Glass and crystal may look the same at first glance, but their structures are very different when viewed under a microscope. Crystals have perfectly ordered, repeating patterns of atoms, whereas glasses have a chaotic structure similar to fluids.
In the world of physics, glass phases are considered a special form of matter. On short time scales, glasses behave like solids. However, over long periods of time, glass behaves like a liquid. It exists at the exact intersection of solid and liquid, and its properties are elusive in conventional classifications of substances.
Quantum researchers and scientists have long been puzzled by the transformation of glass from a liquid to a special solid. Perhaps even more elusive is a phase of material called Bragg glass that exhibits structural properties of both ordered and disordered structures.
Recent advances in the fields of artificial intelligence and machine learning are providing scientists with an opportunity to solve long-standing scientific mysteries. AI and ML algorithms can sift through vast data sets to identify complex correlations and discover patterns that are impossible with traditional analysis methods.
Scientists at the U.S. Department of Energy's (DOE) Argonne National Laboratory, along with collaborators at Stanford University and Cornell University, used ML to discover experimental evidence proving the presence of Bragg glass in the material. did.
Scientists were able to investigate the glass's properties using large amounts of X-ray scattering data and new ML data analysis tools developed at Cornell University. Although theoretical predictions of Bragg glass phases have existed for more than 30 years, concrete experimental evidence has been lacking until now.
“We can collect large amounts of X-ray data in a short amount of time, but when we manually analyze the data, we can't see the trees for the trees,” said author Ray Osborn, a senior physicist in Argonne University's Department of Materials Science. “It may become impossible to see.” About research. ,war “His combination of state-of-the-art X-ray and computational techniques allowed us to reveal unique features of the Bragg glass phase.”
For this experiment, the scientists searched for the elusive Bragg glass state within ErTe3's crystalline matrix. This state is known to have a certain long-range order in its structure, which scientists call charge density waves (CDWs).
About 30 years ago, it was theorized that CDW materials could maintain a Bragg glass state if “chaos” could be introduced into the ordered state of the structure. In this experiment, the scientists randomly distributed palladium atoms in the structure to create some sort of disorder.
(Nico El Nino/Shutterstock)
X-ray scattering was performed on the disordered samples and 3D structural data of each crystal was recorded. To analyze how the structure changes, samples of data were collected at temperatures ranging from 30K to 300K.
They then used machine learning tools to analyze hundreds of gigabytes of data and found that at a certain transition temperature, the sample freezes into a state with significant long-range order, while also displaying local features. It has been confirmed that This confirmed the experimental detection of the Bragg glass phase.
Insights from this experiment also highlight the power of AI and ML algorithms in scientific discovery in the digital age. This discovery may contribute to a better fundamental understanding of phase transitions in materials. It will also help advance the fields of superconductivity and magnetism. Furthermore, the results may lead to the development of new materials for various applications.
Related products
Argonne's EXSCLAIM! software can capture microscopic images from scientific papers
University of Oxford study finds large-scale language models (LLMs) pose risks to science with incorrect answers
Cost-effective GPT-3 fork released to scientists
related
