Characteristics of optical emission intensities and bubblelike phenomena induced by laser ablation in supercritical fluids

In previous works, we examined the characteristics of laser ablation in liquid H₂O.³⁶⁾ In this paper, we report the characteristics of laser ablation in supercritical H₂O in comparison with those in liquid H₂O at the same pressure. The experimental results in liquid H₂O were not the data shown in previous papers but were obtained in this study at a water temperature of 230 °C. The critical point of H₂O is at a pressure of 22.1 MPa and a temperature of 374 °C. For the irradiation of the YAG laser pulses in supercritical and subcritical H₂O, it was difficult to determine the trajectory of the laser beam and to focus it. We adjusted the position of the focusing lens to form a circular ablation crater with a diameter of 1 mm on the target surface when the pressure and temperature of water were 0.1 MPa and 20 °C, respectively. However, we observed the distortion of the shape of the ablation crater when the water was heated to above 300 °C. The defocusing of the laser beam was more serious in the vertical direction (namely, the ablation crater became an ellipse). The vertical size of the ablation crater formed in supercritical H₂O at a pressure of 30 MPa and a temperature of 430 °C was approximately 4.5 mm. In addition, we observed a shift in the ablation point in the vertical direction when we heated the water to a temperature above 300 °C, indicating a bend in the trajectory of the laser beam. The problems in the trajectory and focusing of the laser beam were considered to be caused by the spatial variation in the refractive index of water in the vertical direction. The spatial variation of the water temperature may be the mechanism of the spatial variation in the refractive index. Although we stirred the water inside the vessel, we could not obtain a straight trajectory and the uniform focusing of the ablation laser beam. Accordingly, considering the fact that the physics and chemistry of laser ablation are essentially determined by the laser fluence,³⁷⁾ we adjusted the energy of the laser pulse to keep the same laser fluence of 14 J/cm² on the target surface at all water temperatures.

Figure 1 shows the temporal variation of the optical emission intensity of the plasma produced in supercritical H₂O at a pressure of 30 MPa and a temperature of 430 °C. The maximum intensities of the optical emission images, which were captured at various delay times after the irradiation of the YAG laser pulse, are plotted in Fig. 1. An optical emission image at a given delay time was obtained by averaging 30 results to compensate for the shot-to-shot fluctuation. The temporal variation of the optical emission intensity observed in liquid H₂O at a temperature of 230 °C and the same pressure (30 MPa) is also shown in Fig. 1 for comparison. The amplitudes of the two waveforms were normalized by their maxima. As was reported previously,²¹⁾ the waveform of the optical emission intensity observed in liquid H₂O roughly corresponded to the waveform of the YAG laser pulse. The waveform of the optical emission intensity observed in supercritical H₂O was similar to that observed in liquid H₂O especially in the spike part, as shown in Fig. 1(a) (linear plots). However, we observed that the optical emission intensity observed in supercritical H₂O was stronger in the tail part. The difference between the optical emission intensities in the tail parts of the waveforms is clearly shown in Fig. 1(b) (semilogarithmic plots). We observed exponential decreases in the optical emission intensities in both liquid and supercritical H₂O, but the decay time constant of 29.0 ns observed in supercritical H₂O was much longer than that in liquid H₂O (8.6 ns).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Figure 2 shows the spatial distributions of the optical emission intensities observed in supercritical (430 °C) and liquid (230 °C) H₂O. The water pressure was 30 MPa in both cases. The spatial distributions in the normal direction to the target surface are plotted. These distributions were observed at the maxima in the temporal variations of the optical emission intensities (corresponding to a delay time of 12 ns). As shown in Fig. 2, the plasma with optical emission was transported more efficiently when it was produced in supercritical H₂O. The size (FWHM) of the plasma produced in liquid H₂O was less than 0.1 mm, while that of the plasma produced in supercritical H₂O was greater than 0.2 mm. In addition, in the tail part, the plasma produced in supercritical H₂O reached a distance of up to 1.2 mm from the target at a delay time of ∼10 ns after laser ablation.

