International Journalot Fatigue www: clsevier com/ocateifitigue ELSEVIER International Journal of Fatigue 24 (2002) 1021 Laser shock processing and its effects on microstructure and properties of metal alloys: a review Charles S. Montross *, Tao Wei *, Lin Ye **, Graham Clark *, Yiu-Wing Mai * * Center of Expertise in Damage Mechanics, Center for Advance Materials Technology, School of Aerospace, Mechanical and Mechatronic Engineering, University of Sydney, Sydney, NSW 2006, Australia » dinframes and Engines Division, Aeronautical and Maritime Research Laboratory, 506 Lorimer Street, Fishermens Bend, Victoria 3207, “Australia Received 1 September 2001; received in revised form 20 November 2001; accepted 7 February 2002 Abstract The current status of research and development on laser shock processing of metals, also known as laser peening, using Q-switched high power lasers is reviewed. The influence of processing parameters on the laser-induced shock waves in metal components are discussed and analyzed. Special attention is paid to the residual stresses and improved fatigue performance from laser peenin Which are compared with conventional shot peening results. Modification of microstructure, surface morphology, hardness, and strength by laser peening is also discussed. Finally, applications of laser peening are addressed, Results to date indicate that laser Peening has great potential as a means of improving the mechanical performance of components. © 2002 Elsevier Science Ltd Al rights reserved. Keywords: Laser shock processing; Laser peenng; Residual sess; Fatigue; Microtueture; Morphology 1. Introduction For the past six decades, shot pening has been the most effective and widely used method of introducing compressive residual stresses into the surface of metals to improve fatigue performance. Shot peening is rela- tively inexpensive, uses robust process equipment, and can be used on large or small areas as required. How- ever, the shot peening process has its limitations. In determining the compressive stresses produced, the pro- cess was semi-quantitative and depended upon a metal strip of gage called an Almen type gage to provide an indication of shot peening intensity. This gage did not guarantee that the shot peening intensity would be uni- form across the component being pened, Secondly, the compressive residual stresses were limited in depth, usu- ally not exceeding 0.25 mm in soft metals such as alumi num alloys and less in harder metals. Thirdly, the peen- Comresponding author. Tel: 461-2-9351-4798; fx: +61-2-9351- 7060. E-mail address: [email protected] (L. Ye). PU $0142-1123(02)00022-1 ing process resulted in a roughened surface, especially in soft metals like aluminum. This roughness needs to be removed before use in wear applications and typical processes used removed the majority of the compressive layer. An alternative novel surface processing tech- nology, namely laser shock processing (also known as laser pening), can induce greater depths of residual stress into metal surfaces using high-power, Q-switched laser pulses. The ability of a pulsed laser beam to generate shock waves was first recognized and explored in the carly 19605 [1,2]. Subsequent studies established the con- ditions for enhancing the amplitude of stress waves, ‘making it possible to induce plastic deformation in met- allic targets [3-6]. Prototype facilities were developed and initial feasibility studies were performed at Battelle Columbus Laboratories (USA) for utilizing laser induced shock waves to modify material properties [7-13]. After- wards, many research groups, in particular laboratories in France such as CLFA (Cooperation Laser Franco- Allemande—Areueil Cedex), LALP__ (Laboratoire Application des Lasers de Puissance—Arcueil Cedex) and LULI (Laboratoire °Utilisation des Lasers 23/028 - se font mater © 2002 Elsevier Science Lu. All rights reserved102 ©S Montross eta. / Intemational Journal of Fatigue 24 (2002) 1021-1036 Intenses—Evole Polytechnique, Palaiseau Cedex), expended much effort in the investigation and exploi- {ation of this process for industrial applications [14-19] Through these efforts, laser peening is emerging as an altemative to conventional shot peening processes. The object of this article is to present an overview of the state of the art of laser shock processing, highlighting some applications of this process. This field has been reviewed earlier with an emphasis on some selected top- ies [20-22] but no comprehensive review of process optimization and material aspects is available to date 2. Generation of laser shock waves When the power density of a laser pulse from a Q- switched laser is sufficiently high and when the pulse hits a metal surface, shock waves can be generated. As compared with continuously generated laser beams from helium-neon or carbon dioxide lasers, in Q-switched lasers the laser pulse resonates (or Q-switches) through the lasing medium until the laser pulse exceeds a thres- hold and leaves the laser. In this way, Q-switched lasers can generate GW beams of 1100+ J pulses less than 100 ns in length as compared with kW beams from con- tinuous carbon dioxide lasers The plastic deformation caused by the shock wave while propagating into the metal can result in residual compressive stresses. The physics and mechanisms of shock wave generation by lasers has been investigated intensively [7,23-28]. In carly published experiments [2,4], the irradiated target was in a vacuum and the laser produced a plasma that expanded freely. The resulting peak laser induced shock wave pressures ranged from 1 GPa up to | TPa when the incident laser power density was varied from about 0.1. GW/ent? to 10° GWiem?. The time duration of the shock wave