A Rice research lab’s signature keepsake helped perfect a method for growing patterned diamond surfaces that could help decrease operating temperatures in electronics by 23 degrees Celsius.

In a recent study, Dr. Chander Negi, a postdoctoral research associate in the lab of Texas A&M University professor Dr. Ivan Rusyn, investigated how an emerging “liver-on-a-chip” technology compares with traditional testing methods used in drug safety evaluation.

Superhydrophobic surfaces — those famously “never-wet” materials that make water bead up and roll away — have a stubborn weakness: hot water. Once temperatures climb above roughly 40 degrees Celsius, many superhydrophobic coatings abruptly lose their magic. Instead of skittering off, hot droplets start sticking, soaking into the surface texture and leaving behind wet patches and residue. A new study from mechanical engineers at Rice University describes a surprisingly straightforward fix: Instead of just engineering the surface’s chemistry and texture, they focused on engineering its heat flow.

Researchers at the University of Waterloo have developed a sunlight-driven process that converts plastic waste, including microplastics, into acetic acid—the main component of vinegar—using a bio-inspired photocatalytic system. The approach offers a potential low-emissions alternative to incineration by upcycling mixed plastics into a valuable chemical while helping address pollution in water and other environments, with future potential for scalable, solar-powered recycling and environmental cleanup.

A University of Houston engineering professor is warning that implanted cuff electrodes – widely used in therapy for epilepsy, depression and inflammatory disorders – could trigger unintended nerve stimulation for patients undergoing an MRI scan. 

Scientists have developed a new AI‑based medical imaging method that predicts the bone removal shape needed for cochlear implant surgery using only pre‑operative CT scans. The technique combines two advanced machine‑learning approaches—self‑supervised learning and weakly supervised learning—to learn from “before and after” scans without requiring manual labels. It then refines its predictions using a specialized 3D loss function designed to handle noisy medical data. In tests on 751 scan pairs, the system produced mastoidectomy shape predictions that outperformed several leading medical imaging models, achieving a mean Dice score of 0.72. This advance could improve surgical planning, support future navigation and robotic tools, and reduce the need for labor‑intensive image annotation in complex ear anatomy.