136 ms·

The metastability at STP is particularly intriguing. I wonder what quantum chemistry calculations have been performed on this material. I need to look this up

The metastability at STP is particularly intriguing. I wonder what quantum chemistry calculations have been performed on this material. I need to look this up, and maybe perform a few of my own! Supposing that we can actually generate enough pressure to generate metallic hydrogen at some point, I wonder what the downsides of the material would be.

saalweachter 12y ago
It causes cancer in rats.
throwaway_yy2Di 12y ago
That sounds like a fun weekend hack.
- Xcelerate 12y ago
  I mean, it would have to be its own research project haha. There's no way I could begin to get a simulation like that going in a weekend.
542458 12y ago
> the downsides of the material would be Well, considering how energy-dense it is... As a construction material, wouldn't it be more than a tad dangerous if it ever caught on fire?
- TrainedMonkey 12y ago
  Given temperature and pressure it formed at I am not sure oxidation would be a problem at all.
- andrewflnr 12y ago
  It wouldn't so much catch on fire as revert to gas, straight from solid to H2. Otherwise known as an explosion. Then maybe you get burning hydrogen.
  - ars 12y ago
    Oh, there will be fire. The amount of energy going from H to H2 is more than there is from H2 to H2O. If this really is atomic hydrogen it holds a ferocious amount of energy. https://en.wikipedia.org/wiki/Atomic_hydrogen_welding https://en.wikipedia.org/wiki/Atomic_hydrogen_welding
    - andrewflnr 12y ago
      Yes, my point was that it doesn't make any sense to call that a fire. H(s) to H2(g) is a highly exothermic phase change. Fire is oxidation.
      - ars 12y ago
        That's not a phase change (despite the solid to gas), that's a chemical reaction. Fire is oxidation yes, but despite the name oxidation does not require oxygen, it just requires electron transfer, which happens here.
tfgg 12y ago
You might find a few of these papers interesting. The first challenge is working out the favoured crystal structure, including quantum corrections for the nuclear motion: http://journals.aps.org/prb/abstract/10.1103/PhysRevB.87.184107 http://journals.aps.org/prb/abstract/10.1103/PhysRevB.87.184... http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.110.065702 http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.110... http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.112.165501 http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.112...
- Xcelerate 12y ago
  I had just looked those up actually. The last one is the most interesting to me. I don't like the idea of using DFT for a material whose properties are so unknown experimentally. (d, v)-QMC is much better; however, I think metallic hydrogen could possibly be a prime candidate for the new FCIQMC method. Of course, this is my rough assessment over an hour. I haven't actually read the first two papers in detail.
  - tfgg 12y ago
    Indeed, though there are reasonable arguments that DFT functionals are ok in high pressure regimes, it's obviously best to use the most physically accurate methods when you're in unknown regimes. It's just going to be ridiculously expensive :) I'm not an expert in QMC and haven't really kept up with Alavi's FCIQMC, but that'd be the gold standard if you can also afford to include quantum nuclear motion, which seems to be important here.
    - Xcelerate 12y ago
      > but that'd be the gold standard if you can also afford to include quantum nuclear motion Yeah, which I haven't seen for FCIQMC. (Alavi's group really seems to be the only ones actually using the method so far. I imagine it's still too new for someone else to want to code up unless he distributes the code.) Also, I was perusing your post history (hope you don't mind), and noticed you do DFT calculations. You probably know much more about that than me; I'm primarily MD but keep up with the quantum chemical methods more as a side research project. What are your thoughts using DFT for metallic hydrogen? Is there an exchange-correlation functional that could be good enough?

saalweachter 12y ago

It causes cancer in rats.

throwaway_yy2Di 12y ago

That sounds like a fun weekend hack.

Xcelerate 12y ago
I mean, it would have to be its own research project haha. There's no way I could begin to get a simulation like that going in a weekend.

542458 12y ago

> the downsides of the material would be Well, considering how energy-dense it is... As a construction material, wouldn't it be more than a tad dangerous if it ever caught on fire?

TrainedMonkey 12y ago
Given temperature and pressure it formed at I am not sure oxidation would be a problem at all.
andrewflnr 12y ago
It wouldn't so much catch on fire as revert to gas, straight from solid to H2. Otherwise known as an explosion. Then maybe you get burning hydrogen.
- ars 12y ago
  Oh, there will be fire. The amount of energy going from H to H2 is more than there is from H2 to H2O. If this really is atomic hydrogen it holds a ferocious amount of energy. https://en.wikipedia.org/wiki/Atomic_hydrogen_welding https://en.wikipedia.org/wiki/Atomic_hydrogen_welding
  - andrewflnr 12y ago
    Yes, my point was that it doesn't make any sense to call that a fire. H(s) to H2(g) is a highly exothermic phase change. Fire is oxidation.
    - ars 12y ago
      That's not a phase change (despite the solid to gas), that's a chemical reaction. Fire is oxidation yes, but despite the name oxidation does not require oxygen, it just requires electron transfer, which happens here.

tfgg 12y ago

You might find a few of these papers interesting. The first challenge is working out the favoured crystal structure, including quantum corrections for the nuclear motion: http://journals.aps.org/prb/abstract/10.1103/PhysRevB.87.184107 http://journals.aps.org/prb/abstract/10.1103/PhysRevB.87.184... http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.110.065702 http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.110... http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.112.165501 http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.112...

Xcelerate 12y ago
I had just looked those up actually. The last one is the most interesting to me. I don't like the idea of using DFT for a material whose properties are so unknown experimentally. (d, v)-QMC is much better; however, I think metallic hydrogen could possibly be a prime candidate for the new FCIQMC method. Of course, this is my rough assessment over an hour. I haven't actually read the first two papers in detail.
- tfgg 12y ago
  Indeed, though there are reasonable arguments that DFT functionals are ok in high pressure regimes, it's obviously best to use the most physically accurate methods when you're in unknown regimes. It's just going to be ridiculously expensive :) I'm not an expert in QMC and haven't really kept up with Alavi's FCIQMC, but that'd be the gold standard if you can also afford to include quantum nuclear motion, which seems to be important here.
  - Xcelerate 12y ago
    > but that'd be the gold standard if you can also afford to include quantum nuclear motion Yeah, which I haven't seen for FCIQMC. (Alavi's group really seems to be the only ones actually using the method so far. I imagine it's still too new for someone else to want to code up unless he distributes the code.) Also, I was perusing your post history (hope you don't mind), and noticed you do DFT calculations. You probably know much more about that than me; I'm primarily MD but keep up with the quantum chemical methods more as a side research project. What are your thoughts using DFT for metallic hydrogen? Is there an exchange-correlation functional that could be good enough?