Instead of stumbling upon materials by accident like rubber, or taking months or years of trial and error to experiment while we use up Earth's resources, scientists are using databases of advanced material technologies, with algorithms to theoretically combine materials that would have taken exponentially longer to discover.  The software they use can also mine already published papers to see what has and hasn’t been done, reducing their research time significantly, while sharing necessary information with industry colleagues, globally, to move toward sustainable development goals.

Developing AI Methods of Experimenting in Record Time

The team of scientists at the Department of Energy’s SLAC National Accelerator Laboratory, the National Institute of Standards and Technology (NIST) and Northwestern University, discovered three new blends to form metallic glass 200 times faster than before, as an alternative to steel.

The reason for the quick timing? SLAC’s Stanford Synchrotron Radiation Lightsource (SSRL) is an x-ray beam which was used to scan the alloys so the data could be fed into the machine learning system, "which generated new results and then these were used to create a new sample which was then scanned by the x-ray beam and so on until the best material was made." (Forbes).  

A New Type of Artisan 

Artisans know best what takes apprentices years to follow, simply by trial and error.  One goes to one shop instead of another, because there is “just something” about that product.  There is the idea that quality meets price, but oftentimes if one goes to a discount shop, the same item on the shelves seemingly didn’t turn out right with little to no explanation.  Using advanced material technologies, we can think of a need, think of a resource, think of saving that resource and what else could match it in usability and practicality, and how we could put that material towards the achievement of global goals for sustainable development. 

Scientists at the University of Liverpool have announced a new colleague working non-stop during the lockdown.  Meet Robo-chemist!  "The £100,000 programmable researcher learns from its results to refine its experiments." says the BBC article.  Robo-chemist is working at the lab during social distancing while the human scientists do their research and experiments from home. 

The new employee of the year "could make scientific discovery 'a thousand times faster', scientists say." 

But this type of machine won't take the place of people.  It allows people to work in urgency to find sustainable options and healthcare breakthroughs for our most crucial and demanding needs.

Human scientists invite the idea of having the new colleague and state they would love to have these in different locations around the world with a centralized brain so many people could be helped and everyone has the technology.  Not all places and laboratories have the time and means to test at the rate necessary to keep up with the rest of the globe.  Having this type of machine in many places would mean faster and safer global access to new materials discovery necessary for sustainability.

artificial intelligence materials discovery

Rapid Experimental Timing

The same team at Leverhulme Centre for Functional Materials Design, Materials Innovation Factory and Department of Chemistry, University of Liverpool, Liverpool, UK, and have recently published the scientific findings on about the work that Robo-chemist has done.  There are pages of scientific data running experiments that man could only do in a fraction of the time and with more possible human error.  One being the "[we use] a mobile robot to search for improved photocatalysts for hydrogen production from water. The robot operated autonomously over eight days, performing 688 experiments within a ten-variable experimental space, driven by a batched Bayesian search algorithm".  They offer a detailed explanation of the time of workflow in their findings, far less time than could be achieved by man.  Any machine that could perform this many experiments in rapid timing, could increase the rate at which man discovers new materials, before the resources we have run out.

The Possibilities are Infinite

Material AI aims to share in the process of increasing the world's capabilities and developing faster and more efficient ways of discovering new materials that are urgently needed.

Artificial intelligence can permit a database so swift that scientists would be able to research and find new sustainable materials by taking advantage of not redoing, and as a result, creating something new.  Machine learning allows us to access and process large amounts of data that can lead to new achievements in the digital world, economic growth and a healthier environment.  We can go longer periods of time with similar resources as the ones the planet has exhausted, also a key factor in the desire to speed up the materials discovery process.  The time it takes to discover something new for something that could improve global warming for example, is an urgent matter for man, animal, and global environment.  This will also positively impact the global economy. 

By accelerating the material discovery process, the time it would take to improve, invent, invest, test, patent, and package new materials that lean into global sustainable development goals, would be significantly reduced and long, healthy lifestyles for man, animal, and Earth are really dependent on it.  There is pressure for the emergence of many new materials to alleviate the ones already exhausted.  We need faster materials discovery to see what is stable and consistent in terms of viability and practicality and what needs to be replaced in the environment.  There are an infinite amount of possibilities that can be obtained with artificial intelligence leading the way to materials discovery and design using machine learning.