Posted in | News | Quantum Physics

Machine Learning Can Analyze Potential Quantum Materials Using Virtual Laboratory

Before discovering that carbonized cotton thread burned long and bright in an incandescent light bulb, Thomas Edison notably attempted hundreds of materials and failed thousands of times. Often, experiments are tedious (Edison’s team took 14 months) and costly (in today’s money, the winning combination cost about $850,000). Time and cost exponentially increase while building the quantum materials that will revolutionize modern computing and electronics.

Machine Learning Can Analyze Potential Quantum Materials Using Virtual Laboratory

To make the discovery of quantum material feasible, scientists look for thorough databases as their virtual lab. A new database of understudied quantum materials has been made by scientists at Pacific Northwest National Laboratory (PNNL). This database offers a path to discover new materials that could power gadgets far more efficiently than the lightbulb of Edison.

Beyond Edisonian Trial and Error

We wanted to understand a general class of materials that have the same crystal structure, but different properties depending on how you combine and grow them,” stated materials scientist Tim Pope. This class of materials is called transition metal dichalcogenides (TMDs). It comprises thousands of likely combinations, each of which needs a long reaction of one week to grow flakes of material the size of glitter.

In comprehending what it can do, producing the material is just the first step. According to PNNL computational scientist Micah Prange, every flake is “really small, really delicate,” and quantum features will only arise when researched at super-low temperatures. Basically, “a whole research program could go into each flake.”

In spite of the challenge involved in making and measuring them, every combination holds promise to dramatically enhance batteries, electronics, quantum computing devices, and pollution remediation.

Prange noted that one can imagine the flakes as “fancier graphene with a richer phenomenology and more practical possibilities.” Light, tough, and flexible, graphene has been touted as the material of the future, with applications in all fields, from aerospace to wearable electronics.

The varied properties across this class of materials mean that as we better understand them, one of the combinations could be selected for a desired property and exactly paired to the ideal use or even a brand new application,” said Pope.

Quantum Material Development of the Future

Developing the database started with the Chemical Dynamics Initiative of PNNL, an attempt to utilize PNNL’s power in data science to occupy knowledge gaps left by experimental limitations and measurement challenges.

Such particular quantum materials are developed by different proportions of the 38 transition metals, such as vanadium or tungsten, combined with three elements in the sulfur family. Also, they can be grown in three varied crystal structures, implying there are thousands of likely combinations, all with discreet properties.

With the use of a modeling type known as density functional theory, the scientists computed the properties of 672 unique structures with a sum of 50,337 individual atomic configurations. Prior to this study, there were fewer than 40 studied configurations, with just a rudimentary comprehension of their properties.

Models can work out the quantum mechanics of how atoms are arranged. From this, you can say if the material will conduct electricity or be transparent or how hard the material will be to compress or bend,” stated Prange.

With the use of the database, PNNL’s scientists showed notable variations in the magnetic and electrical behaviors among varied combinations. Significantly, the scientists also found other trends as they varied the transition metal, along with a new comprehension of transition metal chemistry at the quantum level.

Quantum Combinations for Machine Learning

When the crystal structure was overlaid with the database, it matched perfectly,” added Pope, while speaking about the PNNL-grown flakes that are starting to confirm the modeling outcomes.

The idea was really to develop a big data set of theoretical simulations so we could use data analytics to understand these materials. The immediate value of the project is that we did enough different cases to efficiently use machine learning,” stated Prange.

The open-source dataset is published in Nature Publishing Group’s journal “Scientific Data,” which provides scientists a powerful starting point for discovering associations between initial structures and conforming properties. Using this data, they can down-select particular materials for research.

This project is one example of how we can use a large computational dataset to guide the experimental research. Projects like this provide critical data to the machine learning community and could streamline materials development. It is exciting to think about what needs to be understood next to enable synthesis of these materials with atomic precision.

Peter Sushko, Chief Scientist, CDI

The study was financially supported by the Chemical Dynamics Initiative at PNNL.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.