Cafer Acemi, Ph.D. candidate, poses outside Kyle Field at Texas A&M University.
Cafer Acemi, a Ph.D. candidate in the Department of Materials Science and Engineering at Texas A&M University, has received the Acta Student Award, recognizing student-led research published in the Acta family of journals. The award reflects both the caliber of Acemi’s work and the department’s growing role in advancing how new materials are discovered and developed.
“Being selected for this award is meaningful because it recognizes a state-of-the-art alloy discovery campaign that combines computation with real experiments,” Acemi said. “Our department is helping shape the direction of alloy discovery; models guide what to make next, and high-throughput experiments provide the evidence to refine those choices.”
Acemi earned the recognition for his paper, “Multi-objective, multi-constraint high-throughput design, synthesis, and characterization of tungsten-containing refractory multi-principal element alloys.” The study advances a faster, more rigorous approach to discovering metals that retain strength and stability in extreme heat while accounting for early feasibility constraints that determine whether strong lab results translate into real-world processing routes.
Research behind the recognition
Developing high-temperature alloys faces a simple problem with a difficult reality: there are countless possible compositions, and traditional trial-and-error methods take time. Acemi’s work addressed that challenge through a closed-loop discovery workflow that paired computation with experiment at every step.
“Traditional trial-and-error doesn’t scale to the complexity of modern alloy design,” said Dr. Raymundo Arroyave, associate department head for research, materials science and engineering and a co-author on the study. “A closed-loop computational-experimental approach provides a principled alternative: iteratively reduce uncertainty, narrow the search to the most promising regions and strengthen confidence with each cycle. That is how discovery becomes both faster and more reliable.”
The team began with high-throughput thermodynamic modeling, a computational method that quickly evaluates thousands of potential material compositions to identify the most promising ones, prioritizing those expected to remain stable at high temperatures while meeting multiple constraints. Co-author Brent Vela described how early computational screening eliminated weak candidates before any physical testing began.
“The modeling allowed us to discard alloys that likely wouldn’t work before we ever made them,” Vela said. “By narrowing the field early, we avoided costly trial-and-error experiments and accelerated the path to deployable materials.”
To evaluate high-temperature strength, the team built on a physics-informed model as a baseline, then improved its accuracy by updating it with experimental measurements and incorporating statistical learning methods based on algorithms that automatically refine their predictions as new data arrive.
“Machine learning is powerful, but without physics it can wander outside reality when asked to predict something new,” Arroyave said. “Starting from first principles keeps the model grounded, and then the data sharpen it in a disciplined, efficient way.”
Rather than selecting candidates by guesswork, the workflow used algorithmic decision-making to recommend which compositions to synthesize and characterize next.
“By ‘closing the loop,’ we continuously feed experimental outcomes back into our optimization algorithm so the system learns from each result if it will be a success or a failure and then algorithmically recommends the best next compositions to test,” Vela said.
(From left to right) Dr. Ibrahim Karaman, Brent Vela, Cafer Acemi and Dr. Raymundo Arroyave pose together in front of the vacuum arc melter in a Texas A&M materials science and engineering laboratory.
Designed to translate
Throughout the work, feasibility constraints were treated as design requirements rather than afterthoughts. Candidate alloys were assessed not only for high-temperature performance but for realistic pathways to processing and scale-up. Dr. Ibrahim Karaman, department head and co-author, underscored why that distinction matters.
“What makes this significant is that it treats deployment as part of the problem, not something you worry about later,” Karaman said. “It means the alloys we advance aren’t just high-performing; they’re chosen with a realistic path to being made, scaled and used.”
A student-led milestone
The Acta Student Award recognizes research where the student made a major contribution to the published work. For Acemi, the recognition reflects both scientific rigor and research leadership in driving a complex, multi-stage effort that integrated computational screening, experimentation and practical constraints into a repeatable discovery framework.
Work is already continuing beyond the paper.
“This project shows what’s possible when discovery is designed with manufacturing realities in mind, and we’re applying the same approach across a wider set of materials and applications,” Karaman said. “The down-selected alloy has already been gas atomized and is being printed; success now means repeatable processing and dependable performance in component-relevant form.”
About the Acta Student Awards
The Acta journals — Acta Materialia, Scripta Materialia, Acta Biomaterialia and Materialia — present up to 16 awards of $2,000 each year, with four awards designated per journal. Eligible papers must have been accepted and appeared online in ScienceDirect within the calendar year, and the student must have made a major contribution to the published research.
