AI chemist could make oxygen on Mars
nature.com
nature.com
Mars catalysis are one of my validation projects, so it's cool to see work done by other scientists in this area too!
Catalyst discovery is quite fun because the molecular actions are highly varied, and you have to also watch for specificity, selectivity stability, and all sorts of other fun stuff. I have been implementing things like MLP (Machine learned potentials) to speed up the initial search and then higher resolution distirbuted workers that do the quantum chemistry after we switch from "exploratory" to "confirmatory" simulations.
Last weekend I wrote some shaders to visualize electron density grids on the GPU, it was super fun and I learnt a lot
I am hoping to get into y-combinator and find a co-founder, I am highly technical and wrote it all myself, I would also like to connect with deep tech investors too
Turns out it was regular expressions.
It was just a REST API.
"Appy" was quite literally just "API" read as a whole word.
> The OER is a thermodynamically uphill reaction involving four consecutive oxidation steps and O–O bond formation, which requires an applied voltage of no less than 1.23 V to operate. The OER overpotential, which is defined as the extra voltage above 1.23 V required for catalysis to occur, characterizes the voltage efficiency of the electrochemical device. Therefore, we chose the measured overpotential as the primary target of our ML model in searching for the optimal OER catalyst10,11,12. We first created 29,902 unique compositions and simulated atomic structures of resulting high-entropy hydroxides (Fig. 2a) from classical MD simulations (Supplementary Figs. 11 and 12). The obtained structural features, such as averaged metal–metal and metal–oxygen distances (Supplementary Figs. 13 and 14 and Supplementary Table 2), are passed to previously established bimetallic hydroxide models13,14 (Fig. 2b) to determine the OER activity of each multimetallic hydroxide by DFT calculation. Three DFT-predicted OER activity descriptors—including the Gibbs free energy change of hydroxyl adsorption ΔGOH* (ref. 15) and differences between the Gibbs free energy change for oxygen adsorption and hydroxyl adsorption ΔGO−OH (ref. 16), the amount of charge transferred for hydroxyl adsorption on the activate site Δq (ref. 17)—and the paired composition information are used for NN training. As Fig. 2c shows, the NN model can accurately reproduce these DFT results. With the NN model, we can now rapidly predict the OER activity of high-entropy hydroxides obtained from any given composition of selected Martian ores (Supplementary Fig. 15), and these theoretical values are then connected with experimentally measured overpotentials. The ML model achieves remarkable accuracy in predicting true overpotentials (Fig. 2d).
It's my understanding that evidence suggests that Mars once had an atmosphere that has since been lost.
If you could give mars an atmosphere tomorrow the loss of atmosphere won't be a problem on human civilization timescales.
Also, Mars doesn’t have a (strong) magnetic field so going outside would be dangerous because you will be bombarded by radiation.
Do Martian rocks have significant amounts of nitrogen?
If you didn't have nitrogen you would need another inert gas to make up 80% (or whatever) of the atmosphere you were going to make.
Being in a pure oxygen environment does not have a long term impact on animals… if the pure O2 environment is 0.21 atm oxygen. There’s a long history of biological experimentation done by space agencies on this, and the Apollo missions operated on a 0.2 atm pressure for the duration of their missions. Apollo 1 famously had trouble not because it was pure O2, but because it was 1.2atm O2.
There is no reason you need backfill inert gasses to have 1.0 atmospheres.
The vast majority of the biomass on earth, including about half of the fixed nitrogen in your body was synthesized industrially via the Haber-Bosch process over the last 100 years.
Aside from those nitrogen fixing bacteria whose output we have already surpassed with industrial processes (artificial fertilizer), inert nitrogen plays no role in biological metabolism. It is not needed, at all.
Mars colonists can import nitrogen, or mine fixed nitrogen from the few deposits that are known to exist. It doesn’t need a massive atmospheric reservoir.
Mars' (almost non-existant) atmosphere is mostly CO2, and you could use that, except that a CO2 level above 10% is lethal for humans and at 10%, it's still too oxygen rich.
Nitrogen has the advantage that it's inert and non-lethal.
