become profitable for the CO2 to be sequestered rather than released, assuming the CO2 tax applies to negative CO2 emission.

It turns out one place in the US where CO2 capture is actually already being done on a large scale is in bioenergy. As I understand it, 25% of the CO2 produced by ethanol production in the US is captured. I think it's mostly being used for enhanced oil production, but it is being used.

That's why for every very high value use case (see chii sibling post) you keep some low value use cases running that don't really require high energy concentration fuel. Burn shrubs, waste (e.g. bioplastics), whatever for heat and stationary power and capture that CO2. Low concentration fuels are well available in renewable form and will be forever. Bottleneck solved, at least as long as you don't imagine you could keep up fossil age consumption rates with synthfuels (you can't). Leave those many millions of tons of air to processes we already run on them anyways.

You are right (also with that last paragraph), direct air capture is usually a scam. I might be willing to grant a weak exception to projects that put serial self-containedness front and center, e.g. terraform industries (not to be confused with terraform labs...) would certainly be overwhelmed if they tried to tailor a capturing biomass plant to the local situation with each installation. But even they should better spell out in their advertising "optional, could be substituted with:" for the DAC.

> Of course EVs are eating their lunch at this point and they are now actually producing lots of those as well.

Are actual EV sales numbers reflecting this, in Germany?

Despite their loud complaints about electric cars, yes. EVs sales are rising rapidly, and account for >30% of new car sales in the past year (and that's not counting PHEVs)

In the past year EV sales went very well, agreed. But recently I am seeing articles like this one:

https://www.tagesschau.de/wirtschaft/unternehmen/vw-elektros...

A sharp decline in EV sales for Volkswagen, so they are asking for subsidies.

Battery prices are falling very quickly this year, so EV sales globally are set to explode.

That might be the case, but currently EV prices are generally still significantly higher than ICE car prices of the same class, hitting a slow economy in some EU countries. Maybe Chinese EV can fill that gap.

The proposal described was using methane from biogas. But the microbial decomposition of carbohydrates that creates biogas also creates CO2, in equal molar quantity:

(CH2O)2 --> CH4 + CO2

So, if you want to get as much fuel as you can out of the carbon in biomass, you want to use extra hydrogen (and in this case, it would mean they wouldn't have to separate out the methane in the biogas; just add hydrogen and reform the whole mixture). That extra CO2 is available without the need for expensive direct air capture: the plants already captured it. You could even directly hydrogenate biomass (or, perhaps, molecules derived from aggressive depolymerization of biomass) instead of going through biogas. This is Virent's approach. It avoids the need to break all the C-C bonds before recreating them in the synfuel.

There have been efforts to use solar energy to directly gasify biomass. A company called Sundrop, out of Colorado, was trying to commercialize this 15 years ago. But then fracking dropped the price of natural gas and it all collapsed; the remnants of the company were bought by Chesapeake Energy (a natural gas producer) not long after.

> If you are processing a kg of air, you'd be getting only a fraction of a gram of CO2. So if you want to produce say a ton of fuel, you are looking at processing many millions of tons of air.

If this could be within piping distance of lime kilns, the concentrated CO2 emitted in cement production could be used. Both processes need a bunch of heat so maybe they could get that from the same solar collectors.

This still causes net CO2 emission to the atmosphere (cement does end up absorbing some CO2 later, but it cannot absorb all of it.)

Eliminating CO2 emission from cement production is actually a considerable problem. It's either going to require complete sequestration of all this unavoidable CO2 production, or it's going to require production of calcium oxide from silicates. For example:

CaSiO3 + 2 HCl --> CaCl2 + H2O + SiO2

CaCl2 + H2O + heat --> CaO + 2 HCl

The last step occurs at high temperature, driving the reaction to the right.

This sounds very cool, but "annually producing several thousand litres of fuel"?

An average US car consumes ~1800 litres/year... So this 600 kW solar facility is just enough for one or two cars?

A random search shows you need ~2kW worth of solar to charge your electric car daily, so this technology needs to improve a lot to be competitive with batteries. Hopefully they figure it out before they build the full-size plant.

"This heat powers a thermochemical reactor that turns CO2, water and methane into syngas"

The methane has to come from biomass, so this isn't quite closing the loop here on a full solar only process. Pure CO2 and water to synfuels is still wildly energy inefficient.