Zoom In Zoom Out Reset image size

The laser ablation plasma produced in liquid H₂O is so dense that the lifetime of the optical emission is determined not by the radiative lifetimes of excited states but by collisional quenching. This is the reason for the experimental result that the waveform of the optical emission intensity follows the waveform of the laser pulse in liquid-phase laser ablation. The longer decay time constant of the optical emission intensity observed in supercritical H₂O shown in Fig. 1 indicates less significant collisional quenching. The longer lifetime of the optical emission intensity is related to the easier expansion of the ablation plume in supercritical H₂O. This is confirmed experimentally as shown in Fig. 2. The transport length of 1 mm within a delay time of 10 ns indicates an expansion speed of 10⁸ mm/s, which is much higher than the expansion speed of the plume in liquid-phase laser ablation (on the order of 10⁶ mm/s)^21,38) and is similar to the expansion speed of the plume in gas-phase laser ablation.^39,40) Consequently, it can be concluded from the present experimental results that the confinement of the plume in supercritical H₂O is much weaker than that in liquid H₂O at the same pressure. Note here that the less significant quenching and weaker confinement of the plume in supercritical H₂O may not be closely related to the higher laser energy (the laser fluence was adjusted to be the same by changing the laser energy in this experiment), since the quenching and confinement are phenomena determined by the collision rate, where the density of the collision partner in the medium is the most important quantity. However, the rate coefficients of collisions are affected by particle energies, and more investigations are necessary to determine the effects of laser energy on the quenching and confinement phenomena.

In the laser ablation experiment in supercritical CO₂, we examined the effects of the pressure and temperature on the optical emission intensity in detail. We employed a weak YAG laser fluence of 0.6 J/cm² in this experiment by not using focusing lenses. This fluence was below the threshold for achieving the laser ablation of the gold target in gaseous CO₂. However, we observed the production of a laser ablation plasma when the YAG laser pulse with the same fluence irradiated the gold target in supercritical CO₂.

Figure 3(a) shows the optical emission intensities of laser ablation plasmas observed at various temperatures and pressures of CO₂. The optical emission intensities plotted in Fig. 3(a) are obtained by the spatiotemporal integration of the optical emission images. An optical emission image was obtained by averaging 10 results to compensate for the shot-to-shot fluctuation. The critical point of CO₂ is at a pressure of 7.4 MPa and a temperature of 31.1 °C, and hence all the experiments shown in Fig. 3 were carried out in supercritical CO₂. As shown in Fig. 3(a), the optical emission intensity of the laser ablation plasma produced in supercritical CO₂ was significantly dependent on the pressure and temperature. In addition, as shown in Fig. 3(b), the diameter of the laser ablation plasma (FWHM in the spatial distribution) had a similar dependence on the optical emission intensity. Note here that, as shown in Fig. 3(c), the variations of the optical emission intensity and plasma size were not related to the variation of the laser fluence on the target surface. The transmission of the laser beam in supercritical CO₂ was weakly dependent on the pressure and temperature, as shown in Fig. 3(c), and the variations of the optical emission intensity and plasma size were not related to the transmission. In addition, the distortion of the beam shape, which was observed in supercritical H₂O, was negligible in supercritical CO₂.

Zoom In Zoom Out Reset image size

Figure 4 shows a phase diagram of CO₂. We superposed the contour plot of the optical emission intensity in the corresponding region of the pressure–temperature plane. The experimental data are the same as those shown in Fig. 3(a). The dashed curves show the relationships between the pressure and temperature corresponding to the mass densities of 200, 300, and 400 kg/m³. As shown in Fig. 4, the ridge line of the contour plot is roughly along the dashed curve corresponding to a density of ∼300 kg/m³. Therefore, Fig. 4 suggests that the experimental conditions that result in bright optical emissions of the plasmas are concentrated in the region corresponding to a density of ∼300 kg/m³.