pressure pulse was roughly equal to the laser pulse duration, typically 50 ns in length, because of the rapid adiabatic cooling of the laser generated plasma in the vacuum [11,20] ‘The use of coatings transparent to the laser energy as shown in Fig, I has been found to inerease the shock wave intensity propagating into the metal by up to two orders of magnitude, as compared to plasmas generated Focused Pulse Laser ‘Shock Wave ‘Trapped Plasma “Transparent Overlay Absorbent Coating 1 Metal Target Fig.l. Schematie representation ofthe laser peening in a vacuum [7,14,29]. The increase in shock wave inten- sity is achieved because the transparent coating prevents the laser generated plasma from expanding rapidly away from the surface. This results in more of the laser energy being delivered into the material as a shock wave [30]. For a Né:glass laser with a wavelength of 1064 nm, transparent coatings of water, quartz, or glass can increase the shock wave intensity. These transparent coatings are also known as overlays. ‘The use of laser absorbent sacrificial coatings has also ‘been found to increase the shock wave intensity in addition to the protection of the metal’s surface from laser ablation and melting. Metal coatings such as alumi- num, zine, or copper and organic coatings have been found to be beneficial if not necessary to protect the component's surface [7]. Among the absorbent coatings, commercially available flat black paint has been found to be the most practical and effective, as compared to other coating systems [31] ‘When a laser pulse with sufficient intensity irradiates a metal target with an absorbent coating, the absorbent material vaporizes and forms a plasma. Because of the short energy deposition times, the diffusion of thermal energy away from the interaction zone is limited to a couple of microns and is preferred to be less than the thickness ofthe absorbent coating to maintain protection This is important for aluminum alloys since surface ablation processes can detrimentally affect fatigue life [9]. The plasma continues to strongly absorb the laser energy until the end of the energy deposition (11). The hydrodynamic expansion of the heated plasma in the confined region between the metal target and the trans- parent overlay creates a high amplitude, short duration pressure pulse. A portion of this energy propagates as a shock wave into the metal. When the pressure of the shock wave exceeds the dynamic yield strength of the metal, plastic deformation occurs which consequently modifies the near-surface microstructure and properties [22]. This critical pressure Pj, is related to the density of the material (p,) and its elastic wave sound velocity (0), where P,, ‘Typical reported processing para- meters for laser peening have been collated in Table 1 Cost, efficiency, reliability, maintenance, and part- replacement requirements are all important factors in selection of the laser system for laser shock processing, Systems suitable for laser shock processing should have an energy output in the range of 10-00 J/pulse with pulse lengths of less than 100 ns. Currently, the only technology that can realistically deliver this type of energy output with an acceptable pulse length is a Q- switched laser system based on a neodymium-doped glass or yttrium aluminum gamet (YAG) erystal lasing rod which operates in the near infrared at a wavelength of 1064 nm and pulse duration of 10-100 ns. This has been fortunate since of the three wavelengths that have been most commonly used and easiest to pro-025 €S. Montross eta. / international Journal of Fatigue 24 (2002) 1021-1036 a) St aN GH OSH ME HT SUE ss019.0N 53] (am se) 2m EE D-) ad MI, ro DVAPN i] ost 3tON tnd 3201 pad SPN, se) 2U0N oN, s ax ss] SZ (ame 2) sag ts sso ss] oes 518g, ed 40 oasoup re ore 8 oF sR gue aie atsoupe Sse PN a su0N, 0 ox SRN z suo (unt 90 oot ssai:PN ) S808 Se'81 STN uwng) —snsejd pay unusumyy sre ‘on S80PN ELE-Su0L “TE9E-SL0L say 20 nen) und some oso SsoTPN —“ISkL-PE0e “ISeL-Fe0e ove-ve oz nent) se ‘rT ons 91-1909 "Ze1F 9805 arent) pet oe-sz Tro 00s (an ur OF £0) Hem 20 zAENg OTs) sme yoeIg “su “PHT orve 1 00s ©, (eao) (au ame (60) (ujq9) mod soounisjoy —amssang —Kayao maundsueyy, Sueos woquosey Raq Ose WoneMp aging AysUEp HOG — PSP] dK sy jus pores stout yo ussesaud ypoys se] aoj pouodar sossureed Husssooeudns C5. Montross et al. International Journal of Fatigue 24 (2002) 1021-1036 duce, {1064 nm (near infra-red), $32 nm (green) or 355 nm (ultraviolet)}, the near infra-red wavelength has only a modest absorption coefficient in water overlays, sul ficient interaction with the metals surface, and high dielectric breakdown threshold, Although the 532 nm (green) wavelength has the lowest absorption in water, generating the 532. nm (green) or 355 nm (ultraviolet) wavelengths with harmonic single crystals from 1064 rm laser pulses reduces the laser pulse energy output by a factor of two or more due to the conversion process and to associated inefficiencies, Although metals can be highly reflective of light, while keeping the laser power density constant, decreas- ing the wavelength from 1064 to 355 nm increases the photon-metal interaction enhancing shock wave gener- ation. However, this can also decrease the peak pressure possible because decreasing the wavelength decreases the etitical power density threshold for dielectric break- down which in tum limits the peak pressure possible [7,32]. Dielectric breakdown is the generation of a plasma not on the target’s surface which absorbs the incoming laser pulse, limiting the energy to generate a shock wave. In Fig. 2, decreasing the wavelength from 1064 to 532 nm decreases the dielectric breakdown threshold from 10 to 6 GW/em’, resulting in maximum peak pressures of approximately 5.5 and 4.5 GPa, respectively. Fast-firing Q-switched Né:glass lasers at the Lawr- ence Livermore National Laboratory (LLNL) [21,33,34] were recently developed which deliver 200 J pulses with 20 ns pulse lengths at pulse rates of 10 Hz. This delivery rate is 20 times faster than currently available commer- cial systems. If the costs and reliability of the systems can be made commercially acceptable, this LLNL laser system makes laser peening potentially more economical and attractive than current systems, It should also be noted that ultraviolet excimer gas lasers are currently 8 . . 