Laser ablation of overlayers in conservation a numerical model

Document Sample
Laser ablation of overlayers in conservation  a numerical model Powered By Docstoc
					Meeting user needs – small and medium enterprises and research

Laser ablation of overlayers in conservation – a numerical model
Andrzej KOSS and Jan MARCZAK*

Summary
The main aim of this paper is to present a hydrodynamic model of the material ablation process during high-power pulse laser irradiation. Numerical calculations of pressure, temperature, and density as well as the ablation rate of graphite (a model material for encrustation) on aluminium are presented. General conclusions concerning the laser removal of encrustation are given, and experimental results of soot removal from a glass plate are compared with the results of the calculations.

– The absorption coefficient of graphite α ≈ 5×105 cm–1 for λ = 1.06µm is the largest among the absorption coefficients of the usually occurring deposits (e.g. for soot, it ranges from α ≈ 0.9×105 cm–1 to α ≈ 2×105 cm–1). – The specific density of graphite, ρ = 1.782 g/cm3 is approximately equal to the densities of many deposits and is lower than that of aluminium, ρ = 2.7 g/cm3. Generally, the specific density of a deposit is lower than the specific density of a substrate. – Graphite on an aluminium substrate can serve as a model for the compound system of black silver sulphide on silver, occurring, for example, on gilded silver threads.

Introduction
Laser cleaning is a new technique that enables the removal of impurities from a surface using pulsed laser irradiation. This technique, due to its selectivity, precision, controllability and environmental advantages, has recently attracted increasing interest and found wide applications ranging from industry [BAUERLE, 1986; MILLER, 1994; ALLMEN, 1995] to the conservation of works of art and historic buildings [ASMUS, 1973; KAUTEK, 1997; SIANO, 1997]. In the field of conservation, the laser technique is most frequently utilised to remove secondary encrustations. An important limitation that the original material of the substrate cannot be damaged, must be always observed. However, the processes of removing contaminants are in many cases very complicated and not fully recognised. Laser cleaning also involves several non-linear processes that generate shock and elastic-acoustic waves, such as dielectric breakdown, material evaporation and ablation, thermo-elasticity, and others. This paper briefly describes a range of laser-target interactions dependent on a radiation power density value. It presents a hydrodynamic model and selected results of numerical calculations obtained for varying values of density of the laser radiation irradiating a graphite sample on an aluminium substrate. Graphite on aluminium is, to some approximation, ‘a deposit on a substrate’. This combination of materials was chosen for the following reasons: – Graphite can model numerous black and grey deposits occurring on stones, such as sandstone or limestone, as a standard material for the determination of threshold values of the radiation power density.

Interaction of laser radiation with matter
The interaction of laser radiation with matter causes certain changes on the surface of the material and in its bulk. This has been demonstrated for a lot of materials and has led to new scientific and industrial domains, one of which is laser material processing [GREEN, 2003]. All investigations and experiments show that a number of common processes occur within encrustations and original materials irradiated with short, intense laser pulses. The surfaces of the materials undergo ablation only if the irradiation intensity q exceeds the irradiation threshold qthr. Hence, as revealed by the experimental observations, the process of laser radiation interaction with matter can be classified in dependence on the material irradiation intensity q, namely, for q values below the threshold q < qthr, near the threshold q ≈ qthr or above the ablation threshold q > qthr.

The threshold and above the threshold interaction – a numerical model
For a quantitative description of the laser-target interaction for laser pulses with a duration that falls within the range of one to several hundred nanoseconds, a numerical model describing the occurring phenomena has been developed. This model includes the effects of absorption of laser radiation, thermal conduction, radiation transport and ionisation of the material. Figure 1 shows a geometrical diagram illustrating the developed model of interaction of the pulse Q-switched Nd:YAG laser radiation with the object. It consists of a graphite encrustation on an aluminium substrate. Such geometry can be

249

Workshop 3
air graphite aluminium air
8.0 T Temperature [kK], Pressure [Gbar] Density [g/cm3] t = 50 ns 6.0 4.0 2.0 p 0.0 -240 -2.0 -4.0 Air Distance [mm] rAL rC

laser power density Q Z

3 mm Z=0

42 mm

-200

-160

-120

-80

-40

0

40

Target

Figure 1. Geometry of irradiated area used in the numerical calculations for a model graphite-on-aluminium compound system.

Thickness of ablated layer [µm]

applied to each particular case in conservation practice during the removal of secondary contaminants. Fundamental, physical equations for mass, momentum and energy conservation used for numerical analysis are given in [WILKINS, 1984].

Figure 2a. Temperature (T), pressure (p) and density (ρ) of the medium at the ablated surface irradiated with a laser pulse of 10ns, after an interval of 50ns. The laser power density was 500 MW/cm2.

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 10 20 30 Time [ns] 40 50

Results of numerical calculations
Selected results of the numerical calculations modelling the effect of irradiating the aluminium-graphite system with an intense laser beam of a power density of Q0 = 5 × 105 ÷ 5 × 109 [W/cm2] are illustrated below. The triangular shape of the laser pulse was assumed and the base of the triangle (pulse width) was 10ns. Figure 2(a) illustrates the distribution of the plasma parameters (temperature, pressure and density) at the ablated surface irradiated with a laser 10ns pulse, after an interval of 50ns. The negative values of pressure in the target area indicate the expansion of the target material and a resulting extraction of molecules from the surface layer. Figure 2(b) depicts the thickness of the evaporated layer as a function of the time that elapsed from the irradiation of the target with a laser pulse. It can be seen that, even after the laser pulse of 10ns has ended, the ablation of the target material continues to a small extent. This is caused by the interaction of the surface layer with hot, long-life plasma generated during the laser pulse. Figure 3 shows the rate of the graphite ablation process as a function of the radiation intensity. The thickness of the evaporated graphite layer strongly depends on the density of the laser beam power below 100 MW/cm2. One can see that the ablation process is a threshold phenomenon and that the maximal thickness of the ablated layer is around 0.9µm for a radiation intensity of about 500 MW/cm3. A further increase in the laser radiation intensity between 100 MW/cm2 and 10 GW/cm2 does not accelerate the process and even the layer of ablated graphite diminishes. This result can be explained by the ionisation of the air. Plasma formed of ablated graphite strongly heats the ad-

Figure 2b. Calculated thickness of the ablated graphite layer versus time interval after the irradiation with a laser pulse of 10ns. The laser beam power density was 500 MW/cm2.

1.0 Thickness of ablated layer [µm] 0.8 0.6 0.4 0.2 0 0.01

0.10

1

10

100

1000

10000

Power density [MW/cm2]

Figure 3. Influence of the peak power density of the laser pulse (5ns) on the thickness of the ablated layer – solid line: results of the numerical calculations for the graphite-on-aluminium model, single point data – experimental results for the ablation of a soot layer on a glass substrate.

jacent air, which causes air ionisation. At low temperatures air is almost totally transparent for laser radiation. The incidental radiation is, however, absorbed by ionised air and its transport to the target is fully blocked.

250

Meeting user needs – small and medium enterprises and research

Additionally, the experimental results of the ablation of a soot layer on a glass substrate, obtained from a flame of a stearin candle, are presented in Figure 3. The thickness of the soot layer removed from the glass substrate is the arithmetic average of several measurements made for the same irradiation level. Several series of calculations with various pulse durations and various energy densities were also performed. The calculation results are given in Figure 4. Changing the laser pulse duration enables the determination of optimal values of laser power density, which is also depicted in Figure 3. The reason for this is the dependence of the ablated layer’s thickness not only on the power itself, but also on the absorption of laser radiation in the generated plasma layer, as well as other, rather complex processes. The attenuation of the laser beam depends, among other things, on plasma temperature and the thickness of the absorbing plasma layer, which are higher for higher laser power densities.

– Real ablation is observed only above the threshold value of radiation power density qthr. – Below the threshold value q < qthr, which is characteristic of each material, one can observe changes in surface morphology, the generation of defects, and the selective removal of different components of material. – The ablation rate ‘d’ (µm/pulse) is defined as the thickness of a layer removed during one laser pulse. It is a characteristic value for a particular material and is related to the radiation power density. – Ablation rate of every deposit has its maximum ncri depending on the material and the wavelength of the laser. A further increase in the radiation power density causes a reduction of the thickness of the evaporated layer. This is a result of the absorption of the incidental radiation in a plasma layer of a critical density. – The value of the threshold radiation power density qthr decreases together with an increase in the medium absorption coefficient. – The area of damaged material close to the ablated region diminishes with the increase in absorption and a shortening of the pulse duration. Application of the laser radiation technique for the renovation of works of art and architectural decorations is relatively new. The technique is very precise. Its application is limited because of the very high sensitivity of both the substrate and the paint layer with its binding medium to high intensity laser light. Despite the complexity of the laser technology and the sensitivity of the treated medium, this technology is the most versatile technique developed to date for surface cleaning when compared with conventional methods. This is confirmed by numerous results obtained from the work of researchers from various scientific centres [ZAFIROPULOS, 1998; KLEIN, 2000] as well as our own work [MARCZAK, 1997; KOSS, 1999; MARCZAK 1999]. In these projects, secondary encrustations and paint layers on works of art and historic architectural objects were successfully removed using the laser ablation technique.

