material science<\/a> is to synthesise a sample, measure it, learn from it, and then go back and make a different sample and keep iterating that process,\u201d Yager said. \u201cInstead, we made a sample that has a gradient of every parameter we\u2019re interested in. That single sample is thus a vast collection of many distinct material structures.\u201d<\/p>\nThe team then brought the samples to NSLS-II, which generates ultrabright X-rays for studying the structure of materials.<\/p>\n
\u201cOne of the SMI beamline\u2019s strengths is its ability to focus the X-ray beam on the sample down to microns,\u201d said NSLS-II scientist and co-author Masa Fukuto.<\/p>\n
\u201cBy analysing how these microbeam X-rays get scattered by the material, we learn about the material\u2019s local structure at the illuminated spot. Measurements at many different spots can then reveal how the local structure varies across the gradient sample. In this work, we let the AI algorithm pick, on the fly, which spots to measure next to maximise the value of each measurement.\u201d<\/p>\n
As the sample was measured at the SMI beamline, the algorithm created a model of the material\u2019s numerous and diverse set of structures, without human intervention. With each subsequent X-ray measurement, the model updated itself, making every measurement more accurate.<\/p>\n
In just a few hours, the algorithm identified three key areas for researchers to study in more detail. These key areas were imaged with the CFN electron microscopy facility, which uncovered the rails and rungs of a nanoscale ladder, among other new features.<\/p>\n
It is estimated that the researchers would have needed a month to make this discovery using traditional methods, compared to the six hours taken for the experiment.<\/p>\n
\u201cAutonomous methods can tremendously accelerate discovery,\u201d Yager said. \u201cIt\u2019s essentially \u2018tightening\u2019 the usual discovery loop of science so that we cycle between hypotheses and measurements more quickly. Beyond just speed, however, autonomous methods increase the scope of what we can study, meaning we can tackle more challenging science problems.\u201d<\/p>\n
Future uses of the team\u2019s autonomous research method<\/h3>\n \u201cMoving forward, we want to investigate the complex interplay among multiple parameters. We conducted simulations using the CFN computer cluster that verified our experimental results, but they also suggested how other parameters, such as film thickness, can also play an important role,\u201d Doerk said.<\/p>\n
Now, the team is applying its autonomous research method to more challenging material discovery problems in self-assembly. Autonomous discovery methods are adaptable and can be applied to nearly any research problem.<\/p>\n
\u201cWe are now deploying these methods to the broad community of users who come to CFN and NSLS-II to conduct experiments,\u201d Yager said. \u201cAnyone can work with us to accelerate the exploration of their materials research. We foresee this empowering a host of new discoveries in the coming years, including in national priority areas like clean energy and microelectronics.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"
Scientists have used Artificial Intelligence to rapidly facilitate the self-assembly of new nanostructures.<\/p>\n","protected":false},"author":18,"featured_media":28897,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[24429],"tags":[570,833],"acf":[],"yoast_head":"\n
Artificial Intelligence used to facilitate self-assembly of new nanostructures<\/title>\n \n \n \n \n \n \n \n \n \n \n \n \n\t \n\t \n\t \n \n \n \n \n \n\t \n\t \n\t \n