5 a : B : £ 1084 om 22 © 532m i + am i Diolectfic Breakdown Thresholds Power Density (GWiem?) Fig. 2. Peak pressures oblained in the water confining regime as a function of laser power density at 1.064 um [87], 0.832 wm and 0.355 tum laser wavelengths [32] being developed to emit pulses a few tens of nanose- conds in length at a repetition rate better than 10 Hz. ‘The laser pulse delivery rate of 10 Hz would allow rapid laser peening of large areas of components but the cur- rent small output energies per pulse result in smaller laser spots thus limiting the application of this tech- nology. In the future, such lasers may overcome current problems in laser pening such as lower efficiencies and limited wavelengths. 3. Characteristics of residual stresses induced by laser shock peening In a typical laser shock processing operation with an absorbent coating and confining medium, a uni-axial compressive stress is generated along the direction of the shock wave. As the shock wave propagates into the metallic target, plastic deformation occurs to a depth at which the peak pressure no longer exceeds the metal’s Hugoniot elastic limit (HEL), A metal’s HEL is related to the dynamic yield strength according to Johnson and Rhode [35] a where v is Poisson's ratio and 0%" is the dynamic yield strength at high strain rates. Most investigations have concentrated on experimen- tally determining surface and in-depth residual stresses induced in different laser shock processing configur- ations for a number of industrial metals, such as alumi- num alloys [22,36,37], steels [38,39], and titanium alloys [40]. The residual stress measurements are usually per- formed using either of two methods. A non-destructive method involves surface analysis of the laser peened spot by the X-ray diffraction sin? y method. The method used to determine the depth and magnitude of the residual stress is the incremental hole strain gauge rosette method where a small diameter hole, typically I-3 mm, is drilled into the metal between a strain gauge rosette. As the hole is drilled into the metal, changes in the stress distribution are identified by the strain gauge rosette, From this data, the residual stress as a function of depth can be caleu- lated ‘The experimentally measured residual compressive stress across the treated area was found to be a relatively uniform biaxial in-plane distribution after typical laser peening treatment, The residual stress is usually highest at the surface and decreases gradually with distance below the surface. Fig 3 gives a typical depth profile for the residual stress in a 2024-T3 aluminum alloy, showing that the compressive stresses reach a depth of over 1 mm (41).C5. Monross etal. International Journal of Fatigue 24 (2002) 1021-1036 hoas: 100 & Residual Stress [MPa] & ony 05 ry 18 Depth Below Surface fmm] Fig. 3. The depth of the residual stresses in G-mm thick 2034-13 aluminum alloy [41 3.1. Elastic-plastic model for residual stresses ‘The basic mechanics of laser induced plastic defor- mation, relating shock wave stresses with resulting residual stresses, is difficult to analyze because of the discontinuity of the shock wave front and the three dimensionality of the stress system. Traditional shock wave work using explosives preferred to generate large planar shock waves which can be analyzed in one dimension, greatly simplifying analys ‘An early analysis of shock wave propagation behav- iour and basic mechanics attempted using hydrodynamic shock wave codes and compared the predictions with etched specimens [9]. The calculated results crudely matched the experimental results. In line with the histori- cal analyses of explosively driven shock waves, high powered lasers were used to cause spallation of alumi- ‘num and copper foils [42-44]. These experimental results were then compared with the results from the various one- and two-dimensional analytical computer codes with reasonable agreement. ‘A more detailed analytical elastic-plastic model was developed to calculate surface residual stresses induced by laser shock processing with the shock wave propagat- ing into an elastic-perfectly plastic metal half space [45— 48]. The large planar shock wave allowed one-dimen- sional shock wave assumptions to be used and to avoid edge effects, The shock-induced deformation was con- sidered to be uniaxial and planar, the pressure pulse was considered to be uniform in space, and the materials obeyed Von Mises yield criterion, The analytical method used, the method of characteristics, is based on the litera- ture and also assumes that the shock wave unloading is clastic. The computational technique was considered classical, requiring heavy algebra to analyze, and only the results were given [47,48]. ‘The model considers that the plastic deformation increases linearly between 1XHEL and 2xHEL. Above 2xHEL, plastic deformation is assumed to reach a ‘maximum limit and above 2.5xHEL, the surface release waves occur and amplify from the edges of the impacts thus modifying the residual stress field, regardless of the ‘impact shape (circular or square). The plastically