Zoom In Zoom Out Reset image size

The effect of the pressure of CO₂ on the optical emission intensity of a laser ablation plasma at a temperature of 304.5 K³⁵⁾ has been discussed by Kato et al., who observed a bright optical emission intensity around the critical point. They pointed out the importance of the density fluctuation on the optical emission intensity on the basis of their experimental observations. Our experimental results shown in Figs. 3 and 4 partly support the argument of Kato et al., since we observed the optical emission of a laser ablation plasma only in supercritical CO₂, suggesting the importance of the density fluctuation. In the following, we discuss the quantitative correlation between the optical emission intensity and the density fluctuation according to our experimental results. Figure 5 shows the magnitudes of the density fluctuations under the experimental conditions employed in Fig. 3(a). The relative values of the density fluctuations were evaluated from the product of the density, isothermal compressibility, and temperature.⁴¹⁾ By comparing Fig. 5 with Fig. 3(a), it is understood that the variation of the optical emission intensity observed at a pressure of 8 MPa is roughly similar to the variation of the density fluctuation. However, at pressures of 10 and 12 MPa, the correlation between the optical emission intensity and the density fluctuation was rather weak. Therefore, according to our experimental results, which were observed in a regime deviated from the critical point, it is suggested that the density of CO₂ is a more important parameter than the density fluctuation for obtaining a bright optical emission intensity of a laser ablation plasma produced in supercritical CO₂. The paper published by Kato et al.³⁵⁾ suggests the singularity of the critical point where the density fluctuation is significantly large.

Zoom In Zoom Out Reset image size

The bright optical emission can be explained by the less significant collisional quenching of excited states. We observed similar trends in the optical emission intensity and plasma size, as shown in Figs. 3(a) and 3(b), respectively, which can be explained by the easier expansion of the plasma under a condition with less significant collisional quenching. However, we need a quantitative understanding of the relationship among the medium density, density fluctuation, and rate of quenching collisions to explain the experimental results shown in Figs. 3 and 4.

Even if it is possible to attribute the bright optical emission intensity to the less significant collisional quenching, it is still difficult to understand the mechanism of the reduced threshold of the laser fluence for laser ablation. This is because laser ablation is basically a phenomenon of light-matter interaction. The fundamental process of laser ablation is the thermal melting of the target. The thermal melting or the increase in the target temperature is caused by the electron–phonon interaction on the target surface. A possible explanation for the higher threshold fluence for laser ablation in high-pressure gaseous CO₂ is the deexcitation of the electronic excited state on the target surface due to collision with CO₂ molecules. However, the decay time constant of the electronic excited state on the gold surface due to the electron–phonon interaction is ∼1 ps,^42,43) while the frequency of collision between CO₂ molecules and Au atoms on the gold surface is expected to be 2 × 10¹⁰ s⁻¹ at a pressure of 10 MPa. Hence, we could not expect the collisional deexcitation of the electronic excited state under our experimental conditions. Another possible explanation is related to the particular thermodynamic parameters of supercritical CO₂. The threshold laser fluence becomes lower if the target is irradiated in a medium with a smaller thermal conductivity. However, it is known that the thermal conductivity of supercritical CO₂ is greater than that of gaseous CO₂, especially at a density corresponding to significant fluctuation.⁴⁴⁾ Note that the reduced threshold fluence for laser ablation has also been observed by Kato et al. in supercritical CO₂.³⁵⁾ Further investigation is necessary to understand the fundamentals of laser ablation phenomena in supercritical fluids.

Figure 6 shows shadowgraph images observed by laser ablation in supercritical H₂O at a pressure of 30 MPa and a temperature of 430 °C. The shock front was captured as shown in Figs. 6(a) and 6(b) at delay times of 0.9 and 4 µs after laser ablation, respectively. The propagation speed of the shock front was approximately 530 m/s. The region inside the shock front was dark at a delay time of 0.9 µs, as shown in Fig. 6(a). Another small dark region was formed around the ablation point at a delay time of 4 µs, as shown in Fig. 6(b). The latter dark region expanded with time, as shown in Figs. 6(c) and 6(d). The maximum size of the dark region was observed at a delay time of 14 µs. The shape of the dark region was roughly hemispherical up to the maximum size. We observed the distortion of the edge of the dark region after the maximum size, as shown in Fig. 6(e). At a long delay time of 70 µs, the dark region was extinguished into the surrounding supercritical H₂O, as shown in Fig. 6(f).