When you say inefficient, do you mean that the process results in more entropy than work-producing fuel than another method of recomposing molecules into methane, such as biological processes?

If the Hysatas electrolyzer[1] is for real and scales up, making hydrogen is going to get way more efficient. They can make then make synfuels out of hydrogen for about a 40% loss. Eventually, they could put these electrolyzers at major airports and refuel hydrogen planes without having to move the hydrogen much. That's the big weakness of hydrogen is that it can't be moved around easily in pipelines. It can be stored in underground caverns under low pressure for cheap though.

[1]https://hysata.com/news/hysatas-electrolyser-breaks-efficien...

Ok, I see two semi-related assertions now:

>>> Pure CO2 and water to synfuels is still wildly energy inefficient.

> If the Hysatas electrolyzer is for real and scales up, making hydrogen is going to get way more efficient.

From the article, Synhelion has a thermal reactor now: solar radiation directed by heliostats into a thermal reactor that feeds a Fischer-Tropsch process. It seems difficult to establish relative efficiency of the whole to the Hysatas electrolyzer which is one component of a larger chain to capture and transmit energy to perform work.

If I were to assume a similar story for Hysatas where photovoltaics feed electricity to electrolyze water and a whole new infrastructure for fuel is stood up: What measure would measure and demonstrate the relative efficiency between the two processes?

The finding the biomass to make the methane from is the non-scalable, energy intensive part. You have to collect the biomass, load it on to a truck, ship it around on a truck, and then unload it and dispose of it after it's depleted.

If you only need air, electricity and water to make the synthetic fuel, that's going to be scalable to a civilizational level scale.

Are you suggesting scalability is the measure of efficiency?

Several comments... Disclaimer, I work in this space.

a) Surprised that on HN no one has commented on the similarities with Heliogen: https://www.heliogen.com/ This US-based company backed by Bill Gates and Bill Gross similarly focuses on high-temperature heliostat applications, e.g. green hydrogen and concrete etc. They even have similar hexagonal-mirror heliostats. b) Why these CSP startups so often focus on moon-shot 'super hard' applications like the above baffles me. There are LOTS of great applications for lower temperature solar thermal systems - which are much easier to build and operate. Our plastic-molding systems are just one example: http://lm.solar c) It's a little odd to be doing CSP in Germany - Heliostats need collimated light (non-diffuse light, e.g. light that casts a shadow) and Germany has pretty low DNI compared to, say, Morocco. I know the article says they plan to deploy commercially to Spain, but even a test system would be super hard to operate with frequent haze, high cloud layers, etc. To be clear, not saying PV-solar is impractical in Germany - PV can harvest diffuse light just fine.

Funny that the photo of the solar tower / target in the article shows an overcast sky! Global Solar Atlas gives annual average DNI of @ 1000 kWh/M2/year at Jülich, which is way low. https://globalsolaratlas.info/detail?c=50.922093,6.361102,11...

I wish them luck, but there are likely more practical, impactful uses for CSP.

PS Re the 'sunlight is free' comments... yes but if your process is very inefficient and/or requires a huge heliostat array then CapEx goes way up (which has to be financed = cost) and then you get into needing automated cleaning robots to keep your array working well (see Ivanpah - https://en.wikipedia.org/wiki/Ivanpah_Solar_Power_Facility ), etc.

The cascading effects of moon-shot application => huge CSP system => problems (high CapEx, huge physical sites, permitting problems, need for automated cleaning etc) are exactly why we're working on industrial uses for SMALL heliostat arrays. And why grid-scale CSP (electric generation) systems generally get trounced by PV+battery systems.

I have no idea if technology like this will prove to be scalable, economically competitive, or even practical… but it does seem pretty dang cool as a concept

It's a great way to say "Absolutely no more fossil fuels. For things that need hydrocarbons we'll make them. They cost more but it's better than cooking the planet."

With a side bonus of not sending money to oil rich theocracies like Saudi Arabia and Texas.

With methane as an input, I worry that this will be hard to scale up to the point of no more fossil fuels. Can we get that much methane from biomass?

Considering agriculture and forestry combined produce about 5 billion tons of methane emissions a year, I don't think there's a shortage of availability.

