RESEARCH NEWS
Electronic / mechanical metal properties
New research from the University of Birmingham ( in the UK ) shows that the electronic structure of metals can strongly affect their mechanical properties .
The research , published on 26 October 2023 in the journal ‘ Science ’ demonstrates experimentally , for the first time , that the electronic and mechanical properties of a metal are connected . It was previously understood theoretically that there would be a connection , but it was thought that it would be too small to detect in an experiment .
Dr . Clifford Hicks , Reader in Condensed Matter Physics , who worked on the study said : “ Mechanical properties are typically described in terms of the bonding between atoms , while electronic properties of metals are described by states that extend across many atoms . The atomic lattice ( the term used to describe the arrangement of atoms ) of a metal and its mechanical properties are generally thought of as being unaffected by which electronic states are occupied and which are empty but , in this work , we show that this is not always a good assumption .”
The researchers from the University of Birmingham and the Max Planck Institute for Chemical Physics of Solids ( in Dresden , Germany ) conducted experiments on the
Photo by Nils Rasmusson on Unsplash .
superconducting metal strontium ruthenate ( Sr2RuO4 ). By measuring lattice distortion as a function of applied stress , the team found that when Sr2RuO4 is compressed by about 0.5 %, a measure of mechanical stiffness known as the Young ’ s modulus decreases by about 10 % and then increases by about 20 % when the material is compressed further . This change corresponds to a new set of electronic states becoming occupied , at a transition that had been identified earlier through electronic but not mechanical measurements .
Dr . Hicks continued : “ Whilst it is completely standard to measure stress-strain relationships in mechanical engineering , it is not something that has been done to study electronic properties . This is because the metals that have interesting electronic properties tend to be brittle , making it hard to apply large forces . Also , large strains are typically required to meaningfully alter electronic properties . In this experiment , samples of Sr2RuO4 were compressed by up to 1 %. To visualise that , imagine taking a metrestick made of granite , and squeezing it until it is 99cm long .”
To overcome these hurdles , the scientists had to build new instrumentation which could measure small and delicate samples and handle cryogenic temperatures , as electronic measurements are more accurate at lower temperatures . This took five years of planning and design .
“ This research , which was funded by the German Research Foundation ( Deutsche Forschungsgemeinschaft ) and the Max Planck Society , is the first of its kind ,” said the University of Birmingham .
“ Now that this experiment has been completed on one material , the scientists are keen to conduct similar measurements on other metals . A version of the machine developed for this project is manufactured by a UK-based engineering company and , as the apparatus is further developed , it may find application in the study of high-strength alloys . This project provides an example of how curiosity-driven , fundamental research can lead to new technology with practical applications ,” it concluded . n
AI hardware processes in higher dimensions
In a paper published on 19 October 2023 in ‘ Nature Photonics ’, researchers from the University of Oxford ( UK ), along with collaborators from the Universities of Muenster , Heidelberg and Exeter , report on their development of integrated photonicelectronic hardware capable of processing three-dimensional ( 3D ) data , substantially boosting data processing parallelism for AI ( artificial intelligence ) tasks .
“ Conventional computer chip processing efficiency doubles every 18 months , but the processing power required by modern AI tasks is currently doubling around every 3.5 months . To cope with this , we need new computing paradigms . One approach is to use light instead of electronics ; this allows us to carry out multiple calculations in parallel using different wavelengths to represent different sets of data ,” explained the University of Oxford .
In groundbreaking work published in the journal ‘ Nature ’ a couple of years ago , many of the same authors demonstrated a form of integrated photonic processing chip that could carry out the key AI task of matrix vector multiplication at speeds far outpacing
Above : Artistic rendering of a photonic chip with both light and RF frequency encoding data .
the fastest electronic approaches . This work resulted in the formation of the photonic AI company , Salience Labs .
Now the team has gone further by adding an extra parallel dimension to the processing capability of its photonic matrix-vector multiplier chips . This ‘ higher dimensional ’ processing is enabled by exploiting multiple different radio frequencies to encode the data , propelling parallelism to a level far beyond that previously achieved .
“ As a test case , the team applied its novel hardware to the task of assessing the risk of sudden death from electrocardiograms of heart disease patients . It was able to successfully analyse 100 electrocardiogram signals simultaneously , identifying the risk of sudden death with a 93.5 % accuracy . The researchers further estimated that , even with a moderate scaling of 6 inputs x 6 outputs , this approach can outperform stateof-the-art electronic processors , potentially providing a 100-times enhancement in energy efficiency and compute density . The team anticipates further enhancement in computing parallelism in the future , by exploiting more degrees of freedom of light such as polarisation and mode multiplexing ,” added the University of Oxford .
First author , Dr . Bowei Dong , in ‘ Materials ’ added : “ We previously assumed that using light instead of electronics could increase parallelism only by the use of different wavelengths , but then we realised that using radio frequencies to represent data opens up yet another dimension , enabling superfast parallel processing for emerging hardware .”
The work was supported by the EU Horizon 2020 PHOENICS project and the EU Innovation Council Pathfinder project , HYBRAIN . The full paper ‘ Higher-dimensional processing using a photonic tensor core with continuous-time data ’ is published in ‘ Nature Photonics ’. n
14 | ismr . net | ISMR December 2023 / January 2024