5.1 Introduction
In the early years of laser development attempts were already made to shorten the pulse duration initially with special emphasis on spectroscopic applications like e.g. pump-probe techniques. With time an increasing demand for ultra-short laser pulses developed from materials processing applications. The discovery of the self-mode-locked (Kerr Lens Modelocking) Ti:sapphire laser was a milestone for making a reliable ultra-short pulse laser technique available (Spence, 1991). Nowadays laser oscillators on solid state basis provide average powers of up to the 100 W range (Sibbett, 2012) allowing laser materials processing. So far, ultra-short pulse laser machining was done mostly with laser systems with additional amplification, especially on the basis of the chirped pulse amplification technique (Strickland and Mourou, 1985). An oscillator-amplifier laser system working on the principles mentioned above was used for the work described in this chapter.
After the pioneering microstructuring studies on polymers applying femtosecond laser pulses (Küper and Stuke, 1987, Srinivasan, 1987) an increasing number of laser materials processing results for different material classes was published. The most relevant advantages of ultra-short laser pulses compared to longer pulse durations, like the reduction of the laser fluence needed to induce ablation and the improvement of the contour sharpness of the laser-generated structures as a result of a reduced heat affected zone, became obvious. Additionally, even transparent dielectric materials (with a negligible linear absorption) can be laser-processed making use of nonlinear optical absorption phenomena caused by high laser intensities. Significant results of a large number of studies on laser-materials interaction with dielectric materials were reviewed by Balling and Schou (2013) and by Gattass and Mazur (2008). It is worth noting that the reduction of the pulse duration leads to a more deterministic and reproducible processing of the material resulting in clean ablation craters even for those dielectrics showing a significantly larger band gap compared to the laser photon energy. A machining precision down to the 10-nm-level was demonstrated (Lenzner, 1999).
For (linearly absorbing) metals, the heat influence on the contour sharpness can be reduced considerably (Chichkov, 1996). A femtosecond laser machining of metals is possible with comparatively low energy densities (fluences) of the order of 0.1 J/cm2 (Krüger and Kautek, 1999).
At present, various applications of femtosecond laser-treated materials showing distinct surface micro- and nanostructures (e.g. micro/nano holes, ripples, cones) have been developed (Bäuerle, 2011). These partly new structures (like sub-wavelength ripples) can be used for changing the wetting behaviour, for colouring of surfaces, sensor and biomedical applications as reviewed by Sugioka and Cheng (2014) and by Vorobyev and Guo (2013).
All processes mentioned above include removal of material either from the bulk sample itself (surface structuring) or of unwanted species from the surface (laser cleaning) or the (chemical) modification of the material. So far, no femtosecond laser process has been reported which fuses two materials with completely different band gap (dielectric material and metal) together. Such a process was introduced by Symietz et al. (Symietz, 2010) and is described here with the aim to produce metallic implants with a bioactive ceramic coating on the surface to accelerate the incorporation time into the living tissue. The experiments were made with an ultra-short oscillator/amplifier laser system with a pulse duration of 30 fs, a single pulse energy of up to 1 mJ at 790 nm centre wavelength and a maximum repetition rate of 1 kHz.