A D V A N C E D M A T E R I A L S & P R O C E S S E S | A P R I L 2 0 2 1 5 MACHINE LEARNING FOR AEROSPACE ALLOYS A new research initiative called Project MEDAL (Machine Learning for Additive Manufacturing Experimental Design) aims to accelerate the product development lifecycle of aerospace components by using a machine learning model to optimize additive manufacturing (AM) processing parameters for new metal alloys—and at a lower cost than current approaches. The project is being led by two U.K. entities, Intellegens and the University of Sheffield, along with Boeing, headquartered in Chicago. Intellegens is a University of Cambridge spin-off specializing in artificial intelligence. Project MEDAL’s research will concentrate on metal laser powder bed fusion—the most widely used industrial AM process—and will focus on the key parameter variables required to manufacture high density, high strength parts. The project is part of the U.K.’s National Aerospace Technology Exploitation Program. Intellegens will produce a software platform with an underlying machine learning algorithm based on one of its previous innovations. The project was conducted for a leading OEM and a new alloy was designed, RESEARCH TRACKS developed, and verified in 18 months instead of the typical 20-year timeline, saving about $10 million. While the new method is being developed with aerospace in mind, the team believes it will have applications in other sectors as well. “The opportunity for this project is to provide end users with a validated, economically viable method of developing their own powder and parameter combinations. Research findings from this project will have applications for other sectors including automotive, space, construction, oil and gas, offshore renewables, and agriculture,” says Ian Brooks, AM technical fellow at University of Sheffield North West. intellegens.ai. MORE EFFICIENT MATERIALS FOR ELECTROLUMINESCENCE New research from an international team of scientists offers insight into how electroluminescent materials could be designed to work more efficiently. Two years ago, theoretical chemist Andrew Rappe of the University of Pennsylvania (Penn) visited the lab of Tae-Woo Lee at Seoul National University (SNU) to see if they could develop a theory to explain some of their experimental results. The material being studied was formamidinium lead Machine learning will be used to make 3D printing of metallic alloys cheaper and faster for the aerospace industry. bromide, a type of metal-halide perovskite nanocrystal (PNC). Results collected by the Lee group indicated that green LEDs made of this material were working more efficiently than expected. PNCs like formamidinium lead bromide are used in photovoltaic devices, where they can store energy as electricity or convert electric current into light in LEDs. To make sense of the SNU results, the team developed a computational model of the material’s unexpected efficiency and designed follow-up experiments. Using their new model, researchers found that the PNCs were more efficient if the size of the quantum dots were smaller, although reducing the size also meant increasing the surface-to-volume ratio, creating more surface area prone to defects. The team found that replacing formamidinium with a larger organic cation called guanidiniummade the particles smaller while also preserving structural integrity. Building on this approach, the team found other strategies to improve efficiency, including the addition of longchain acids and amines to stabilize surface ions and adding defect-healing groups to repair any vacancies that might form. Besides Penn and SNU, other researchers came from the Korea Advanced Institute of Science and Technology, Ecole Polytechnique Fédérale de Lausanne, University of Tennessee, University of Cambridge, Universitat de Valencia, Harbin Institute of Technology, and University of Oxford. upenn.edu. A new study explores how a class of electroluminescent materials can be made to work more efficiently. Courtesy of Penn.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==