It'll be a lot more work to capture and direct all that, but the biosphere spins off plenty of decomposition byproducts.

Is it more or less energy dense than oil?

    The consumption of oil has steadily increased over the last three decades, totaling 4.53 billion metric tons in 2023,

That's oil shipped and used, not methane drifting free from biowaste across the plant.

How much can we actually capture, how much energy is that, and how great is the conversion challenge to shift most cars to methane, etc?

It appears a little larger than a handwave problem IRL.

> Can we get that much methane from biomass?

I know that food waste generates a heck of a lot of methane in landfills, but I’d love if someone could add some rough napkin math on this!

Biomass availability in the US:

https://www.energy.gov/eere/bioenergy/2023-billion-ton-repor...

One would like to focus on existing streams before having to increase area used to grow biomass.

be afraid.

methane hydrates;

strip mining ocean floors could keep fossil fuel consumption going at current rates for millennia.

but before that we could get we stuff in the permafrost

all terrible ideas ... so probable.

https://duckduckgo.com/?q=methane+hydrates&t=lm&ia=web

Texas?

Yup

At the moment, Oklahoma next door is even more prominent: they just mandated teaching the bible in public schools. But it's much the same.

Were you thinking of Louisiana?

How is Texas a theocracy?

If they get some traction, it might be interesting. It reminded me of another project which I have almost forgotten about.

David Doty is a respected name in a narrow circle of nuclear magnetic resonance scientists for the hardware that his company builds [1]. At some point about a quarter century ago, he became obsessed with what he saw as an impending energy crisis, and started to look precisely into the technical nuances and economics of Fischer-Tropsch process. It seemed like a potential solution to turn excess of renewable energy into an energy-dense liquid fuel, which could then be distributed using the already existing infrastructure.

Doty was always exceptionally meticulous in anything he did, and so he went with a fine comb to find and eliminate inefficiencies in the fuel synthesis process, wherever it was physically possible. He funded a small team working on this, and they came up with some improvements [2] which they have patented, presented at conferences, etc. But despite all this work the economics of the process was still not favorable.

[1] https://dotynmr.com/ [2] http://www.dotyenergy.com/

At this feeble level of efficiency we’re probably better off using the energy to split water into hydrogen or producing ammonia.

It’s a trade-off ultimately, either we get to use currently deployed ICE systems and feed them with this solar fuel, or deploy new engines to leverage hydrogen or ammonia based vehicles.

Most probably better to bet on electricity storage tech catching up and just switch everything to full electric.

There are some things that will be tricky to convert, notably airplanes. And it would be convenient to have synthetic gas for driving peaker plants on the occasions where they're needed.

But yeah, ground transport needs to be electrified yesterday, and the grid upgraded to support it. That can be done incrementally, using existing technology.

In this case isn’t biogas a much better option efficiency-wise?

I think all of these approaches are not "or" but are instead "and" ...

We need both PV solar and "let's get some heat" solar; we need LiFePO4 and nmc batteries; we probably even need to keep fossil fuels and biodiesel and other biogas.

Every little incremental bit helps... It's a race between turning into venus and turning into mad max's thunderdome.

I like how mad Max thunderdome is a given at this point and the env work is just to prevent Venus #2

> the env work is just to prevent Venus #2

There is no science-based predictions that climate change as we understand it today, will lead to Venus #2. That idea is just fodder for climate hysteria and sci-fi fantasy films.

We've got only a very limited understanding of exactly how the current climate works. We have (and have had for a very long time) crude understanding that more CO2 == more energy from the sun retained == more hotter overall.

We've got a Chesterton's fence here with our existing climate as it was between a long time ago and 1940. Sure, sometimes it gets too cold and sometimes it gets too hot, but it's mostly something we've gotten used to.

Once we get into a situation where all sorts of enormous systems with enormous inertia get out of whatever balance they've been in until now change, we're off into radically uncharted territory. It's not something we should gleefully jump into just to see what happens.

We're dancing on a cliff in the fog, we may even be wile e coyote dancing on a cloud.

"mad max" is "turn off all fossil fuels"

"venus" is an example of what can happen with too much CO2 in the air

One of those is actually a fictional thought experiment, the other's actually quite real. The magnitude of how bad the real one is is pretty terrible.. I guess I should just trust you that we can't possibly become venus; is 1/100th of venus tolerable?

