Solar Splitting of CO2 with 3D-Printed Hierarchically Channeled Ceria Structures
onlinelibrary.wiley.com
onlinelibrary.wiley.com
Like this is a very weak claim:
> For example, when compared to e-fuels, the solar thermochemical pathway bypasses the solar electricity generation, the water electrolysis, and the reverse-water gas shift steps, to directly produce solar syngas of desired composition for FT synthesis, i.e., three steps are replaced by one.
Reducing the number of steps might lead to some amazing efficiency breakthrough, but if you mention the step reduction and then stop, I'm going to assume it doesn't.
There's also the question of what kind of system it is embedded in. Typically people have made a CO + H2 syngas using pyrolysis of coal or rubber tires or municipal waste or something like that. You could make methane out of that or make gasoline or diesel compatible fuels, not to mention just about any chemical feedstock you need:
https://en.wikipedia.org/wiki/C1_chemistry
The "water shift" and "reverse water shift" reactions are about what to do if your ratio of CO and H2 are wrong for what you are trying to make so you can basically turn CO into H2 or vice versa. That reactor looks like it would make pure CO without any H2 so if you didn't have an H2 source you could use the energy in the CO to split water and make H2 ("water shift") Similarly an e-Fuel factory might primarily have an electrolizer that makes H2 and you would use the reverse water shift to turn some CO2 into CO.
It doesn't (yet) and they do mention it:
"To date, maximum measured values of efficiency are still in the single digits, namely 5.6% for pure CO2-splitting and 4.1% for co-splitting CO2 and H2O."
Not great so far. Also Cerium is a rare earth, even though quite widespread.
It is even much cheaper than it could be expected from its abundance, because it accumulates as a byproduct of producing REEs for which there is a higher demand, e.g. neodymium, and its storage is expensive for the REE producers. They would be almost willing to give it away for free, to get rid of it.
In recent years most public information about the current price of metals has disappeared, but 10 years ago the price of pure cerium was around $10/kg, in the same range with the prices of pure copper, pure chromium or pure titanium.
Cerium compounds, like the ceria used here, can be significantly cheaper than the metal (because the metal requires a lot of energy for extraction).
A Google search indicates that the current price of ceria (i.e. cerium oxide) is around $1.65 per kilogram, so the price of ceria must be a negligible part of the price of such a CO2 splitting device.
There are so many of these bad schemes for making "green" burnable fuels by splitting CO2 that there is probably some group subsidizing these articles. Not to the point that the technology actually works, just to the point that it generates PR.
It's been proposed that you could sequester carbon by making plastic pellets and burying them, for instance. (For that matter you can partially burn the trees to make a carbon-rich "biochar" which is one of the most powerful soil amendments known.)
People who just want to take carbon waste from an industrial process or the atmosphere and just "make it disappear" are mostly happy to concentrate the CO2 and then pump it underground and not have to do the "work" of converting low energy CO2 into some other product. From a material handling perspective you can pipe the supercritical fuel 200 miles away without loaders and dump trucks and trains and similar hassles.
I think though there is an "e-fuels" market in that someone is going to want to have fossil fuel independent chemistry: for instance the U.S. Navy would love to use nuclear electricity to make aircraft fuel on an aircraft carrier so that the carrier never needs to slow down to take on fuel. With a sufficiently high carbon tax, for instance, e-Fuels could be cheaper to end users than fossil fuels.
I think it's an interesting technology for space exploitation in that the factory of a space colony has to produce all sort of carbon, hydrogen, oxygen, nitrogen and similar organic compounds to support biology and technology.
Some asteroids contain huge amounts of "coal" and carbonates and also water and silicates and it would need some system to turn the coal into organic chemicals (like 20 billion kg of Kapton plastic film for solar sails, sun shades, stuff like that) They would never dispose of waste CO2 because it so precious and certainly they have a system that recycles the carbon... One of those solar reactors would be perfect for always shining sun out there.
Still better than burning it though.
- PLA degrades very, very poorly in landfills [0], which leads me to think that if you were to dig boreholes and dump PLA inside it, wouldn't it effectively sequester the CO2 w/o releasing much/any CH4?
- PLA (as far as I know) is produced from glucose derived from e.g. corn. In theory this is a process which sequester CO2 from the air and puts it into solid blocks. No fossil fuels or fossil fuel extraction required.
So I guess my question is, what's the catch? I suppose agriculture has its own CO2 footprint, but I feel we've gotten pretty efficient at making glucose.
[0] https://all3dp.com/2/is-pla-biodegradable-what-you-really-ne...
