PNNL researchers have developed a new, physics-informed machine learning model that accurately predicts how heat accumulates and dissipates during friction stir processing.
A switchable single-atom catalyst is activated in the presence of surface intermediates and reverts to its stable inactive form when the reaction is completed.