Graph Networks for Materials Exploration

The linked paper from Lawrence Berkeley National Lab is almost way cooler, automated wetlab material science experiments: <a href="https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41586-023-06734-w" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41586-023-06734-w</a>

I don&#x27;t think identification of possible new materials is a rate-limiting step for discovery of better catalysts, batteries, etc. The problem is not coming up with new materials -- it&#x27;s coming up with new materials that _have desired properties_ and _can be cheaply synthesized_.<p>It&#x27;s like if you asked a chemist to draw a few possible structures for organic molecules that have never been synthesized. They can do that. But not all of those possible molecules they came up with will be easy to synthesize. And neither they nor anyone else (without doing a lot of experimental work) will be able to tell you which of those possible structures, if any, would work as a painkiller or an oncology drug.<p>Still, I do think this is a nice demonstration of how more data enables very accurate predictions of energies that would otherwise require expensive DFT calculations. That part is definitely interesting.

Are applications like batteries, semiconductors, solar panels, etc. bottlenecked by the number of available materials? Also, I wonder if the discovered materials are kind of &quot;interpolating&quot; between materials that are already known, or if they expand the convex hull in some way. (Though perhaps it&#x27;s difficult to precisely define what the convex hull of materials is.)

Nice trick, but it’s almost useless.<p>We have been using for decades integer programming to explore all the possible permutations with hard constraints that include manufacturability.<p>Their references list is lacking, to say the least.

Isn&#x27;t the problem how to actually scale these discoveries to industry processes? Yes you can create some crazy materials a low levels, but scaling up the small scale stable processes is difficult.

I put together a gpt which lets you ask questions about and visualize the materials discovered by the GNoME material discovery project discussed here. Fun quick little project.<p><a href="https:&#x2F;&#x2F;chat.openai.com&#x2F;g&#x2F;g-5Kt4lhwvF-unofficial-gnome-materials-discovery-ai" rel="nofollow noreferrer">https:&#x2F;&#x2F;chat.openai.com&#x2F;g&#x2F;g-5Kt4lhwvF-unofficial-gnome-mater...</a>

There is an ongoing effort to simulate living cells using big computers. I wonder if deep learning will get there before others. A system like AlphaZero can start with small cells and keep going up in complexity just how we imagine life came to be...

&gt; The GNoME project aims to drive down the cost of discovering new materials. External researchers have independently created 736 of GNoME’s new materials in the lab, demonstrating that our model’s predictions of stable crystals accurately reflect reality.<p>It seems like a neat project.<p>I wonder, though, what does an unsuccessful prediction look like? They successfully created 736 of the materials. I’m sure they didn’t make 380000-736 bad predictions, hahaha!<p>Would it be interesting to know about materials in their set where fabrication was attempted but didn’t work out? Or maybe it is much more complicated than that; maybe it is assumed that there are crystals in the set that are basically impossible to fabricate for complicated engineering reasons, and but that’s fine because it is just the beginning of the investigation.

For all the automation effort, there&#x27;s always something that has to be done by hand...<p>From the experimental paper: &quot;The XRD sample holders must be cleaned manually when the lab has depleted its stock&quot;

Gray goo next. Please stop this madness and quickly regulate this before we are doomed (: