In order to improve upon the resolution of photolithography, a technique that is used to produce features for today's micro and nanodevices, techniques must move beyond e-beam and deep-UV sources. Multiphoton absorption polymerization (MAP) uses near-infrared light for the creation of complex, three-dimensional features on the sub-100 nm scale. The resolution of MAP can be enhanced further using a two-beam technique called resolution augmentation through photo-induced deactivation (RAPID) to the reach feature sizes as small as 40 nm.

The mechanism and kinetics of photo-induced deactivation are not well understood. To better understand these processes, studies of different photoinitiators have been performed. We find that some photoinitiators are so efficient at deactivation that they are capable of undergoing self-deactivation by addition of another photon from the excitation source. This phenomenon is manifested in a polymerization trend in which feature size has a proportional velocity (PROVE) dependence, the opposite of the conventional velocity dependence. We also demonstrate that the velocity dependence can also be tuned between PROVE and conventional dependences.

Kinetic models have been formulated to account for the observed deactivation. By reconciling experimental data for some sample photoinitiators with the kinetic model through the use of simulations, kinetic rate constants are determined. The self-deactivation efficiency of each photoinitiator was determined. The lifetimes of intermediates in the radical photopolymerization process were also determined. The kinetic rate constants associated with photoinitiators should allow for the customization of photoinitiators for specific applications and make RAPID a more efficient process capable of reaching resolution on the level of 30 nm and below.