with water. Dichloromethane is a chlorohydrocarbon and is immiscible with water. N-butyl acetate is a long-chain aliphatic solvent having good miscibility with water. A set of four substrates (numbers 57-60) were cleaned with the standard cleaning procedure followed by a drag wipe with one of these organic solvents. The first laser irradiation scan produced a very high chamber pressure (see part numbers 57 through 60 in table 2) and all three groups of peaks with large amplitudes. For example, figure 12 shows the RGA spectrum for the first scan from part number 59 wiped with dichloromethane. Three groups of peaks were present: Note the very large amplitude of peaks 18 (H2O), 28 (CO, CH2 CH2, N2), and 44 (CO2, CH3-CH2-CH3). The residue, which was most Tikely left by the impure solvent, was desorbed by the first irradiation scan. Figure 13 shows the RGA spectrum for the first scan on part number 60 wiped with n-butyl acetate. Once again, three group of peaks were detected, but with a different peak structure than obtained from dichloromethane (fig. 12). Similar observations were noted for the first irradiation of parts wiped with xylene or MEK. The second and subsequent scans produced small water peaks (18, 17) and very small peaks like 15, 28, etc. This implies that one irradiation of sufficient laser power was able to remove most of the volatile organic residue and water from fused silica substrates, regardless of the nature of the organic contamination. It should be noted that the cracking patterns caused by the solvents employed in the laser irradiation experiments were totally different than those obtained by conventional gas analysis with a RGA spectrometer. In the irradiation experiments, small hydrocarbons or water components were detected, regardless of the complexity of the organic molecule present at the substrate surface. For example, the largest peak obtained from an RGA spectrum of a substrate with MEK should be 43, but laser irradiation produced 16 and 15 as the predominant peaks. This might be due to fragmentation of the surface molecules caused by the high surface temperature (>1000°C) created by repetitive laser irradiation. It should be noted that the highest mass detected in this study was 45, independent of the laser power and the organic solvent used in the substrate cleaning. 25 MASS 4. Summary Cleaning of substrates outside the vacuum chamber was not a complete process. Only an in-situ cleaning process in vacuum could remove the residual surface contaminants. A single irradiation scan of sufficient laser power can remove the volatile organic residue and water vapor from the substrate surface, independent of the nature of the solvent present on the substrate. The water and other peaks detected after repetitive laser irradiation scans originated from the areas adjacent to the directly irradiated area and not from the irradiated area itself. There was very little redeposition of organic contamination on the substrate 12-16 hours after the laser irradiation. Only small hydrocarbons or water components were detected regardless of the complexity of the organic molecule at the substrate surface. This might be due to the molecules at the surface breaking due to the high surface temperature caused by the laser irradiation. The rise in chamber pressure during irradiation corresponded to the rise in amplitude of peaks detected by the mass spectrometer. Thus, the effect of laser irradiation could be monitored by noting the total chamber pressure. As soon as the maximum chamber pressure from one irradiation no longer differs from that of the previous irradiation, a complete degassing has taken place. From a practical application, laser degassing or cleaning can be monitored in such a manner by a simple pressure gauge rather than with a mass spectrometer. The authors wish to acknowledge the help of Jerry Kienle and Art Westerfeld for fabricating the UHV fixtures required to perform the experiments. This effort was sponsored by the Air Force Weapons Laboratory, Air Force Systems Command, United States Air Force, Kirtland AFB, New Mexico 87117 and was funded by the Air Force Office of Scientific Research. 5. References [1] Dylla, H.F. Glow discharge techniques for conditioning high-vacuum systems. J. Vac. Sci Technol. A6 (3): 1276-1278; 1988. [2] Holloway, P.H. and Nelson, G.C. Preferential sputtering of Ta205 by argon ions. J. Vac. Sci Technol. 16: 793; 1979. [3] Piper, L.G.; Spencer, M.N.; Wodward, Wodward, A.M.; Green, B.D. Optical System Contamination: Effects, Measurements, Control. SPIE Proc. 777; 1987. 320-332. [4] Domann, F.E. and Stewart, A.F. Laser induced particle emission as a precursor to laser damage. Presented at 1987 meeting of the Boulder Damage Symposium, to be published in Nat. Inst. Stand. Tech. (U.S.) Spec. Publ. COMMENTS Did you ever see any evidence of coating material in the RGA or in the substrate material? Question: Answer: I don't think we had any solute remover. Question: Answer: Question: Answer: Question: Answer: Question: Answer: Question: Answer: Question: Answer: Question: Answer: Question: Answer: Have you tried to correlate your results with the study that Allen and Porteus did 4 or 5 years ago on laser desorption of water? Our goal was slightly different so we really did not try to do that. That was partially my question. Are you aware of any work on the correlation No, I have my personal opinion on low discharge. It is a dirtying process, not Did you try dielectrics, metals, different things? No, we tried a few silica. We believe there should be two or more dielectrics investigated. And the second question is did you ever observe onset of damage after repetitive shocks, did you look for it and was there a damage onset after a number of shots? No, no, we have here a pretty low power