"Between madmax and venus" is the path we (civilized, happy people) need to follow over the next 100 years. Hopefully it's a pretty broad path.

The Earth, without humans, has had much higher CO2 levels in the past and yet it isn't like Venus.

> It's a race between turning into venus and turning into mad max's thunderdome.

It is hard to tell how serious you are in that comment, but just in case, worrying about runaway greenhouse gas effect turning Earth into another Venus is climate hysteria.

Poorly thought out energy policies due to climate hysteria have a good chance at creating considerable political unrest though (not sure about "thunderdome" level of unrest).

Turns out when you restrict access to energy (e.g., by increasing its cost), people get upset.

Ammonia will likely not be used for cars and only be used for large industrial applications where there aren't any good alternatives like container ships because it's so toxic. For the same reason we don't have a make semiconductors at home with hydrogen fluoride kit, we won't have ammonia fuel in consumer oriented applications.

In my view, it sounds like they're trying to tackle way too much at the same time.

Just the core system that converts input chemicals and heat to a fuel is a major undertaking. Focus on that.

Adding their own mirror system might be doable, but only if they use a well known and simple solution.

Trying to add an "AI drone mirror autocalibration" is a major undertaking, enough to keep a medium sized company busy for a number of years.

Likewise with the "solar energy storage system". Just run when the sun is shining, and produce as much fuel you can from that.

If you have more incoming power from the sun than what you can produce fuel from, concentrate on solving that instead. Or just build something that is economical even if the mirror system is way oversized in order to always keep the fuel conversion busy.

> What if they could reverse combustion and turn carbon dioxide and water back into fuel?

Humbling thought that the green weed outside your window is doing exactly that - plus, depending on the species, tens of thousands of variations on a carbon theme.

Seems like cool concept, basically it's just solar furnace with extra steps - to me it is a bit nonsense that it is placed in Germany because there are far better locations on a globe like Chile with 10 times more solar energy per m2.

I am glad, however, that the idea of solar furnace is still being explored. Yesterday I was wondering if such installation could be put on a large ship vessel which would solve problem of year season. Also with our knowledge about tornado formations the ship could be put in places with max solar input.

Fischer–Tropsch process

https://en.wikipedia.org/wiki/Fischer%E2%80%93Tropsch_proces...

They should come up with a way to turn leaves into oil. Send it down fractured wells and pump it out 10 years later.

There used to be a company that tried to do this to directly generate electricity: eSolar (https://www.cbsnews.com/news/an-interview-with-esolars-bill-...)

There are lots of companies that used heliostats to generate electricity, but that's not really a problem anymore. My understanding is that:

1. Photovoltaic panels absolutely cratered in price, so they are, generally, much more economical.

2. Not mentioned in TFA but these things have environmental concerns because they are basically bird killing machines. Any bird unfortunate enough to fly through the concentrated "death rays" is instantly fried to a crisp. So these things need solutions like drones to keep birds away.

Generating gas from solar is an attractive option because it's then basically just another form of storage. But it would be great if we could generate gas from solar to power existing gas power plants. The primary problem is that the efficiency of generating gas from (solar-powered) electricity is quite low.

Reading the article I get a feeling that it's more like proof of concept, than a economically feasible project, as there's no mention of costs.

Many years ago I remember seeing this same focused sunlight design being used to generate electricity directly from the heat.

This concept is not quite smoke and mirrors, since there's nothing wrong with the science, but this article definitely reads more like a breathless press release than something truly ground-breaking. More notes below:

> Synhelion was founded in 2016 as a spin-off from ETH Zurich, sparked by what the company founders describe as a crazy idea they had: what if they could reverse combustion and turn carbon dioxide and water back into fuel?

This is not a "crazy idea", but rather a straightforward description of the chemistry involved. We call one implementation of this process "photosynthesis", but there are others.

> The technology they’ve developed relies on four key components. Mirrors – known as heliostats – that track the sun to focus its energy on to a solar receiver. This in turn produces very high process heat at temperatures exceeding 1,500°C. This heat powers a thermochemical reactor that turns CO2, water and methane into syngas, which can be processed via Fischer-Tropsch into fuels.

