12 10 11 10 10 109 ΔΗ 108 N = Noe Foc 107 0.02 Fluence (mJ/cm2) Figure 8. Change of the absolute neutral Zn yield as a function of the inverse of fluence. The solid line is a least-squares fit of an Arrhenius expression to the three data points obtained at low ablation fluences. The dashed line is a least-squares fit to the power law using all data. In figure 7, the increase of the absolute neutral Zn yield is depicted as a function of the inverse temperature. The Zn yield follows an Arrhenius relation for fluences between 20 and 60 mJ/cm2 with an activation energy of 2.5 eV. This is approximately the heat of vaporization of ZnS which indicates that a thermal process, such as sublimation, may account for the emission [14]. At higher fluences, the measured temperatures deviate significantly from the straight line. This deviation also occurs as an abrupt step, similar to figure 6, indicating that a second process has started. Figure 8 shows the absolute neutral Zn yield as a function of the inverse fluence. Since for thermal evaporation at low fluences 1/F 1/T, a straight line should be expected. For fluences higher than 60 mJ/cm2, the yield shown in figure 8 deviates from the straight line. This change accompanies the jump in temperature in figure 7. This behavior suggests that the plasma interacts with the surface leading to a rise of the yield. Since we are only detecting neutral Zn particles once a plasma is formed, the total particle density may be significantly higher than the measured value. The relationship, N∞ F。6, appears to fit the data very weli. Such power laws are often used to explain multiphoton processes [16-18]. In this case, however, the ablation laser wavelength was in the region of interband absorption; thus, there is no obvious physical meaning to this power law dependence in our experiment. Figure 9 shows the Zn yield, the kinetic temperature derived from the velocity distribution, and the first occurrence of visible damage for a multipulsed experiment on a single spot at a fixed fluence of 45 mJ/cm2. It can be seen that after 30,000 shots the particle density increases nearly exponentially with consecutive laser shots. The kinetic temperature was about 1900 K and did not change very much during the entire experiment. After 150,000 shots (corresponding approximately to a removal of 60 monolayers), the first visible damage occurred. The observed increase in particle emission with consecutive laser shots could be attributed to a change in the surface morphology, which in turn could lead to better coupling between the laser light and the surface as well as to changes in the surface stoichiometry. Clearly, figure 9 shows unambiguously that microscopic damage increases with the number of laser shots and, consequently, leads to catastrophic failure at lower laser fluence than the single-pulse damage threshold. Figure 9. Plot of the kinetic temperature and particle density as a function of the number of laser shots striking one site. The results reported in this paper can be rationalized on the basis of the following model. The high kinetic temperatures obtained for low laser ablation fluences are due to reduced thermal contact of the surface region with the bulk. This can result from cracks in the crystal produced by repetitive laser pulses heating the surface, thus causing localized stresses due to differential thermal expansion [19]. This effect may also explain the large differences in kinetic temperatures of the ejected atoms obtained in single-pulse [1] compared to multiple-pulse experiments. The sudden rise in temperature at a particular fluence indicates that a critical combination of particle density and laser intensity leads to formation of a plasma. The plasma thus formed interacts with the surface and leads to catastrophic failure. The very small size of the observed damage spots (um) in comparison to the irradiated area (mm) suggests the possibility that local breakdown between the plasma and the surface may be leading to damage. In both single-pulse and multiple-pulse experiments, high-energetic neutral particles were observed at fluences near the visible damage threshold. Combined LFS TOF measurements indicate that these neutrals are products from molecule or cluster fragmentations caused by the excitation laser. Presumably, electron impact ionization could also fragment such species. The emission of these fast molecules or clusters are believed to be the result of cleaving or cracking of the surface. Clearly, additional measurements are necessary to verify this damage model and to gain more detailed information on the mechanism for atomic and cluster emission during laser irradiation. 