and we did not see any damage. Did you see any indication at all, either through your mass spectrographic analysis or even just a visual inspection, whether there was ablation of the fused silica? I don't think the powers were high enough to do that and you know from the mask data that we did not see any. I think if you try and calculate what that temperature would be at the surface and under those conditions, I think you can reach a temperature high enough to begin vaporizing the surface. Yes, that is true. I was wondering if you have any reason to expect that laser radiation is better than baking the substrates. This is a surface phenomena and temperatures are high. Baking will not do. Manuscript Received Cavity Ringdown Measurements of High-Reflection Mirrors at 1.06 μm L. John Jolin, Virgil R. Sanders, Thomas P. Turner Los Alamos National Laboratory Los Alamos, NM 87545 A 1064-nm-cavity-ringdown reflectometer was used to study reflectances of a variety of high reflectance dielectric coatings and silver mirrors. Reflectances of the dielectric coatings were measured at near-normal angle of incidence. The multilayer dielectric high reflectors included ZrO2, HfO2, SiN4, TiO2, and Al2O3, which were used in combination with SiO2 in conventional quarter-wave designs. Reflectors from more than a dozen vendors were surveyed in this effort. Metallic reflectors investigated were silver and silver protected with a thin (10 Å) alumina overcoat. These reflectors were tested at both normal incidence and at high (grazing) angles. The results of these studies show that, although high reflectances (R > 0.9990) can be achieved for 1064-nm-multilayer dielectrics, the majority are not of this caliber. Reflectances for dielectric designs ranged from 0.96 to >0.9990. The silver mirrors exhibited predictable behavior with reflectances of 0.992 at near-normal angle. Key words: multilayer dielectrics; Nd:YAG; high reflector; reflectance; reflectometer; Reflectance is a key parameter in understanding fundamental relationships between mirror characteristics and laser damage susceptibility. An accurate, high precision reflectance measurement capable of measuring reflectances that are near unity is essential. A cavity-ringdown reflectometer utilizing a 1.06-um laser source has been developed for this purpose. Adapted for our use is the ringdown device thoroughly described by Anderson et al. [1] in 1984. The source for this reflectometer is a repetitively pulsed Nd:YAG laser operating at the funda– mental wavelength of 1.06 μm. The repetition rate is variable but is set at 3,000 pps for these measurements to allow for long decay times. The laser temporal profile is near Gaussian with a time duration of 150 ns (FWHM). The average power is approximately 7 W (2.3 mJ per pulse). Dielectric-thin-film polarizers are used, external to the laser cavity, to produce highly planar polarized light. At the test specimen the ratio of S-polarized light (E vector perpendicular to incident plane) to P-polarized light (E vector parallel to the incident plane) is greater than 300 to 1. The reflectometer cavity is a folded configuration consisting of three mirrors - two end mirrors and the test sample (figure 1). The end mirrors are multilayer-dielectric coatings deposited on the concave side of a plano/concave fused silica substrate with a radius of 4.5 m. The test sample is flat. The cavity is folded about the test specimen at an angle of 2 degrees. Both end mirrors are equidistant from the test specimen creating a round-trip cavity length of 11.47 m. To accommodate insertion of the laser beam into the ringdown cavity the entrance end mirror has slightly lower reflectance than the second end mirror. Because the entrance mirror is a high reflector, a little less than 1% of the available laser light is actually coupled into the cavity. The remainder is reflected, dispersed and absorbed in a beam dump. In order to minimize atmospheric effects the entire reflectometer is encased in a Plexiglas housing. Figure 2 is a photograph of the ringdown device and laser source. The cavity-intensity decay (ringdown) is monitored by measuring the leakage through the second end mirror. A commercially available InGaAs detector having a 2-mm diameter is used. A short focal length lens, in close proximity to the exit mirror, focuses the light to a small spot size on the detector. This conveniently corrects for minor cavity misalignments by resteering the misdirected light back to the detector. Accurate alignment is quite easily accomplished because misalignments are recognized by simple sinusoidal oscillations superimposed on the decaying signal as the light scans on and off the detector during subsequent bounces. A low-power beam expander is used between the laser and the ringdown cavity to increase the laser-beam spot size on the test sample. The spot size on the test sample is approximately 3-mm diameter for near-normal angle testing. Non-normal angle testing is accomplished by unfolding the cavity and increasing the size of the test specimen to 4-in. diameter to accommodate the elongated spot. As mentioned previously, the laser light at the test sample is S polarized. A shorter cavity is used for these measurements but is accounted for in the calculations. Figure 3 shows the digital-storage oscilloscope display of a cavity-intensity decay. The quality of the ringdown and the decay constant is determined manually by using the inherent capabilities of the oscilloscope. As described by Anderson, et al., the reflectance product of a ringdown cavity is, 1 Rp=| 1 (1) where tc is the cavity-intensity-decay time constant, L is the cavity round-trip optical path length, and C is the speed of light. For the folded ringdown cavity represented schematically in figure 1 the reflectance product is, |