Scientists at Cornell University have been using a modified household microwave oven to help overcome a significant obstacle to practical 2nm semiconductor production. The resulting microwave annealer borrows inspiration from TSMC’s theories about microwaves and silicon doping with phosphorus. As a result, semiconductor manufacturers could push past a previous phosphorus concentration limit using the newly devised equipment and techniques.
For semiconductor processes to continue shrinking, silicon must be doped with higher and higher phosphorus concentrations to facilitate accurate and stable current delivery. As things stand, with the industry starting mass production of 3nm components, traditional annealing methods are still working effectively. However, as the industry reaches beyond 3nm, concentrations of phosphorus that are higher than its equilibrium solubility in silicon need to be ensured. As well as achieving higher concentration levels, consistency is vital in making functional semiconductor materials.
TSMC had previously theorized that microwaves could be used in the annealing (heating) process to facilitate the increased doping concentrations of phosphorus. However, microwave heating sources previously tended to produce standing waves, which are bad for heating consistency. In simple terms, previous microwave annealing devices heated their contents unevenly.