5.3 Femtosecond laser treatment
It is of fundamental importance for the method of laser-induced fixation (explained below in section 5.4) that the substances involved differ in one essential property, the ablation threshold. In general, if a powder lies on top of a solid plate it can potentially be removed from the surface by the laser energy (laser cleaning) before it can be fixed. On one hand, a laser energy density (fluence) is required to be harmless to the powder because the crystalline structure has to be preserved. On the other hand, the laser fluence should be high enough to cause a strong binding between the powder and the underlying material.
The pulses from a Ti:sapphire femtosecond laser (Femtolasers, Vienna, Austria) have a duration of about 30 fs, a repetition rate of 1 kHz and the infrared centre wavelength of 790 nm. The focussed laser beam had a beam diameter (1/e2) of about 200 µm. The beam has a Gaussian profile. Multi pulse laser ablation thresholds of both substances were measured using the D2-method (Liu, 1982). The titanium alloy Ti6Al4V was found to have an ablation threshold of Fth = 0.15 J/cm2. Above this fluence a thin surface layer of the metal is disintegrated and removed, leaving behind a shallow ablation crater. The modified metal surface can form different characteristic nano- and microstructres, depending on the local fluence. The structures produced on the metal consist of small parallel ridges on the nanometer scale formed at fluences just above the ablation threshold, and larger rods in the dimension of a few micrometers (Fig. 1). The cone like spikes (Fig. 1b) in zones of a large fluence grow as a result of simultaneous ablation and accumulation next to each other. The other unusual structure of nanoscale ripples (Fig. 1c) gradually fades out toward the edge of the ablation zone. There is no obvious sign of melting, and clearly no lateral mass transport.
Such prominent modifications appear only with the ultra-short laser pulses. All experiments described in this text were made under ambient conditions, room temperature in air, no protective atmosphere was used. The changes in the Ti6Al4V surface shown in Fig. 1 are a much too strong impact, compared with what is needed for the desired powder fixation. A simple melting of a thin surface layer near the Ti6Al4V ablation fluence of 0.15 J/cm2 is desired. The dip coated ceramic layer could not be destroyed at such a relatively low fluence. Figure 2 illustrates a typical dip coated layer of the fine ceramic powder before laser treatment. Only above the ablation threshold Fth = 0.4 J/cm2 it was possible to ablate the ceramic. This is much higher than the value for the titanium alloy, meaning that the ceramic is much more stable than the metal. In the right image of Fig. 2 (optical microscope, Nikon Eclipse L200) a crater produced by laser pulses with an energy density >0.4 J/cm2 is depicted. In the centre of the Gaussian beam the whole GB14 layer was erased, uncovering the underlying metal, including damage visible in the Ti6Al4V. The ceramic layer in this example had a thickness of 35 µm. Such samples, the metal plates covered with a dip coated ceramic layer, were tested for their behaviour at various laser fluences. The surface was irradiated both pointwise and in lines. The latter was done by moving the samples perpendicular to the beam direction with a computer controlled positioning system (Micos, Eschbach, Germany). The lateral movement with a constant speed of typically 100 µm/s resulted in linear patterns written into the ceramic layer. With decreasing fluence the visible trenches disappeared until, below 0.4 J/cm2, no more influence of the laser radiation was visible on the sample surface. The interesting process remains invisible because the fixing of ceramic particles occurs under the otherwise undisturbed surface of the powder layer.