4. Conclusions It can be concluded that the LFS TOF technique is a very sensitive and effective method for detecting ejected neutral particles from laser-irradiated surfaces for a variety of elements and transitions as well as their characterization by yields and velocity distributions. Neutral particle emission was identified as a precursor of observable damage. The unique combination of Dopplershifted LFS and TOF techniques is a useful tool for separating prompt from delayed emission of particles ejected from laser-irradiated surfaces as well as for probing possible molecular or cluster fragmentation. To get more detailed information about the physical processes involved in particle ejection and optical damage, measuring of the spatial dependence of the emission and quantitative relative yields of various surface constituents have to be included in future investigations. [1] Chase, L.L.; Smith L.K. Laser Induced Surface Emission of Neutral Species and Its Relationship to Optical Surface Damage Processes, in the Proceedings of Laser Induced Damage in Optical Materials (NIST Special Publication 756, (1987), in press. [2] Chase, L.L.; Lee, H.W.H. Accumulated Surface Damage on ZnS Crystals Produced by Closely Spaced Pairs of Picosecond Pulses, to be published in the Proceedings of the 20th Symposium on Optical Materials for High Power Lasers, (1988). [3] Arlinghaus, H. F.; Calaway, W. F.; Young, C. E.; Pellin, M. J.; Gruen, D. M.; Chase, L. L. Laser Fluorescence Spectroscopy of Zinc Neutrals Originating from Laser-Irradiated and lonBombarded Zinc Sulfide and Zinc Surfaces, in the Proceedings of Laser Induced Damage in Optical Materials (NIST Special Publication 756, (1987), in press. [4] Arlinghaus, H. F.; Calaway, W. F.; Young, C. E.; Pellin, M. J.; Gruen, D. M.; Chase, L. L. HighResolution Multiphoton Laser-Induced Fluorescence Spectroscopy of Zinc Atoms Ejected from Laser-Irradiated ZnS Crystals, J. Appl. Phys 15 (1988). [5] Arlinghaus, H. F.; Calaway, W. F.; Young, C. E.; Pellin, M. J.; Gruen, D. M.; Chase, L. L. Analysis of lon-Bombarded and Laser-Irradiated ZnS and Zn Surfaces via High-Resolution Multiphoton Laser-Induced Fluorescence Spectroscopy, J. Vac. Sci. Technol. May/June A (1989), in press. [6] Arlinghaus, H. F.; Calaway, W. F.; Young, C. E.; Pellin, M. J.; Gruen, D. M.; Chase, L. L. Laser Damage Studies of ZnS via Neutral Zn Particle Emission, in the Proceedings of the 4th International Laser Science Conference (1988), in press. [7] See, for example, Lee, C. S.; Koumvakalis, N; Bass, M. A. Theoretical Model for Multiple-Pulse Laser-Induced Damage to Metal Mirrors, J. Appl. Phys. 54, 5727 (1983). [8] See, for example, Jee, Y.; Becker, M. F.; Walser, R. M. Laser-Induced Damage on Single-Crystal Metal Surfaces, J. Opt. Soc. Am. B5, 648 (1988). [9] Arlinghaus, H. F.; Calaway, W. F.; Young, C. E.; Pellin, M. J.; Gruen, D. M.; Chase, L. L. IsotopeShifts of Zn Neutral Atoms Measured by Two-Photon Doppler-Free Laser-Induced Fluorescence Spectroscopy, submitted to Opt. Lett. [10] Wood, R. M. Laser damage in optical materials. Bristol: Adam Hilger; 1986. [11] De Maria, G; Goldfinger, P; Malaspina, L.; Piacente, V. Mass-Spectrometric Study of Gaseous Molecules, Trans. Faraday Soc. 61, 2146 (1965). [12] W. H. Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T. in Numerical Recipes: The Art of Scientific Computing, Cambridge University Press, Cambridge (1986). Touloukian, Y. S.; DeWitt, D. P. Thermophysical Properties of Matter, IFI/Plenum, New York (1972), Vol 8., p. 1217. [13] Bechtel, J. H. Heating of Solid Targets with Laser Pulses, J. Appl. Phys. 46, 1585 (1975). [14] Landolt Börnstein, N.S., Vol. 17b, Springer-Verlag, Berlin (1982), pp. 110-114. [15] [16] [17] [18] Bräunlich, P.; Schmid, A.; Kelly, P. Contributions of Multiphoton Absorption to Laser-Induced Matthias, E.; Nielsen, H. B.; Reif, J. Multiphoton-Induced Desorption of Positive lons from Petzoldt, S.; Elg, A.P.; Reif, J.; Matthias, E. Surface Damage Thresholds of Fluoride Crystals: Dependence on