As to dumping PLA down bore holes. You can just burry the plant matter without turning it into PLA. So there’s little reason to turn plant matter into PLA to burry unless you have some other use for the PLA at which point it’s a small portion of your waste stream.
My reasoning was that PLA degrades less readily than plant matter, so less likely to release CH4. But I hadn't thought of the fact that if you bury plant matter deep enough, CH4 emissions likely aren't as high a concern.
And plastic leaks methane from dumps.
Unfortunately, this figure is fatal to any solar-thermal project. While the color temperature (hence theoretical limit) of sunlight is a blistering 5000 K, real concentrators achieve much less. For practical purposes they are divided into two categories:
- single-stage concentrators, which employ a single heliostat (array of mirrors) or parabolic trough to focus sunlight onto a collector;
- double-stage concentrators, which use a second mirror to achieve very high temperatures.
The largest double-stage concentrator ever built is the Odellio solar furnace operating at just one megawatt:
https://en.wikipedia.org/wiki/Odeillo_solar_furnace
For practical purposes in the foreseeable future, single-stage concentrators are the technology of interest, currently representing around 7 GWp of installed capacity around the world and growing rapidly. These reach a maximum temperature of about 550 C, which is just barely enough to run the copper-chlorine cycle, but not many of the more fanciful solar fuel cycles. Some single-stage "dish" systems reach 750 C at much smaller sizes (the startup that was selling this pivoted to batteries [1]). Some research has proposed ways to boost power tower temperatures to as high as 800 C [2].
But 1500 C? In a single-stage system? It would require a level of accuracy that just isn't available. Such a system requires a solar concentration ratio of around 2500 suns [3]. This is extraordinarily difficult, because the sun subtends a finite angle for an Earth observer and hence reflects off a mirror with a significant divergence; furthermore an ideal reflector is not planar but slightly curved, and every mirror on a flat field would be curved slightly differently, creating impossibly difficult manufacturing issues. Real-world systems hover around 500 suns, with 1000 suns being a commonly stated practical limit [4].
1: https://en.wikipedia.org/wiki/TEXEL
2: https://asmedigitalcollection.asme.org/solarenergyengineerin...
3: https://opg.optica.org/oe/fulltext.cfm?uri=oe-24-14-A985&id=... (See Figure 4)
4: https://www.sciencedirect.com/science/article/pii/S0038092X2... (Some thermal efficiencies in Figure 13; cost figures are, in my view, highly optimistic)
or you could heat the reactor with microwaves
of course that's not what the paper proposes, and it requires using five times as much land, but it is eminently practical and quite cheap. i'm not sure why anyone builds csp anymore
Design a custom 3D printing process so you can build an optimally shaped structure for absorbing concentrated sunlight and converting CO₂ to carbon monoxide.
1. Build a structure made of CeO₂ (ceria = Cerium Oxide). The structure is shaped to absorb as much sunlight as possible to heat it to 1500°C, reducing to CeO and O₂ at 0.1 mbar
2. Pass a gas of CO₂ and H₂O over the structure at 900 °C and 1 bar, oxidising the CeO back to CeO₂, and producing CO and H₂ (syngas) in a ratio suitable for potential production of specific hydrocarbon fuels.
Decide the structure should be layers of open grids with finer dimension of the grid from top to bottom. Sunlight is absorbed by interior sidewalls of the grid cells. Top layer grid might be a 1x1 cell, second layer 0.5x0.5, next 0.25x0.25 final layer 0.1x0.1. Light comes in at the top layer and light either hits cell wall or passes through cell void to reach the next layer.
Previously investigated printing a scaffold and adding ceria to the outside of the structure - some problems.
So decided to design a custom 3D extrusion printer that could print the ceria structure directly without any scaffold.
Designed a viscous custom fluid containing suspended ceria particles with temperature dependent plasticity. Used that fluid to print 3D structures.
Compared efficiency of syngas conversion process from sunlight energy depending on the 3D structure design.
Sunlight comes from concentrating solar dish and solar tower configurations. This approach uses the entire solar spectrum.How many would it take to remove ~40-50 giga-tons (40x10^12 kg) of CO2 per year? Two trillion (2,000,000,000,000) trees per year. The total number of trees on Earth today is estimated to be about three trillion.
So we would need to plant ~2 trillion per year to just keep up with the amount of CO2 being produced today. Do that many saplings get produced every year? How much land would it take to create that many saplings? How much time and labor? And what steps should be taken so that none of them end up rotting once they are dead (and release the CO2)?
How many more would it take to use the biomass to create something that can offset the existing fuel usage that keeps billions of people alive?
Hence the reason to pursue more efficient forms of carbon capture than just "plant a tree"