Zoom In Zoom Out Reset image size

The dark region shown in Fig. 6(a), which was observed inside the shock front at a delay time of 0.9 µs, may be attributed to the high densities of species ejected from the target. The size of the dark region was approximately 0.4 mm, which was similar to the size of the main body of the plasma with optical emission shown in Fig. 2. The shadowgraph image shown in Fig. 6(a) indicates the rapid expansion of the laser ablation plume into supercritical H₂O.

The dark region in the shadowgraph images shown in Figs. 6(b)–6(d) indicates that the refractive index of this region is different from that of the surrounding supercritical H₂O. The different refractive index may be caused by the rarefied medium density. The shadowgraph images of the dark region are similar to those of a cavitation bubble observed in liquid-phase laser ablation. The cavitation bubble in liquid-phase laser ablation is induced by the rapid vaporization of the liquid medium. The inside of the cavitation bubble is in the gas phase at a pressure on the order of 1 kPa.⁴⁵⁾ In this paper, we call the dark region shown in Figs. 6(b)–6(d) "a bubblelike hollow", since we are not sure whether the inside of the dark region is in the gas phase at a reduced pressure.

A markedly different feature of the dynamics of the bubblelike hollow from the cavitation bubble was that the bubblelike hollow never collapsed. It is well known that a cavitation bubble induced in liquid-phase laser ablation shrinks after it reaches its maximum size and collapses at a certain delay time after laser ablation.^17,46,47) In contrast, the bubblelike hollow shown in Fig. 6 never shrank or collapsed. Another difference of the bubblelike hollow from the cavitation bubble was in its maximum size. Figure 7 shows a comparison of the maximum sizes of the bubblelike hollow and a cavitation bubble induced by laser ablation in liquid H₂O at the same pressure of 30 MPa and a temperature of 20 °C. The maximum size of the cavitation bubble was 0.06 mm, which was observed at a delay time of 0.6 µs. After that, we observed the shrinkage and collapse of the cavitation bubble. On the other hand, the maximum size of the bubblelike hollow observed in supercritical H₂O at the same pressure was 0.25 mm. The delay time corresponding to the maximum size was 14 µs. This indicates much easier expansion of the bubblelike hollow in supercritical H₂O than the cavitation bubble in liquid H₂O at the same pressure.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

A comparison between the bubblelike hollows observed in supercritical and gaseous CO₂ is shown in Fig. 8 . The laser fluence in this experiment was 10 J/cm², achieved by focusing the YAG laser beam using a lens. We observed the formation of a bubblelike hollow by shadowgraph imaging even in gaseous CO₂ at a pressure of 5 MPa and a temperature of 40 °C, as shown in Figs. 8(a)–8(c). The shadowgraph images of bubblelike hollows observed in supercritical CO₂ (at a pressure of 8 MPa and a temperature of 40 °C) were similar to those observed in gaseous CO₂, as shown in Figs. 8(d)–8(f), except for the smaller size. A smaller bubblelike hollow was observed in supercritical CO₂ at a higher pressure. Kato et al. observed similar shadowgraph images in supercritical CO₂ at different temperatures and pressures.³⁵⁾ The experimental results shown in Fig. 8 suggest that the dynamics of supercritical fluids as the medium of laser ablation are rather similar to the dynamics of ambient gas in gas-phase laser ablation. Sasaki and Watarai investigated the dynamics of ambient gas in gas-phase laser ablation by laser-induced fluorescence.⁴⁸⁾ They clearly observed the formation of a region with a rarefied ambient gas density around the ablation point. The dark regions observed by shadowgraph imaging in supercritical fluids may be similar to the region with a rarefied ambient gas density in gas-phase laser ablation, and are different from the cavitation bubble in liquid-phase laser ablation. However, the experimental results shown in Figs. 6 and 8 were observed in "gas-like" supercritical fluids. The comparison with bubblelike phenomena in "liquid-like" supercritical fluids is an interesting future work.

Zoom In Zoom Out Reset image size