affec- ted depth, Z,, and the maximum surface residual stress Gouc 6an be calculated from Ballard et al. [46,47]: CoCxit Cau @ CoCyit %C=Cy, 8) where C, and Cy are the elastic and plastic shock wave velocities, respectively; 7 is the pressure pulse duration; 2. and fare the elastic Lame’s constants of the target; Ly is the plastically affected depth; and r is the radius of the impact. These equations are considered valid only when P, the shock wave pressure, is greater than twice the material’s HEL. The finite element method (FEM) was first used by Braisted and Brockman [49] to analyze and predict the residual stresses from a laser generated shock wave. ‘There were some problems matching theory with experi ‘mental results published elsewhere, in particular due to the release wave affecting stresses in the center of the spot. Ding et al. [50] also used the FEM to analyze single and multiple laser generated shock waves propagating into a 35CD4 steel. The ABAQUS/Explicit FEM pro- «gram was used to simulate the propagation of the shock wave with the ABAQUS/Standard FEM program used to calculate the resulting residual stresses. The residual stresses were calculated in two dimensions, taking advantage of the circular symmetry of the round laser spots to simplify the analyses. ABAQUSIExplicit is a nonlinear elastic-plastic explicit time integration FEM code which is especially suited for high speed dynamic events needing a high resolution solution. A close match between these simulated results with experimentally measured residual stresses was found. 3.2. Residual stresses and absorbent coatings In laser pening with no transparent overlay, the laser induced plasma absorbs the incident laser energy and rapidly expands away from the solid surface. Conse- quently, the absorbed, incident laser energy cannot be efficiently converted into a pressure pulse and therefore cannot induce compressive stresses in the substrate by a shock wave effectively. For example, when a Hadfield steel specimen was laser shocked with no transparent overlay, the resulting surface stress on the specimen was only slightly compressive [51]. This occurred even with a very high laser power density of 2.4 TW/em? as com-6 C5. Montross et al. International Journal of Fatigue 24 (2002) 1021-1036 pared with the power densities of 1 GWyem? commonly used elsewhere. As discussed in Section 2, absorbent coatings in a typical laser peening operation are used to protect the substrate from thermal effects of the ablation [10] and can be used to increase the shock wave amplitude on the surface of a substrate [52]. As shown in Fig. 4, while the specimens with coatings have high compressive stresses in the surface, uncoated specimens have very high tensile stress values even with a water overlay. These tensile stresses were attributed to severe surface melting during the laser shock processing, To reduce processing costs of coating/painting for an aluminum surface to be laser peened, attempts have been made to identify laser pulse parameters that will reduce or eliminate severe surface ablation, eliminating coating requirements [53,54]. On the bare surface of a 2024-T62 aluminum specimen, a laser pulse with a duration of 30 nis was found to be effective to shock the surface and produce residual compressive stresses without melting the surface. Pulses of a duration of 13 ns were not long enough to generate sufficient shock pressures while pul- ses of a duration of 50 ns were too long, resulting. in melting and damage of the surface. Pulses of a 40 ns duration were found to result in the melting and damag- ing of a 2011-T3 aluminum surface [55] but seemed to have minimal effects on tool steel alloys. 3.3. Laser spot size and shock wave propagation The laser spot diameter can be varied and is limited only by the power density required and the laser power available. Varying the spot size affected the propagation behaviour of 1.2 mm and $ mm diameter impacts on SSCI steel foil specimens 620 um thick [56]. The smaller diameter shock wave expanded like a sphere, which resulted in an attenuation at the rate of I/r. The larger diameter shock wave behaved like a planar front, z on Ee = | (Coating 7 i= == : rm Spotd a 8 twtr — Fig. 4. Average residual stress values determined atthe SSCI stel surface aller laser peening [52] which attenuated at the rate of 1/ The net result was that the energy attenuation rate was less for the larger diameter, planar shock wave which therefore propagated further into the metal. This was also seen in shock wave propagation in rock where the shock wave from a 10 J pulse was thought to decay like a spherical shock wave over 10 mm [57]. The shock wave from the 100 J pulse behaved like a planar shock wave and propagated 25 mm into the rock. 3.4, Residual stresses and laser shock intensity ‘The amplitude of the surface stress increases with the ‘magnitude of the laser-induced pressure, which is related to the incident power density. When the incident power density exceeds a threshold, residual stresses increase with depth but decrease at the surface because of surface release waves. This threshold indicates that there are optimal shock conditions. For instance, surface com- pressive residual stresses in an A356-T6 alloy specimen increased up to 145 MPa for a laser-induced pressure increase from 1.3 to 1.5 GPa for a laser power density range change of 1.5-2 GWiem?, respectively. However, a further power density increase to 3 GW/em: tended to reduce the stress level to 100 MPa, whereas the in-depth compressive residual stress level continued to increase {58}. 3.5, Residual stresses and multiple laser shocks Compressive residual stresses can also be generated deeper below the surface by using successive shocks. Fig. 5 shows the results of two successive laser shots at the same spot on a Ti-6Al-4V titanium alloy. It can be seen that two successive laser shots generated residual Residual Stes [MPa] 0 02 Oo 06 08 Dept into Material frm] Fig. 5. Comparison between residual compressive stresses induced by two successive laser shots and one single laser shot in Ti-AI-4V titanium alloy using laser pulse energy density of 200 Jem? and a pulse duration of 30 ns [33]C5. Monross etal. International Journal of Fatigue 24 (2002) 1021-1036 to? stresses deeper than a single shot [59]. The same trend was observed in other alloy systems. For a 0.55% carbon steel, as the number of shots on the surface increased from one to three, the depth of compressive residual stresses increased from 0.9 to 1.8 mm [60]. For a 7075 aluminum alloy, repeated impacts had a very beneficial effect on the compressive residual stress levels on the surface. A 4 GWiem? treatment generates a residual compressive stress of 170 MPa at the first impact, 240 MPa at second impact and 340 MPa after the third impact [58] 3.6, Residual stresses and overlapping of laser spots Due to power density requirements and laser power availability, even laser systems with the potential for 200 J per pulse will be limited in the area covered per pulse. Overlapping of the laser spo is the method used to laser peen larger areas or large components. Overlapping of laser spots has been investigated for many industrial metals such as 1026 steel and ductile cast iron [41], A356, AII2Si, and 7075 aluminum alloys [58], SSCI steel [52] and titanium [40]. The results showed that there was a relatively uniform distribution of compress- ive residual stresses across the overlapped regions after laser shock processing. No indication of tensile residual stresses in the overlap regions was reported by the researchers, which indicated that little or no degradation of the properties would occur when larger areas were treated ‘Two laser impact diameters with the same power den- sity of 5 GWiem? were used to investigate the interaction of both overlapping and spot sizes on material properties [52]. Laser shock processing was performed on S5CI steel with 6 mm diameter and 1 mm diameter overlap- ping spots. The large, 6 mm diameter spots were big enough to span the notch lengthwise along the specimen and were overlapped 50% in one direction only across the width of the specimen’s notch. The smaller | mm diameter spots were overlapped 25% in two directions along the length and width of the specimen’s notch. Residual stress measurements showed that, with a small impact configuration, the plastically affected depths were strongly reduced. This confirms the reduced propagation distance results for small laser spots [56] Proper alignment and overlap of the laser spots will be critical. Analysis was conducted with a micro inden- tation system of a polished cross section through laser peen spots that were adjacent but not overlapping [61] ‘A decrease in hardness between the two laser peen spots was found. It was hypothesized that the shock wave affected the previously shocked zone with an expansive release wave. Expansive release waves have been found to follow intense shock waves [38] 3.7, Residual stresses and laser shock processing thin sections For thin sections, the use of shot pening is not practi- cal due to the potential damage from the process. Laser peening is considered suitable for thin sections [62]. In thin sections, increasing the intensity of the laser pening process did not increase the depth of compressive stresses, but did increase the magnitude of the in-depth compressive stresses. This observation was seen by Masse and Barreau [60] for a high strength 4340 steel sheet with a thickness of 1.5 mm. The sheet was laser shock peened with one and five shots from both sides simultaneously and the residual stresses are plotted in Fig. 6. The depth of the compressive stress was nomin- ally 0.5 mm for both peening conditions, but the magni- tude of the compressive stresses was higher for the five shots. However, the tensile residual stress at the mid- thickness of the sheet were also higher for the five shots, which can be controlled by carefully selecting the pro- cessing conditions. 3.8, Comparison between laser peening and shot pening ‘The actual depth of laser peening induced stresses var- ies depending on processing conditions and material properties, and generally ranges from 0.5 to over I mm. Moreover, smaller surface stress gradients are found after laser peening, which is beneficial because it is known to be important in reducing or eliminating eycli stress relaxation. Residual stresses induced by laser peening of Inconel 718 alloy are shown in Fig. 7, and compared with the typical results achieved by conven- tional shot peening [33]. Residual stresses are much deeper for laser pening than for conventional shot peen- ing. © Aseceived 1 Shot, LSP, © 5Sh0s, LSP _Residval Stress [MPa] Beess obs 1200 00 01 02 03 04 05 06 O7 08 Depth [mm] cl shet after different laser (60). Fig. 6. Residual stress profiles in 4340s peening intensities (the mid-thickness of the sheet i at 0.75 nons, C5. Montross et al. International Journal of Fatigue 24 (2002) 1021-1036 — /comentonaly / shotPooned Laser Peened 1000] 1259) ‘B00 028 0s0 Ors ODS Dp into Material [rm Fig. 7. Residual stresses in the surface of Inconel 718 induced by laser pening and conventional shot peening (33] A review of both the shot peening and laser shock processing technologies was conducted by the US Air Force [63]. They focused on the leading edge of a tur- bine fan blade with an emphasis on reducing high cycle fatigue failure due to foreign object damage. Shot peen- ing