However people can wear breathing masks when outside of buildings. Plants cannot. Most terraforming proposals focus on supporting surface plant life, and plants thrive in high-CO2 environments. These proposals involve toxically high (for humans) level of CO2 in order to support lush plant life. Once the O2 and temp gets high enough, people can survive on the surface without space suits, so long as they carry a bottle of oxygen and a breathing mask.
Longer-term, atmospheric pressure could be supplanted with a couple of Kuiper Belt objects consisting of frozen nitrogen and organics. Let it burn up in the thickening atmosphere and contribute to the developing pool of gases.
Which requires supply from somewhere. How far you can go with ISRU depends on how big the colonisation development gets.
But even without access to such deposits, fixed nitrogen in the form of ammonia could be brought from earth in sufficient quantities to support each settler. Organic waste would be composted and thus the nitrogen recycled with very little loss. N2 waste produced by decomposition would be recaptured by the atmospheric regulators, and either fixed by certain plants brought for that purpose, or industrially fixed by an arc or Haver-Bosch process in the bigger cities.
I think we should try to overcome these hurdles, but getting humans out there is a much taller order than establishing a human presence on Mars.
"CelebrityDude had the amazing idea of flying around by building a set of robot arms that would reach down and pull really hard on your own shoelaces, which is something we could do today..."
It's not clear why we would care about that except as a reason to point and laugh at CelebrityDude.
> You think he's a genius?" she said, raising her eyebrows. The high-[noble] twit?
> "I don't know him quite well enough, yet. But I suspect so, a part of the time."
> "Can you be a genius part of the time?"
> "All the geniuses I ever met were so just part of the time. To qualify, you only have to be great once, you know. Once when it matters."
-- Komarr by Lois McMaster Bujold
For a great many discoveries or inventions or ideas, digging a little further into the history shows the famous-people often aren't the first to say/try something, but the ones who attempted it again at a more favorable time or place, were more powerful or influential, or just plain got lucky.
I don't believe in great men, but I don't believe innovation just stochastically appears either. Behind all those little steps upwards is someone who's stumbled on some kind of insight and did something about it. Sometimes an idiot can get lucky and have it work out once, but Musk has done it several times.
I don't hold him on any pedestal. But saying he's just a marketing guy isn't correct either.
My point is, you’d probably get a lot farther by talking to the experts he pays to be close to rather than listening to him.
But if we did, it would have to be replenished but it would be lost on geological not human timescales. Think significant amounts of atmosphere lost after 100,000 years.
You could do the same with the moon.
https://www.youtube.com/watch?v=KsBveF-PwAY
This is the illustrative YouTube video:
1. The moon is closer.
2. We can't even take care of Earth. I'm no expert but I have to believe it would be far cheaper and easier to solve Earth's problems.
This is a popular view but I think it is wrong. Yes, we have problems on Earth. But if you compare Earth today, in terms of "carrying capacity for human beings", or "utils generated per day" or some other metric you like, vs pre-human Earth, I don't think it's a close contest.
The few early humans were hungry, naked and vulnerable to all manner of dangers. Now there are billions of us, and most live in relative safety, comfort and material luxury. If we can do for Mars what we did for Earth, that will be the best thing that has ever happened to Mars.
As for why not moon, I'm not a terraforming expert but my understanding is that the more massive planet will be capable of retaining an atmosphere more easily than our small satellite would.
Maybe let's turn down the optimization on a single metric before we're all turned into paper clips.
Still, I think we should build a moon base for the same reason we should go to Mars: even if these projects don't yield direct benefits, they are probably necessary steps to exploring the solar system and beyond.
In terms of your "why explore space at all while children are dying in Africa?" argument -- we can do multiple things at once! Anyway, people's interests aren't fungible: you can't force the people who want to dedicate their lives to exploring space to instead dedicate their lives to curing malaria. Even if you could, people should be allowed to do what they like. And complaining that person X is doing prosocial activity A rather than prosocial activity B is stupid and antisocial.
But financial capital, political capital, and public attention are fungible and directly trade off. Want a space program at the top of a national political party platform? What are priorities are you going to give up fighting for to prioritize that?
I'm not sure that's true in any simple way.
