Cavity-enhanced detection of surface photovaporization

Cavity-enhanced detection’-3 can be used as a sensitive means of monitoring laser-surface interactions. In particular, it can detect the amount of material ejected from a solidor liquid-film surface when a laser beam is focused onto it. The tiny ablation plume is monitored inside an optical cavity by a fixed frequency continuous-wave (cw) probe beam. The photovaporization signal is produced by changes in the optical path length, absorption, and scattering of the probe beam by the laser-ablation plume. The signal is amplified by the quality factor of the high finesse optical cavity, since the probe beam frequency is resonant with a cavity mode. The new technique can be used to monitor and control (via feedback) laser etching, micromachining, and annealing of surface films. In addition, delicate surfaces can be cleaned of particulate and micronsized contaminants. In this letter we report details of laser-ablation plume formation for a variety of refractory and low melting point materials. The absorption of a tightly focused pulsed laser beam on a surface creates localized surface heating. This localized heating results in both melting and vaporizing. Thermal surface effects have been previously modeled4 for the case of steady-state material removal. The calculation is based on the assumptions of one-dimensional fluid flow of the molten material on the surface and of an optically thin vapor plume. Extrapolation of the calculations to pulsed laser ablation indicates that liquid expulsion from the surface can be greater than the vaporization losses, even at the low laser powers used in the present study. Figure 1 depicts the experimental apparatus. The vaporization is produced by a tunable pulsed dye laser which is focused onto the sample located inside a spherical etalon. The pulsed dye laser produces pulse energies in the range of 0.35-35 PJ at ;Z = 640 nm with a pulse width of 10 ns. When focused to a spot size of 20 ,um, this corresponds to an intensity of 0. l-l 1 W/pm2. The material removed from the surface is monitored by a weak KeNe-laser probe beam. In the present experiment, a mode-matched HeNe beam is split into two beams, and both are injected into the etalon. One beam is used as the probe beam and grazes the sample surface above the focal spot of the pulsed dye laser beam. The other beam is used to stabilize the &talon by locking it to the side of a transmission fringe and, thereby, to the I-IeNe-laser frequency, In all our measurements the focused laser spot size was kept the same. Using for size reference a calibrated Ronchi grating as the target, the spot size was measured to have a diameter on the surface of 20 pm. The photovaporization signal as a function of position on the Ronchi grating is shown in Fig. 2. More than 25 different materials were examined to investigate the various vaporization signals. Where possible, the samples were polished with a fine grit and occasionally were also acid etched. The most interesting features were obtained from metals and opaque films. Three features, namely, the magnitude of the vaporization amplitude, the temporal evolution of the plume, and the velocity of the ejected material were studied in detail. The results are compiled in Table 1. The strongest photovaporization signals were obtained from aluminum surfaces which had an 80% peak attenuation of the probe beam. The weakest signal occurred for tungsten, which produced a 1.5% peak attenuation. Transparent materials, such as glass and Plexiglas, and highly reflective materials, such as Teflon and Spectralon, did not produce any observable signals at the present low laser powers. The most common temporal structure observed in the vaporization signals consisted of a single asymmetric peak, as shown for aluminum in Fig, 3. It has a rise time of about 0.5 ms and a much slower decay time of about 14 ms. For some samples, a second broad peak, centered about 2 ms after the laser pulse, occurred in the averaged data, as is shown for lead in Fig. 3. This broadened peak displayed considerable variations in shape when observed on a shotto-shot basis. For the dye laser pulses with the highest