Back to Table of contents

Primeur weekly 2019-08-12

Focus

The quest for European HPC research and innovation funding ...

Quantum computing

Unique electrical properties in quantum materials can be controlled using light ...

Light for the nanoworld ...

Middleware

Ohio Supercomputer Center to host seventh meeting of the MVAPICH Users Group ...

Hardware

Cray reports second quarter 2019 financial results ...

Mellanox Ethernet and InfiniBand solutions deliver breakthrough performance for AMD EPYC 7002 processor-based data centres ...

Phison at the forefront of PCIe Gen4 storage market with a portfolio of products ...

Researchers embrace imperfection to improve biomolecule transport ...

2nd Gen AMD EPYC processors set new standard for the modern data centre with record-breaking performance and significant TCO savings ...

HPE ProLiant shatters 37 world records ...

Cray awarded contracts with the U.S. Army Research Laboratory and the Army Engineering and Research Development Center ...

Cray Shasta supercomputer to power weather forecasting for the U.S. Air Force ...

GRC partners with Prasa to bring liquid immersion cooling to data centres in India ...

Excelero debuts Excelero NVEdge, software for creating NVMe All Flash Arrays (NAFA) from high-availability (HA) servers ...

Excelero honoured with Flash Memory Summit 2019 Best of Show Award for third year in a row ...

Simulation technique can predict microstructures of alloy materials used in jet engines - before they are made ...

Supermicro now offering AMD EPYC 7002 series processor-based systems to customers who want to transform their data centres ...

Boston now offers AMD EPYC 7002 series processor-based systems to customers ...

Lenovo and Intel announce multiyear global collaboration to extend HPC and AI leadership ...

Xilinx expands Alveo portfolio with industry's first adaptable compute, network and storage accelerator card built for any server and any Cloud ...

Penguin Computing expands Altus product family with AMD EPYC 7002 series processor-based systems, reaching new levels of data centre performance ...

Applications

Stanford researcher develops data standards for brain imaging and applies rigorous computational methods to work ...

HPE advances its intelligent data platform with acquisition of MapR business assets ...

Turbulence meets a shock ...

The Cloud

UC San Diego, UC Berkeley, and University of Washington announce 'CloudBank' Award ...

Simulation technique can predict microstructures of alloy materials used in jet engines - before they are made


The new "First-principles phase field" method to better predict complex microstructures of alloys. Credit: Yokohama National University.
1 Aug 2019 Yokohama - Japanese researchers were able to rapidly and accurately predict the microstructure of Nickel - Aluminum (Ni-Al) alloys that are commonly used in the design of jet engine turbine parts. Predictions of the microstructure of these alloys have so far been time-consuming and expensive. The findings have the potential to greatly advance the design of materials - made up of a range of different alloys - that are used to make products in several different industry sectors.

Alloys are durable materials made up of two or more metals. The current high cost and design limitations of traditional alloy manufacturing processes have driven the need to create more efficient design methods. One key challenge has been how to accurately predict an alloy's microstructure - the very small-scale structure that is only visible by microscope - which can greatly influence physical properties such as strength, toughness, resistance to corrosion, hardness and/or wear and tear resistance.

The authors were able to predict alloy microstructures by using the "first-principle phase field method". This procedure predicts the microstructure of alloys based on the fundamental laws of physics alone - first principles - and then uses those parameters to model microstructure formations (phase field). This is contrary to empirical modelling, or predictions based on experiments or previous observations alone. Furthermore, the researchers conducted their modelling experiments under high temperatures that mimic those of jet engine turbines (~1027oC).

The research was published in Nature Communications on August 1, 2019.

The quest of new materials with desirable properties requires microstructure engineering of materials based on changing several variables, such as composition, morphology, pressure, temperature, doping, casting and forging.

A reliable simulation technique that can help with the design and production of new materials based on a theoretical principle alone could make production faster and cheaper. However, most of the current theories of material design are phenomenological and derived from experimental observations and empirical experiences. These are both time-consuming and expensive.

