Artificial intelligence has no limits. It has spread to all corners of the planet and brought about remarkable changes. It is used in face recognition, auto correct/spellchecker, Google maps, and so on. The possibilities are endless. Some Ingenious AI methods that can assist in discovering new compounds.
It is not just the technological developments that have led to the discovery of rare earth compounds (also known as rare earth elements or rare earth metals).
For Your Information!
There are 17 rare earth elements (REE), including scandium and yttrium, that are nearly unrecognizable gleaming silver-white soft heavy metals.
In this article, we learn how ingenious AI methods have paved the way for the discovery of new rare earth compounds.
Using ingenious AI methods and advancements in the field, scientists have explored a variety of materials. To assess the stability of new rare earth compounds, researchers from Ames Laboratory and Texas A&M University trained machine learning (ML).
According to materials scientist Prashant Singh:
“Machine learning is really important here because when we are talking about new compositions, ordered materials are all very well known to everyone in the rare-earth community.
However, when you add disorder to known materials, it’s very different. The number of compositions becomes significantly larger, often thousands or millions, and you cannot investigate all the possible combinations using theory or experiments.”
LDRD (Laboratory Directed Research and Development) program at Ames Laboratory brought forward the work. The framework (that was discovered) relies on current state-of-the-art methods to experiment with compounds and understand their chemical instabilities.
Did You Know?
Since the middle of the 20th century, Ames Laboratory has been a leader in research on rare earth elements.
With the help of innovative AI methods, scientists have been able to find and study rare earth elements and materials that can be used in energy storage, clean energy technologies, and permanent magnets.
Wondering how? For those who don’t know, machine learning (ML) is a type of artificial intelligence-driven by algorithms and based on data usage.
Scientists at Ames used their Rare Earth database (RIC 2.0) and high-throughput Density-Functional Theory (DFT) to build an ML model.
It is necessary to understand that high-throughput screening is a computational scheme that enables researchers to analyze and test a variety of models in a matter of seconds.
This model was specifically developed to test new rare earth compounds.
(DFT) Density-Functional Theory is a quantum mechanical method used to study the thermodynamic and electronic properties of various body systems.
By Ingenious AI methods such as ML (machine learning) use regression learning to analyze the phase stability of compounds based on this data.
Lowa State University graduate student Tyler Del Rose conducted foundational research for DB (database) by writing algorithms to search the web for information and supplement the database and DFT calculations.
He has also worked on AI predictions’ experimental validation and improved ML-based models by ensuring they are representative of reality.
Let’s move on to Prashant Singh who described:
The material analysis is based on a discrete feedback loop in which ingenious AI methods (ML) are updated and that too, using new DFT database.Prashant Singh
Don’t forget that the database is built using phase information from experiments and their real-time structural data. The transfer of data from one step to another is ensured by this process.
Yaroslav Mudryk, the project supervisor, shared some thoughts here. According to him:
“The framework was excitingly introduced/designed to find and inspect new rare earth compounds due to its technological importance. But, the execution and usage are not limited to rare earth elements (REE).
Interestingly, the same approach can be used to train the new ML model (a form of ingenious AI methods) to optimize mechanical behaviors, anticipate magnetic properties of new rare earth compounds, and process transformative manufacturing controls.”
He went on to say that
“This was not just about discovering the rare earth metals. The main focus was to design a FRAMEWORK to predict new rare earth compounds.”
One of the best aspects of this experiment is that the team is still exploring new frameworks, tools, and ways to discover and inspect compounds.
As a result of this amazing experiment, we can recognize a great advancement in artificial intelligence. Research done at Ames Laboratory was remarkable and had been documented all time.