was seen to have several limitations for high cycle fatigue. Shot peened blades did not meet the fatigue life- time requirements. Shot peening also creates an unac- ceptable surface finish, especially when deep residual stresses are attempted. ‘A quantitative comparison between the loading con- ditions induced by laser pening with a water overlay and shot peening is presented in Table 2. The most dis- tinctive change in the impact conditions involves the duration of the induced peak pressures, which are 10- 20 times longer in the case of shot pening, Shot-peened surfaces are subjected to more multiaxial, intense load- ings than the laser peened surfaces. ‘An investigation combining laser peening with shot peening of 7075 aluminum alloy specimen showed that this combination can result in enhanced properties with increases in both in-depth and superficial compressive residual stresses [58]. Further studies should provide interesting information on the mechanical improvements generated by the combination of the two processes. Table 2 4. Modifications in surface morphology and microstructure The surface morphology of metals has a great effect on fatigue behaviour. Many investigations related to the surface morphology of laser shock processed materials have been performed with SEM observations and rough- ness measurements. When no protective laser-absorbent coating was used on the substrate, the laser shock pro- cessing causes severe surface melting and vaporization, particularly in aluminum [9,64]. This can result in resol- idified droplets and craters leading to very rough sur- faces. These problems can be solved with energy absorb- ent coatings as discussed in previous sections. In contrast, shot peening generates @ rough surface with large inereases in the value of the mean and peak roughness. This can be advantageous for paint adhesion but is detrimental to wear and fatigue properties. A com- parison of the roughening effects of laser peening and shot peening on A356 and 7075 aluminum alloys are shown in Table 3. Similar results were observed on a Hadfield steel specimen [51]. For wear applications, removal of the roughened surface is a necessity. How- ever, due to the thinness of the shot pened compressive layer, removal of the rough surface also produces a sig- nificant decrease in the compressive layer thickness. Laser pening is a mechanical process for metals which is accompanied by significant microstructural and phase changes. These changes in microstructure have been investigated by means of TEM, SEM and X-ray diffraction analysis. Microstructural changes induced by Table 3 Comparative toughening eects with laser shock processing and shot peening (58) Material and processing (um) Rum) A356 as milled 07 62 ‘A386 LSP (2GWiem?, wo impacts) 11 15 ‘A386 Shot pecning (F38-50N, 0.3 mm 38 3 beads) 17075 a8 milled 06 52 1075 LSP (4GWiew, thee impacts) 13 u 7075 Shot peening (20-23A, 125%, 0.6 3.7 2 sm beads) Comparative loading conditions induced by laser shock processing and shot peening [58] Process Peak pressure (GPa) Diameter of impacts Pressure duration (is) Mechanical impulse (GPa Strain rates!) (enn i) isp. 06 Las os 0-03 10" Shot pesning 3-10 02-1 Ost 110 10°©S Montross eta. / Intemational Journal of Fatigue 24 (2002) 1021-1036 029 LSP have been related to the laser process parameters and the heat treatment condition of the alloys. The dislocation density increased significantly in laser peened aluminum alloys such as welded 5086-H32, 6061-T6 [10], 2024-T62 [36] but no quantitative analysis, has been noted. High dislocation densities were also a prominent microstructural feature in low carbon steels after laser peening [39]. Laser pecning of Hadfield manganese steel was found to induce extensive forma- tion of é-hep martensite and high density dislocations in the #Hfec austenite matrix [51]. Investigations of the effect of laser peening of weld zones in 18Ni(250) maraging steel showed that after the laser pening treatment, the austenite weld phase reverted to martensite and the dislocation density quali- tatively increased in the martensite matrix [39,63] Numerous twins as well as o-phase embryos located at the twin intersections were found in laser peened 304 austenitic stainless steel [16] and 3161. stainless steel [64] A Fe-Ni alloy was laser shock processed with a laser power density of 100 GWiem? and 10 TWiem? without ‘transparent overlay and absorbent coating [38]. Very thin twinned grains were found on the surface because of melting and rapid solidifying. A martensite transform- ation zone was found at the back face of the laser shocked Fe-Ni alloy sample due to reflection of the shock wave from the back face. ‘No fundamental understanding in the laser peening literature has been noted on the interaction of the miero- structure with laser induced shock waves and the resulting changes in the microstructure. The majority of, the reports on microstructural changes have been quali- tative with few quantitative details such as dislocation density The lack a fundamental understanding of the laser peening process was most clearly seen with the unexp ted property changes when the weld region of a 6061- T6 weld specimen was laser peened [66]. Laser shock waves of 3.5 and 6 GPa shock pressures, were used to laser peen various spots of the weld specimen from the base metal through the heat affected zone (HAZ) into the weld, One of the first property changes was the change in tie modulus in the laser pened aluminum which was not previously noted in the literature. The hardness of the metal increased as noted in the literature but the depth of the property changes varied with respect to the weld center and HAZ. The depth of the changes in hardness and clastic