What makes the first-principles phase field method so advantageous, according to the authors, is that it bridges the accurate small-scale (first principles) calculations and large-scale (phase field) model by renormalization theory, a concept in physics that essentially makes infinite degrees of freedom finite, or continuous variables discrete. In other words, by using their method, they were able to overcome time-consuming and expensive experimental procedures and still produce materials that were in agreement with experimental methods.

"First-principles phase field method was invented as the world's first innovative multiscale simulation technique. Using this method, we were able to successfully predict complex microstructures of any compositions of Ni-Al alloys from first-principles - basic laws of physics - without using any empirical parameter, and our results agree quite well with experiments", stated Kaoru Ohno, corresponding author and professor at Yokohama National University.

Kaoru Ohno and co-authors from the National Institute for Materials Science in Japan said the method can be used to predict mechanical strength of alloys because the local force distributions as well as the microstructures can be easily calculated.

The method that the authors present can also be used to predict microstructures of multicomponent alloys, or alloys that are made up of more than two metals. "These studies highlight the fundamental nature of steels and other alloys that have so far only been demonstrated based on empirical observations. As such, the proposed method is a powerful theoretical tool to quickly predict the most suitable alloy that can realize the desired strength, toughness, ductility, plasticity, lightness, etc. as much as possible", added Kaoru Ohno.

In the future, the authors plan to apply the method to various steel materials and other multicomponent alloys in order to predict the dependence of microstructures and local stress distributions on their initial compositions and better understand their characteristics.

This research was supported by MEXT as a post-K computer project - the HPCI priority issue 7E; Project IDs: hp170268, hp180125.

Source: Yokohama National University

Back to Table of contents

Primeur weekly 2019-08-12

Focus

The quest for European HPC research and innovation funding ...

Quantum computing

Unique electrical properties in quantum materials can be controlled using light ...

Light for the nanoworld ...

Middleware

Ohio Supercomputer Center to host seventh meeting of the MVAPICH Users Group ...

Hardware

Cray reports second quarter 2019 financial results ...

Mellanox Ethernet and InfiniBand solutions deliver breakthrough performance for AMD EPYC 7002 processor-based data centres ...

Phison at the forefront of PCIe Gen4 storage market with a portfolio of products ...

Researchers embrace imperfection to improve biomolecule transport ...

2nd Gen AMD EPYC processors set new standard for the modern data centre with record-breaking performance and significant TCO savings ...

HPE ProLiant shatters 37 world records ...

Cray awarded contracts with the U.S. Army Research Laboratory and the Army Engineering and Research Development Center ...

Cray Shasta supercomputer to power weather forecasting for the U.S. Air Force ...

GRC partners with Prasa to bring liquid immersion cooling to data centres in India ...

Excelero debuts Excelero NVEdge, software for creating NVMe All Flash Arrays (NAFA) from high-availability (HA) servers ...

Excelero honoured with Flash Memory Summit 2019 Best of Show Award for third year in a row ...

Simulation technique can predict microstructures of alloy materials used in jet engines - before they are made ...

Supermicro now offering AMD EPYC 7002 series processor-based systems to customers who want to transform their data centres ...

Boston now offers AMD EPYC 7002 series processor-based systems to customers ...

Lenovo and Intel announce multiyear global collaboration to extend HPC and AI leadership ...

Xilinx expands Alveo portfolio with industry's first adaptable compute, network and storage accelerator card built for any server and any Cloud ...

Penguin Computing expands Altus product family with AMD EPYC 7002 series processor-based systems, reaching new levels of data centre performance ...

Applications

Stanford researcher develops data standards for brain imaging and applies rigorous computational methods to work ...

HPE advances its intelligent data platform with acquisition of MapR business assets ...

Turbulence meets a shock ...

The Cloud

UC San Diego, UC Berkeley, and University of Washington announce 'CloudBank' Award ...