British biochemist Professor Frederick Sanger was awarded the second Nobel Prize in Chemistry in … [+]
The average person, it seems, is somewhat familiar with generative AI. Those regular consumer products that produce images like the Dall-E are, to some extent, household names. But fewer of us realize how far generative AI can go in introducing new solutions to our world.
This is especially true at the biological level. In the same way that genAI can “hack” the ability to simulate text, photos and videos, it can turn to biology and find new ways to identify biological structures.
Take this example: a team at Microsoft is just figuring out how to use a diffusion model to find new materials that can fit high-tech projects.
“Finding a new material for a target application is like finding a needle in a haystack,” write the authors of a blog post at Microsoft, where they’ve been working on just such a program, something aptly named , MatterGen. “Historically, this task has been done through expensive and time-consuming experimental trial and error. Recently, computer screening of large material databases has allowed researchers to speed up this process. However, finding scarce materials with the desired properties still requires screening millions of candidates.”
In a way, this reminds us why AI is the perfect partner for the Internet. It can take hours, days or even weeks to scour the internet for a particular fact or piece of information – or you can ask a model like ChatGPT and get it in seconds. The Internet is a vast array of information that spans the globe – AI is a super productive researcher who can crunch the results of the Internet in the blink of an eye.
It is simply many degrees more efficient than a human in searching the internet. It is also apparently many orders of magnitude more efficient than humans at finding new types of matter, materials with a specific chemistry, magnetism, conductivity, etc.
Architecture of Diffusion and the Biological World
To understand what MatterGen is doing with biological science research, you need to understand the diffusion model itself.
In general, when generative artificial intelligence is applied to diffusion, you start with a defined object, an image or a protein or some kind of biological item, for example, and you corrupt it, with a “noise” process. You insert noise, to distribute the original article in an abstract. The system then goes back and “discovers” the target, assembling it into a new result, a new item or structure that has the desired attributes or properties that the researcher, the human user, wants.
Microsoft researchers report that when the system was given over 608,000 stable materials from the databases, it was able to generate good candidates for new materials.
Dealing with compositional disorder
One challenge, the team noted, is disordering the composition, where atoms can change their positioning within a particular synthesized material.
“The community has been exploring what it means for a material to be novel in the context of computationally designed materials, as widely used algorithms will not distinguish between pairs of structures where the only difference is a permutation of elements of similarly in their respective countries.” explains the team.
Clear as mud?
As I mentioned before, you can also use ChatGPT to better understand what these researchers are talking about. I put the above into ChatGPT and got this.
“Recently, researchers have been looking into what makes a material ‘new’ when it comes to computer-designed materials. This is important because common algorithms often cannot distinguish between structures that differ only in the way similar elements are arranged in their positions.
I also looked for a concrete example of composition disorder and got this:
“A concrete example of compositional disorder can be found in metallic alloys, such as brass. Brass is made from the combination of copper (Cu) and zinc (Zn). In a perfectly ordered structure, the atoms of copper and zinc will are arranged in a regular pattern. However, in reality, these atoms can randomly exchange sites within the crystal lattice, meaning that some sites that were supposed to have copper atoms could have zinc and vice versa.This randomness in the arrangement of copper and zinc atoms is an example of compositional disorder, which can affect material properties such as strength, corrosion resistance, and electrical conductivity.
I learned a lot about the attributes of brass and why it is soft.
“Brass is generally considered soft, especially when it has a high percentage of zinc.” ChatGPT is provided. “While compositional disorder can affect the material’s properties, bronze’s softness is largely due to the way the copper and zinc atoms interact in its crystal structure. When bronze is made, zinc atoms can replace copper atoms in the crystal lattice. This creates a solid solution, where the different sizes and properties of the copper and zinc atoms disrupt the orderly arrangement of the atoms, making it easier for the layers of atoms to slide over each other. This slip is what leads to the material being ductile (able to be drawn into wires) and malleable (able to be molded into shapes), which are characteristics of softness. So while compositional disorder can contribute to softness by making the structure less stable, the main reason for brass’s softness is the specific arrangement and interaction between the copper and zinc atoms. The relative proportions of these metals in the alloy also play a key role; Higher zinc content generally leads to softer bronze.”
One of the key points here is that some bonds can have combinations of smaller and larger atoms, with the smaller atoms fitting between the larger ones. It is the compositional equality of copper and zinc atoms that leads to some of the attributes of the brass material. The model and other sources also point out that when you add more zinc, the brass becomes softer.
Innovative battery technology
At the start of their announcement of this new generational AI solution, the Microsoft team talks about lithium-ion battery innovations and how the original lithium-ion battery was designed with the discovery of lithium cobalt oxide in the 1980s. As they mention they, lithium-ion batteries are critical to everything from modern smartphones to electric and hybrid vehicles; doing without them would set our world back a bit.
Now, researchers are trying to figure out how to replace traditional lithium-ion batteries with better, more modern battery designs—ones that use less lithium because it’s hard to get and hard to make. take out.
And they are managing to use this biological technology. Going back about a year, you can see the results from PNNL (Pacific Northwest National Laboratory) where researchers obtained a design with 70% less lithium required.
It is likely that we will continue to have better iterations through the use of the AI process, where diffusion shows us what might be possible in chemical engineering.
I’ve written before about some other AI-powered biological solutions, but this one is exciting in part because of the potential for lithium-ion battery advances, and for other mission-critical changes in the way we source enterprise materials.
Think about it in terms of simple logistics – you want a better supply chain, safer materials, higher quality projects and better delivery to customers – and all of this can help us get there.