As far as the government goes, it has a role, but this isn't the 60s. SpaceX has shown that a company can launch rockets. As long as there are people willing to work on the necessary problems and people willing to take the necessary risks, I see no reason (that isn't degenerate) why anyone would want to obstruct those efforts. It's not like making SpaceX illegal is going to somehow fix climate change or cure malaria.
Anyway, I think terraforming is getting ahead of ourselves. Step one is getting supplies to Mars. Step two is building a base there. Step three is making that base as self-sufficient as possible. Down the road from that we can worry about generating an atmosphere.
I was merely asking what the point was, from the comparative costs and difficulty of terraforming other planets vs Earth.
Nor I, but maybe the Earth's problems can't be solved at all, so -at least some people think- the next best thing would be to have an alternative for saving the human species.
We tend to seek technological solutions to our problems, but where has that led us? Cars, fossil fuel use, massive cities, most people wasting away as drones for global corporations.
The problem behind this sort of technology is that it's just an intellectual exercise that is fun for researchers and is trendy because it's about AI, but I sincerely doubt it will contribute any good to the world.
The hard part of taking care of this planet is that every inch is covered with intricate and interconnected systems already. Almost anything you do is guaranteed to have significant side effects and the effort required to mitigate them is extreme. In space this is much less of a problem, comparatively speaking.
Yeah, if mass production and mass consumption is causing the problem, distributed across the commons, then it's pretty obvious that an individual choosing to "opt out" isn't going to solve the problem.
By the way, as creatures that evolved into a world already engaged in tool use and cooking on camp fires, we are fundamentally technological. That smartphone is as natural as a shrub. It's just new, that's the only difference.
GP’s motive is to take care of the planet (and reading between the lines: at least our ability to inhabit it comfortably, since of course the piece of rock will be just fine no matter what we do).
No, age is obviously not the only difference between old technologies and new technologies.
Age is the difference between an old hand axe made out of stone and a new hand axe made out of stone.
The difference between an old hand axe made out of stone and an iPhone 15 goes… umm… a bit deeper.
The intricate and interconnected systems are what is required to make Earth habitable! If you wanted to live on Mars, you'd have to replicate the same complexity synthetically. Hopefully we can learn to treat the Earth with the same care that you'd treat the oxygen production system for your Martian bubble.
HOWEVER, I am very optimistic about upcoming technology in green chemistry, For example our retrosynthesis models are finding molecular pathways that are much more green - I have one project for example that is removing dependence on oil using a chemical pathway that we found that uses a molecule thats easily harvested from discarded orange peels.
I think a lot of the past in chemistry (1800 -> 1940-50s) was done with little regard to the environment but many places are much more concious of it now. I would be hesitant to be critical of the current state of things, which is a product of the past, as a measure of current innovation, I think there's a logical error there.
- this method or a similar one being used on earth to produce oxygen (if we ever need to)
- these AI method being applied in other fields - solving earth-problems here and now
> Cars, fossil fuel use, massive cities, most people wasting away as drones for global corporations.
These are some grim examples, but we are also using technology for good. We have a lot of work ahead of us in medicine and mental health, and AI has the potential to liberate people from chronic pain, mental decline, genetic diseases and generally just living happier and healthier. Maybe your quarrel is more with capitalism than technology, we would be drones for global corporations in some way or another even with pen and paper.
We can and should do both. Remember, you can't just take an astronaut and ask them to farm crops. I mean, you can, but that's not the best use of resources. If we want to use our talent pool appropriately, we need to invest in several different areas.
> The problem behind this sort of technology is that it's just an intellectual exercise that is fun for researchers and is trendy because it's about AI, but I sincerely doubt it will contribute any good to the world.
That's what's always been said about any sort of research that doesn't immediately produces results. Lasers were at some point considered a novelty with zero practical application. Look at what we are now, there's a fiber optic cable right at my home (that's probably using LEDs instead of laser diodes; LEDs were another once very theoretical area of research).
Even Einstein's work was completely theoretical with no know practical applications... for a time.
Research has a way of bleeding through other fields. Many technologies that are required to colonize Mars are also useful right here.