modulus varied from 1.5 mm in the as- received metal to up to 2.5 mm in the HAZ for the 3.5 GPa laser peen spots. For the 6 GPa laser peen spots, the depth of changes varied from 2 mm in the as- received metal to 3 mm in the HAZ. These depths were significantly greater than the depths reported in the literature such as by Peyre ot al. [58] The minimal change in hardness of the bulk material outside the HAZ from the 3.5 GPa pressure laser pen pulse has been noted before [13]. Laser peening has been reported to improve the hardness of underaged 2024- T351 but not peak aged materials like a 2024-7851, 075-T6S1, or 7075-T73. However, the fatigue and ten- sile strength was improved for all materials. For the 6061-T6 specimens unaffected by the welding, there was no change in hardness or strength reported [9]. Tt was hypothesized that the precipitation hardening in the T6 condition is significantly large enough to mask eny shock wave strain hardening. It was also hypothesized that exceeding a threshold shock wave pressure of 7.5, GPa was required to significantly change the properties of peak aged aluminum alloys (12). From the investi- gation of property changes with the UMIS [66], a shock wave with a pressure of 6 GPa was suficient to signifi- cantly increase the hardness in the bulk 6061-16 material Another example of the lack of fundamental under- standing of the laser peening process in metals is the self-limiting hardness changes in 6061-T6 specimens [67]. For 3.5 GPa shock wave pressures, one repetition increased hardness but not five repetitions. When 6 GPa shock wave pressures were used, both one and five rep- etitions increased hardness. Increases in the metal’s Hugoniot elastic limit and changes in the metal’s dynamic yield strength with each laser shock wave were identified as the two mechanisms responsible for the self-limiting hardness changes. With each laser gener- ated shock wave, the pressure must remain significantly greater than the increased Hugoniot clastic limit and dynamic yield strength for further shock processing and surface hardness changes to occur. 5. Effect on the mechanical properties Many materials display pronounced improvements in fatigue life with laser shock processing. The beneficial effects of laser shock processing may originate from sur- face compressive stresses in the large affected depth and surface quality, which delay the development of fatigue cracking. Investigations on several different aspects of the fatigue behaviours, such as fatigue life, fatigue strength and fretting fatigue, have been reported. 5.1. Fatigue life and strength Investigations on aluminum alloys, steels, and titanium alloys have shown that laser pening can increase fatigue strength. The initial research work focused on the effect of laser peening on the fatigue crack growth of pre-existing cracks, using different laser spot shapes on the pre-cracks [62]. The two different laser peening spot configurations around a hole in a 2024-T3 aluminum alloy specimen ate shown in Fig1030 C5. Montross et al. International Journal of Fatigue 24 (2002) 1021-1036 8(a). The specimen had a centered hole with small starter notches machined into its sides. The region around the hole was shocked simultaneously on both sides using split laser beams. Two different laser spot arrangements were used for laser pening, The first laser spot arrange- ment used a solid spot to treat the entire region around the hole, The second laser spot arrangement used only an annular-shaped area around the hole and notched region. The fatigue life increases for both laser spot arrangements are shown in Fig. 8(b). The results showed that the laser peened 2024-T3 specimens with the solid laser spot had a fatigue life about 40 times longer than the non-shocked ones, whereas those with the annular laser spot had a life about three times longer than the non-shocked ones. This work was later verified where fatigue specimens were laser peened around the pre- crack [41]. The fatigue lives were almost the same range as those of the laser peened materials tested without a pre-crack The effect of laser peening and shot peening on the early and later stages of crack propagation for notched 7075-17351 and cast aluminum alloy specimens was also examined and compared by Peyre et al. [58]. The notching process was used to localize any crack (a) cibzw hmser 04a : Seer Grain x 0364 1010] lfowction ae 0.188 dia ine Sasite BiB o2s ne (b) i son Sp & 4 a i er 1 1 ‘o Fig. 8, Increased fatigue life in 2024-73 aluminum after laser poen= ing [62]. (a) Specimen configuration and laser shocked region shape (dimensions in inches). (b) Fatigue life. Laser energy densities for 160 sides of laser peened specimens are: 7 and 75 Jem? (W); 81 and 80 em? (4; 82 and 78 Jem? (@); 84 and 78 Jem? initiation to the notch root, with a stress concentration of K=1.6-1.7, to minimize the amount of laser peening necessary. The laser peening process consisted of three square laser peen spots with a 50% overlap of the spots for the cast alloys and an overlapping rate of ~67% for the 7075 alloy specimens. Fatigue testing was done under three point bending with the stress ratio, R=0.1 at 40-50 Hz. Fig. 9 shows the fatigue crack growth behavior of as- received, shot peened, and laser peened specimens for an applied fatigue stress of 260 MPa, with R=0.1. Laser peening dramatically improved the fatigue life with clear differences in the early and later stages of crack growth. Compared to the as-received specimens, the fatigue life improvements from laser peening can be separated into a seven-fold increase in the early erack growth stage and only a three-fold increase in the later propagation stage. In contrast, shot peening only provided a