Conclusions
The interaction of laser radiation with the surface of a solid is a complex process. Many processes resulting from the absorption of the laser radiation take place. These phenomena depend mainly on the chemical and physical properties of the surface, as well as on the laser beam parameters. It can be stated, on the basis of numerous experimental studies [ASMUS, 1973; COOPER, 1992; SIANO 1997], including our own work [MARCZAK, 2001] on the laser-induced ablation of various materials, that the theoretical model presented here confirms the conclusions drawn from experiments and can be utilised for the qualitative prediction of object encrustation/ substrate behaviour during laser renovation. Moreover, these conclusions are common for various media:
2.0

Thickness of ablated layer [µm]

50ns

1.5 25ns 1.0 15ns 0.5 10ns 5ns

References
ALLMEN M. (Editor), Laser-Beam Interactions with Materials, Physical Principles and Applications, Springer-Verlag, Berlin and Heidelberg, 1995. ASMUS J.F., MURPHY C.G. and MUNK W.H., ‘Studies on the Interaction of Laser Radiation with Art Artifacts’, Proc. SPIE, 41, 19–27, 1973. BAUERLE D., Chemical Processing with Lasers, Springer Series in Materials Science, Vol.1, Springer-Verlag, Berlin Heidelberg, 1986. COOPER M.I., EMMONY D.C. and LARSON J.H., ‘The Use of Laser Energy to Clean Polluted Stone Sculpture’, J. Photographic Science 40, 55, 1992.

0 0.001

0.01

0.1

1

10

100

1000

Energy density [J/cm2]

Figure 4. Calculated thickness of the graphite layer evaporated from the aluminium substrate on irradiation with laser pulses of variable energy density and duration – numbers denote half-width. The triangular shape of the laser pulse and a homogeneous graphite layer were assumed.

251

Workshop 3
GREEN M. (Editor), The Industrial Laser User, ISSN 1366963X, Association of Industrial Laser Users, www.ailu.org.uk. KAUTEK W. and KOENIG E., (Editors), LACONA I, Laser in the Conservation of Artworks, 4-6 October 1995, Heraklion, Crete, Greece; Proc. LACONA I, Verlag Mayer & Comp., Vienna, 1997. KLEIN S. STRATOUDAKI T., MARAKIS Y., ZAFIROPULOS V. and DICKMAN K., ‘Comparative Study of Different Wavelengths from IR to UV Applied to Clean Sandstone’, Applied Surface Science, 157, 1–6, 2000. KOSS A. and MARCZAK J., ‘Laser Cleaning of Stone Parts of the Tomb of the Unknown Soldier in Warsaw’, Ochrona zabytków 1 (204), LII, 39–44, 1999. MARCZAK J., ‘Renovation of Artworks Using Laser Radiation’, Przegląd Mechaniczny, No 15-16, 37–40, 1997. MARCZAK J., ‘Laser Radiation used for Renovation of Monuments and Artworks’, Proc. SPIE, 4238, Laser Technology VI: Applications, 219–225, 1999. MARCZAK J., ‘Surface Cleaning of Art Work by UV, VIS and IR Pulse Laser Radiation’, Proc. SPIE, 4402, Laser Techniques and Systems in Art Conservation, 202–209, 2001. MILLER (Editor), Laser Ablation, Springer Series in Materials Science, Vol. 28, Springer, Berlin and Heidelberg, 1994. SIANO S., SALIMBENI R. and PINI R., ‘Cleaning Processes of Encrusted Marbles by Nd:YAG Lasers Operating in Freerunning and Q – switched Regimes’, Applied Optics, 36, No. 27, 7073–7079, 1997. WILKINS M.L. (Editor), Modelling the behaviour of materials, Pergamon Press, London and New York, 1984. ZAFIROPULOS V. and FOTAKIS C., ‘Laser in the conservation of painted artworks’, Chapter VI in Laser cleaning in conservation: an introduction, Ed. M. Cooper, Butterworth Heinemann, Oxford, 79-90, 1998.

Jan Marczak* Institute of Optoelectronics Military University of Technology ul. Kaliskiego 2 00-980 Warszawa, Poland Graduated in 1973 from the Military University of Technology (MUT) and received his Ph.D. in 1988 from the same university. Specialist in the domain of laser interaction with matter, laser rangefinders and simulators. Head of the Department of Laser Devices and Techniques at the Institute of Optoelectronics of MUT. He has received a number of awards, including a gold medal for ‘Laser device ReNOVALaser for restoration/renovation of art works’ during the EUREKA’96 exhibition. He is a coordinator of the EUREKA !2542 Renova-Laser Project, as well as a contact member for the MUT team involved in the COST G7 Action ‘Artwork conservation with lasers’ and the COST G8 Action ‘Non-destructive testing of museum objects’. Member of the Polish Optoelectronic Committee and the International Society for Optical Engineering.

Andrzej Koss Interacademy Institute of Conservation and Restoration of Works of Art, Academy of Fine Arts Wybrzeże Kościuszkowskie 37 00-379 Warszawa, Poland

252


				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:21
posted:11/12/2008
language:English
pages:4