Double-Pulse Ultrafast Laser Bulk Modification of Silicon

In recent years, more and more insight has been gained about the propagation of high-intensity ultrashort laser pulses in silicon [1]. Filamentation [2], an inevitable consequence of the extreme nonlinear properties of silicon, has been identified as the main reason for the difficulties to produce internal modification. Due to this nonlinear interaction, delocalization of energy deposition is observed even at pulse energies in the nanojoule range, which in turn hinders modification. To address these limitations, various methods have been proposed, including the use of solid-immersion lenses [3], longer pulses in the picosecond range [4]–[6], or pulse trains with pulse-to-pulse delays short enough to benefit from cumulative processes [7], [8]. To date, demonstrations of modifications solely in the bulk of silicon have been achieved only by using numerical apertures NA ≥ 0.85. This is a major bottleneck, as it implies limited working distance and thus restricted writing depths. Therefore, the ability to reduce the required NA is an important step to meet the demands of emerging industrial processes. A promising approach to achieve bulk modifications with NA < 0.85 is an adapted temporal pulse profile. While initial steps have been taken in the field of temporal pulse shaping to identify suitable conditions for bulk modification [4]–[8], important aspects remain unexplored.