Silicon is the backbone of today’s semiconductor industry. Despite electronics, silicon photonics plays an increasing role due to the large interest of integrating photonic and electronic devices on the same chip. However, conventional approaches are limited to the surface resulting in 2D planar photonic solutions.

Within the past decades, ultrashort pulse laser structuring has been established as a powerful tool for direct inscription of optical functionalities like waveguides, Bragg structures or artificial birefringence in various glasses and crystals. Moreover, cutting and welding of different glasses has been demonstrated as well with ultrashort pulses. All these processes rely on a well-controlled nonlinear energy deposition inside the transparent material. However, the transfer of these processes to semiconductors is difficult not only due to the high refractive index and different transparency ranges, but mainly due to the significantly higher nonlinearities, hindering a precisely localized energy deposition inside the material.

Here, we report on our recent investigations to directly inscribe waveguides into the bulk of crystalline silicon by using infrared ultrashort laser pulses. We carefully analyze the induced structural changes as well as the resulting positive refractive index changes of up to 4⨯10^-3 enabling the waveguiding. The guiding properties are characterized and correlated with the refractive index profiles for different pulse energies.

In addition, we demonstrate the first semiconductor-metal ultrafast laser welding illustrated with silicon and copper. To do so, we measure the nonlinear propagation of the laser pulses inside silicon. Imaging the light evolution during propagation yields excellent agreement with a semi-analytical model based on self-focusing theory. By precompensating the nonlinear focal shift, we were able to optimize the energy deposition at the interface between the two materials. The resulting welds show remarkable shear joining strengths (up to 2.2 MPa) compatible with applications in microelectronics. Insight into the physical mechanisms involved during the interaction is gained by characterizing the samples after the fracture test.