2 ms·

It is exceedingly likely that any sequestration will take substantially more energy than burning fossils fuels produced. I couldn't explain it properly in physi

by graeme 1y ago

It is exceedingly likely that any sequestration will take substantially more energy than burning fossils fuels produced. I couldn't explain it properly in physics terms but when you burn fuel you are releasing stored energy and when you sequester carbon you are storing the energy. If we could store energy cheaper than we could use it we'd have a perpetual motion machine, I think? Fairly sure this would be physically impossible. Where this might be wrong is if we found a process to use another energy source (the sun, something that uses the sun, etc) to do it for us, but we haven't go anything that works in that vein either. Trees are actually one of the better options, but to reverse climate change you'd need to reforest the earth AND sequester all the oil we burned. Burning carbon is effectively debt. If we stopped burning carbon right this second, billions would die, as our whole system depends upon it. But if we don't stop burning it, we increase our future problems. This is unpleasant to reckon with so most don't. I don't think it makes the problem intractable, but it gets harder the more we delay. We do need sequestration because simply eliminating all carbon sources wouldn't be enough, we also have to reduce current levels. Also there are some cases where fossil fuel might be the best solution (rocketry?) so we'd need to be able to deal with the waste.

kragen 1y ago
This is not correct. Your reasoning would be correct if carbon sequestration involved reducing carbon dioxide back to elemental carbon, or hydrocarbons, or whatever form you burned it in. In fact, though, almost all proposed methods for carbon sequestration sequester the carbon dioxide, not just the carbon. Consequently, point-source carbon capture uses only a fraction of the energy released by burning the fuel. It's a significant fraction, though, and atmospheric carbon capture uses more energy because it has to extract the carbon dioxide from air, which is 99.96% things that are not carbon dioxide. As I understand it, the energy dissipation thermodynamically required by that separation is quite small, but getting anywhere close to that thermodynamic limit is going to be a large engineering effort.
- graeme 1y ago
  Ah, thanks, you're right. Though I was thinking mainly of direct air capture. Point source is great, but not actually net sequestration. Need to look into this a bit more, but what would you say the theoretical efficiency is, could we reach a point where you can actually burn fossil fuel to net extract CO2 via direct air capture or another sequestration method that can be scaled?
  - kragen 1y ago
    Point source capture is net sequestration if the fuel is made by direct air capture. That sounds stupid but biomass fuel actually achieves this. It probably can't scale high enough, though. I don't know enough about thermodynamics to calculate the fundamental limits. I suspect that low-temperature sorbents like triethanolamine can currently do direct air capture for less than the energy produced by the fuel, but the process is complicated, involving things like embodied energy in fan motors and hard-to-predict maintenance costs. In a cousin comment (https://news.ycombinator.com/item?id=44461370 https://news.ycombinator.com/item?id=44461370) I did an upper-bound calculation with a very-well-understood atmospheric carbon capture process that people have been doing inadvertently in a no-net-sequestration fashion for thousands of years. It came up with burning no more than 2 kg of coal per kg of carbon dioxide removed, which is theoretically significant net sequestration (if you do point-source capture on the coal burning) but far too expensive to make a dent. Sequestration processes like serpentinization are actually exothermic. The idea there is that you react carbon dioxide with olivine and get serpentine and heat. There's vastly more olivine available than any crustal rock such as limestone. You can do this in lots of ways; for example, you can pump concentrated carbon dioxide down a well into fracked olivine, where it reacts, or you can crush olivine from an olivine quarry into olivine sand and just dump it on beaches, or build artificial islands out of it in Dubai or disputed areas of the South China Sea. Crushing the olivine costs energy, though, and it's energy that's mostly not stored for the process; it just provides more surface area for the same reaction. I don't actually favor this approach (what do you do if you decide the beaches are removing too much carbon dioxide? Beaches don't have emergency stop buttons) but it does show that in principle the effective energy consumption of direct air capture doesn't even have to be positive.

by graeme 1y ago

kragen 1y ago

This is not correct. Your reasoning would be correct if carbon sequestration involved reducing carbon dioxide back to elemental carbon, or hydrocarbons, or whatever form you burned it in. In fact, though, almost all proposed methods for carbon sequestration sequester the carbon dioxide, not just the carbon. Consequently, point-source carbon capture uses only a fraction of the energy released by burning the fuel. It's a significant fraction, though, and atmospheric carbon capture uses more energy because it has to extract the carbon dioxide from air, which is 99.96% things that are not carbon dioxide. As I understand it, the energy dissipation thermodynamically required by that separation is quite small, but getting anywhere close to that thermodynamic limit is going to be a large engineering effort.

graeme 1y ago
Ah, thanks, you're right. Though I was thinking mainly of direct air capture. Point source is great, but not actually net sequestration. Need to look into this a bit more, but what would you say the theoretical efficiency is, could we reach a point where you can actually burn fossil fuel to net extract CO2 via direct air capture or another sequestration method that can be scaled?
- kragen 1y ago
  Point source capture is net sequestration if the fuel is made by direct air capture. That sounds stupid but biomass fuel actually achieves this. It probably can't scale high enough, though. I don't know enough about thermodynamics to calculate the fundamental limits. I suspect that low-temperature sorbents like triethanolamine can currently do direct air capture for less than the energy produced by the fuel, but the process is complicated, involving things like embodied energy in fan motors and hard-to-predict maintenance costs. In a cousin comment (https://news.ycombinator.com/item?id=44461370 https://news.ycombinator.com/item?id=44461370) I did an upper-bound calculation with a very-well-understood atmospheric carbon capture process that people have been doing inadvertently in a no-net-sequestration fashion for thousands of years. It came up with burning no more than 2 kg of coal per kg of carbon dioxide removed, which is theoretically significant net sequestration (if you do point-source capture on the coal burning) but far too expensive to make a dent. Sequestration processes like serpentinization are actually exothermic. The idea there is that you react carbon dioxide with olivine and get serpentine and heat. There's vastly more olivine available than any crustal rock such as limestone. You can do this in lots of ways; for example, you can pump concentrated carbon dioxide down a well into fracked olivine, where it reacts, or you can crush olivine from an olivine quarry into olivine sand and just dump it on beaches, or build artificial islands out of it in Dubai or disputed areas of the South China Sea. Crushing the olivine costs energy, though, and it's energy that's mostly not stored for the process; it just provides more surface area for the same reaction. I don't actually favor this approach (what do you do if you decide the beaches are removing too much carbon dioxide? Beaches don't have emergency stop buttons) but it does show that in principle the effective energy consumption of direct air capture doesn't even have to be positive.