Again, this is well-understood industrial process chemistry - absolutely a good thing, in my opinion, but not new and sexy by any stretch.

> And finally, a thermal store to release energy when the sun goes down to allow the solar-powered facility to operate around the clock.

This actually IS new and interesting in this application (or at least, it is to me) - a shame that this isn't fleshed out more in the article. I tried to see if there was more about this aspect of their process on the Synhelion website, but their pages were loading slowly and I lost patience. Sorry, team.

> The company says the design of its ultra-thin hexagonal mirrors are key to achieving such high process heats.

Any physicists out there who have a speculation about why the thinness of the mirrors makes a difference here? My understanding is that the maximum temperature that mirrors can get you is limited by the surface temperature of the sun, rather than the mirrors themselves, but I'm certainly no expert on this point.

This bit smells like trying to shoehorn in an application of "AI" where it's not really needed - what's the actual improvement using "drones and AI" over just pre-calculating a tracking curve based on latitude + time of day/year? Or just putting down twice as many mirrors and not bothering to make them track?

> “... The inauguration of DAWN marks the beginning of the era of solar fuels – a turning point for sustainable transportation. Our founding dream of producing renewable fuels from solar energy is becoming a reality.”

This is hyperbole, as eg. Prometheus was doing this two years ago. Additionally, Synhelion will be hamstrung on growth as long as they depend on biomass methane as a feedstock, but they can solve that by buying methane from Terraform :)

Couple of points, but not a physicist or scientist.

1. I believe most existing concentrating solar plants operate below 1,000C. 1,500C might be well understood in producing syngas using other energy sources, but operating a CSP at that temperature is not. To the extent that a company is bringing a CSP plant to market with 1000+ temperature operation, that is novel.

2. Heliostat design seems like a core focus of improving CSP performance. A lot of research direction is on lowering the cost and complexity of deploying and maintaining the mirrors. See: https://www.nrel.gov/csp/solar-fields.html and https://www.energy.gov/eere/solar/heliocon

1,500 C sounds like it's way past what any normal metals can handle?

But in this case, maybe it is enough to have just a small part of the central reaction vessel handle 1,500 C? Maybe using ceramics?

Or can you 3D print an Inconel chamber with builtin cooling channels that makes it possible for the Inconel surface to handle the incoming heat?

I think you nailed it on the ceramics. Pretty sure they are looking at falling ceramic particles in the tower to capture and transfer the concentrating solar.

Silicon carbide would work up to 1800 C. Various refractory oxide ceramics could also work.

This whole approach needs direct sunlight, so it's not great for a place, like Europe, where it's often cloudy. It would be better for deserts, like in Chile, Namibia, or maybe the Arabian Peninsula.

It's not great for a production site. But for engineering tests, you can do a lot of stuff with no sun, and the occasional clear skies are used for sun testing.

>> And finally, a thermal store to release energy when the sun goes down to allow the solar-powered facility to operate around the clock.

> This actually IS new and interesting in this application

Yes, it's a nice thing to have. But it is a major research and engineering project just in itself. It is by no means a solved problem where you can buy a working solution from someone.

> This bit smells like trying to shoehorn in an application of "AI" where it's not really needed - what's the actual improvement using "drones and AI" over just pre-calculating a tracking curve based on latitude + time of day/year? Or just putting down twice as many mirrors and not bothering to make them track?

Open loop (just pre-calculate) is pretty inaccurate in a system like this. Especially over time as the mirror positioning degrades with accumulating inaccuracies.

So there might be a business case for their AI drone calibration system. But just that part in itself is also a major undertaking that requires a significant investmet over a long time to make it a production system.

"Just add more mirrors" sounds like a better business proposition, coupled with a much simpler autocalibration. Perhaps an off-center bullseye target where each mirror periodically aims itself, and a central vision system figures out the azimuth/elevation offsets?

> This is hyperbole

I agree in that I'd be very surprised if they ever get to the point of mass production.

But I agree with the point that generating synthetic fuels from the sun in a 100% renewable way would be very significant.

Sure, in 30-50 years electrical transportation might meet 99% of our needs. But we're not there yet.

I’m looking forward to times when this silly technology kills all plants because no CO2 remained in the air but people need fuel

Reply on news.ycombinator.com