Surface Quality and Irradiation Mode, to be published in the Proceedings of the 20th Symposium on Optical Materials for High Power Lasers, (1988). [19] Krutyakova V.P.; Smirnov, V. N. Electron Emission from Alkali Halide Crystals Irradiated by a Pulsed CO2 Laser, Sov. Phys. Tech. Phys. 24, 1085 (1979). Comments Question: And what is the initial process responsible for damage? Answer: We could talk a whole day about it, and there's a lot of Russian literature - Manuscript Received 1. Substrate Cleaning in Vacuum by Laser Irradiation Tilak Raj, D. E. McCready and C. K. Carniglia Martin Marietta Astronautics Group, Laser Systems Technology The cleaning of a substrate prior to the deposition of a coating on its surface is an important factor influencing the quality of the coating. In most cases, the cleaning of the surface takes place before the substrate is placed in the coating chamber. This paper presents the results of an investigation of the effects of the irradiation of the surfaces of fused silica substrates in a UHV chamber using a CW CO2 laser. The laser beam was rastered over the substrate surface, and desorbed gases were detected using a residual gas analyzer. Water was desorbed from the surface upon irradiation with a low-power beam, while hydrocarbons required higher laser power for removal. determined that a single irradiation of a substrate was effective in removing organic contamination and water from its surface. The laser power required to remove the majority of the contamination was lower than the power which resulted in damage to the surface as observed using an interferometer and a polariscope. With the part maintained in a UHV environment, the surface remained uncontaminated for several hours. A series of surfaces was cleaned with various solvents. Residual gas analysis indicated that the solvents had not been removed completely during a high-temperature vacuum bake of the system. Laser cleaning of optical surfaces immediately prior to coating deposition is expected to improve adhesion and to affect the nucleation, growth and other characteristics of the deposited coating. Key Words: CW CO2 laser; fused silica; substrate; laser irradiation scan; mass spectrometer; residual gas analysis; substrate cleaning Introduction The cleaning of a substrate before the deposition of a coating on its surface is an important factor influencing the quality of the coating. In most cases, the cleaning of the surface is performed before the substrate is placed in the coating chamber. However, cleaning of a substrate in vacuum before coating deposition would be more desirable, but is not always practical. Some of the techniques which could be used for precleaning in vacuum are glow discharge, sputter ion etching and laser irradiation. Since a glow discharge cleaning [1] is not confined to the substrates, particulates are removed from substrates, fixtures and surrounding chamber walls. The resulting redeposition of the contaminants onto the substrates is a serious drawback of this technique. Sputter ion etching with a noble gas like argon has been employed successfully for precleaning elemental substrates. But, in the case of a dielectric substrate, preferential sputtering [2] and damage to the substrate surface occur. This paper describes a successful application of CO2 laser irradiation in removing organics and water from the surface of fused silica substrates placed in an ultrahigh vacuum (UHV) chamber. It was determined that a single irradiation with a scanned high-power laser beam was most effective in removing the volatile contaminants. The highest laser power which can be used is limited by the surface damage. A redeposition of contaminants on the substrates in the UHV chamber did not occur for several hours. It should be noted that laser irradiation of optical components has been used in other contexts. For example, laser irradiation was suggested to remove ice from the surface of metal mirrors used at cryogenic temperatures [3]. As another example, a study of C and H20 removal from optical coatings by laser irradiation was reported recently [4]. 2. Experimental An x-y scanned CO2 laser beam was directed at the substrate placed in a UHV chamber. gaseous components released from the substrate surface were detected by a mass spectrometer. experimental details are described below. The The |