Saturday, February 2, 2019

Supercomputing helps study two-dimensional materials - (Gauss Centre for Supercomputing)

graphene sheets
A multi-layer of lithium atom-  Image CREDIT Dr. Mahdi Ghorbani-Asl, HZDR. See M. Kühne, et al., Nature 564 (2018).


Supercomputing helps study two-dimensional materials - (Gauss Centre for Supercomputing)


Whether it is high-temperature superconductors and the storage of improved energy to bendable metals and tissues that can fully fuse, scientists study materials and understand the physics of interacting atoms in solids and finally find ways to improve the materials we use in every aspect of daily life. The high-performance computing helps researchers understand experiments to monitor the movement of lithium atoms in real-time in bi-layer graphene, paving the way for the design of new materials for batteries and other electronic devices.

The extent of material science research is not in alchemical trial, alkaline testing and error, though; to better understand and improve content today, researchers should be able to study the physical properties at the atomic level and in extreme conditions. As a result, researchers have rapidly come to rely on simulations to complement or inform experiments in the properties and behaviors of the material.

The researchers of the Helmholtz-Zentrum Dresden-Rossendorf under the leadership of a physicist Dr. Arkady Krasheninnikov and partners with experimental experts were able to answer basic questions about the properties of materials and they were able to observe a major breakthrough experimentation in real time lithium atoms' behavior when placed between two graphene sheets. A graphene sheet is that the researchers consider a 2D material because it is only an atom thick, which has made it possible to observe lithium atomic motion in a transmission electron microscopy (TEM) experiments.

The team of Krasheninnikov was able to use the Hazel Hen supercomputer- High-Performance Computing Center Stuttgart (HLRS) with access to supercomputing resources through the Gauss Centre of Supercomputing (GCS) to simulate, expand and confirm the experimental results of the team. The collaborative work and study report were recently published in Nature.
Dr. Arkady Krasheninnikov said in a statement “2D material exhibits exciting and useful properties and can be used in various applications, not only as support in transmission electron microscopy (TEM). Essentially, 2D content materials are state-of-the-art research. There are thousands of potentials about a couple of these materials, and about 50 have actually been made.

Under the microscope

In order to understand the 2D experimental material, researchers now use TEM. The method allows researchers to suspend the thin, small pieces of material, then run a high-energy electron beam on it, which ultimately creates a magnified image of the material that researchers can study, much like film projector that takes pictures from the reel and projects them on a big screen. With this approach in material, experimentalists can better estimate the status and arrangement of charts and atoms.
The high-energy beam can do more than help researchers to inspect the material, however - it is also a tool to study the electronic properties of 2D material. In addition, researchers can use high-energy electrons from TEM to separate atoms from a substance with high precision to see how the behavior of the material changes depending on the structural change.

Recently the experimentists of Max Planck Institute for Solid State Research, Stuttgart and University of Ulm wanted to better understand how lithium particles interact between two atom-thin graphene sheets. Having better lithium understanding, or keeping lithium among layers of any other material helps researchers to develop new ways to design better battery technologies. The experimentalists obtained data from TEM and asked Krasheninnikov and his colleagues to rationalize the experiment by using simulations.

Simulation allows the researchers to see the atomic structure of material from different angles, and they can also help speed up the test-and-error approach to designing new materials through the experiment. "Simulations cannot complete the job, but they can really limit the number of potential variants, and show the direction on which route to go. Simulation saves money for fundamental research and people working in the industry. And, as a result, computer modeling is getting more and more popular”, said Krasheninnikov.

In this case, Krasheninnikov and his colleagues found that the atomic coordinates of the laboratories or the position of the particles in the material will not be stable, which means that the material will define the laws of quantum mechanics. By using simulation data, Krasheninnikov and colleagues proposed a different atomic structure, and when the team re-runs the experiment, then it got a perfect match with simulation.

"To understand the atomic structure based on experimental results, sometimes you do not need a higher principle, but at other times it is impossible to understand the structure without actually having accurate computational approaches that go hand in hand with experimentation", said Krasheninnikov.

The experimentalists and experts were able to see in real-time for the first time how lithium graphics behave when two graphene sheets are placed between them, and with the help of simulations, get insights into how to organize the atoms and how to identify their arrangement. It was previously believed that in such a system, lithium would be structured as a single atomic layer, but simulations showed that lithium could produce two or three layers, at least two-layer graphene, leading researchers to look for new ways of research to improve battery efficiency.

Charging forward

Krasheninnikov said that while simulation has made great progress in the last decade, there is still room for improvement. The team can periodically execute the first principles of the 1,000-Atom system to inspect the short-term (nanosecond time scale) material interaction. Big core matters in the next generation of supercomputers will allow researchers to include more atoms in their simulation, which means they can make a more realistic model and meaningful slices of the material in question.

According to Krasheninnikov, the biggest challenge relates to the extent to which researchers can simulate physical interactions. In order to study phenomena that occur over longer periods of time, such as how to form stress and spread a crack in metals, for example, researchers need to be able to simulate minutes or even hours to see how materials change. However, researchers also need to take very small time steps in their simulations to develop a precise model of ultra-fast atomic reactions. Just using more computing cores allows researchers to make calculations for larger systems faster, but they can not make each time step faster if a certain level of "parallelization" is reached.

Breaking this predicament will require researchers to rework algorithms to calculate more efficiently at each time step across a large number of cores. Krasheninnikov pointed out that the design of codes based on quantum computing may enable simulations capable of monitoring physical phenomena that occur over long periods of time - quantum computers may be ideal for simulating quantum phenomena. Regardless of the direction of the researchers, Krasheninnikov pointed out that access to supercomputing through PRACE and GCS enables him and his team to maintain progress.


Story Source: ScienceDaily, February 2019: Supercomputing helps the study of two-dimensional materials.


Journal Reference: Nature, 2018; 564 (7735): 234: Reversible superdense ordering of lithium between two graphene sheets




No comments:

Post a Comment