homogeneous two- to three-fold increase in both early and late stages of growth when compared to the as-received specimens. ‘This difference in behaviour between laser peening and shot peening was attributed to surface embrittlement and surface roughening due to the shot peening process which creates sites at which cracks develop rapidly and tends to reduce the beneficial effects of the compressive residual stresses The bending fatigue properties of notched 7075- T7351 aluminum alloy specimens having received shot peening or laser peening are compared in Fig. 10. Shot peening provided an 11% increase in the fatigue strength at 10” cycles, while laser shock peening provided a 22% increase, as compared with the as-received untreated specimens. This improvement was explained by the greater depth of the residual compressive stress fields induced by laser peening as compared with shot peening. The effect of different laser spot sizes and configur- ations on the fatigue behaviour of steel and aluminum alloys was further investigated by Peyre et al. [52] a Tad —S cadirgtatre — az peened Initiation a 2 i 61 og o. Fig. 9. Comparison of initiation (crack development to a specified size, and usually the detectable limit) and late cracking stages at Gnur=260) MPa for crack detection tests on 7075-17351 aluminum alloy [58].C5. Monross etal. International Journal of Fatigue 24 (2002) 1021-1036 ost 108 108 108 10 108 Number of Oycles Fig. 10. om-¥ curves for untreated, shot-peened and lserpeened 7075-17351 alloys [58]. Notched specimens were again used as described pre- viously but fatigue testing was conducted by four point bending of the notched specimen with R=0.1. For the 'S5CI steel specimens, the small I mm laser peen spots had a strongly reduced residual stress depth as compared with the 6 mm spots. However, the surface residual stresses were approximately the same for both I and 6 mm spots with the same coating and conditions. The fatigue specimen laser peening process consisted of either four 6 mm diameter spots laser peen spots with a 50% overlap of the spots or 50, | mm diameter spots with a 25% overlap. From Fig. I, it can be seen that the small (1mm), overlapped laser spot configuration displayed an approximately equivalent fatigue strength improvement (490 MPa) at 2x10* eyeles with the larger (6 mm), over- lapped laser spot configuration (470 MPa). This is a sig- nificant improvement to the fatigue strength of 380 MPa at 2x10° cycles for the as-received material. This also 650 oon é. Ese Bs a é 490 MPs © 480 s70.MPa = 380 MPa 00 ‘0 08 10? Number of Cycles Fig. LI. SW curves of notched bending SSC1 steel samples treated by laser peening [52] indicates that laser-shock processing with small impacts could be considered as a potential method for improving the fatigue life of structures without having to use larger, more costly, and more difficult to use laser systems. In an industrial magazine, laser peening was reported to have improved the fatigue life of notched, 1.5 mm thick AISI 4340 hardened steel specimens with a Rockwell hardness of R.=54 before laser peening [68] With a notch radius and notch thickness of 7.62 mm and a laser peen spot size of 9.91 mm, fatigue testing showed that the fatigue strength was increased by 60-80%, from approximately 552-612 MPa to approximately 966— 1035 MPa. Other details such as R and testing frequency were not given, Laser peening also successfully improved the fatigue behaviour of titanium alloys, such as Ti6AI-4V which is used in turbine compressors, and Inconel super-alloys used in turbine hot sections. Initial tests of laser pened blades showed a 10-40% improvement in metal fatigue strength [34], allowing engines to operate at higher stress loads. Laser peening also significantly increased the resistance of titanium fan blades to early fatigue failure caused by foreign object damage (FOD). After treatment by laser peening, the fatigue life of the damaged blades was found to be equal to or higher than that of undam- aged blades without laser peening [69] Laser peening can also increase the fatigue strength of the welds. For example, fatigue strength of welded joints of I8Ni(250) maraging steel showed a 17% increase at 2x10° eycles by laser shock hardening the heat affected zones (39. The fatigue life of $456 alumi- rum alloy welds was also extended by laser peening [20]. For an applied stress amplitude of 158 MPa, the fatigue life increased from less than 50,000 eycles for untreated specimen to more than 3-6 million cycles without failure for laser peented specimens. 5.2. Fretting fatigue The fretting fatigue properties were investigated on laser peened 7075-T6 aluminum alloy [13] using dog- bone type specimens as shown in Fig. 12(a), Both sides of the regions around the fastener hole in the specimen ‘were laser peened simultaneously and the pad were laser peened with 13 mm diameter laser spots. The fretting fatigue results shown in Fig. 12(b) indicate that at a stress amplitude of 96.6 MPa, the fretting fatigue life was increased by at least two orders of magnitude. At the highest stress level of 110.4 MPa, the fatigue life ‘was still twice that of the unpeened specimens. 5.3. Stress corrosion cracking While there has been extensive investigations on the effect of laser peening on the fatigue behaviour of met- als, only two works have been noted on stress corrosion1032 CS Montross eta. / Intemational Journal of Fatigue 24 (2002) 1021-1036 (a) ome tmnt, ot aso Laser Shocked "3 [+ Head (3k) n Stop See Pas Es . 7 « 1 @ 17 Gdesto Fare Fig. 12. Inereased resistance to ftting fatigue around fastener holes aller laser peening for 7075-16 aluminum [13]. (a) Fretting fatigue Specimen configuration (dimensions ia mm), (b) Fring fatigue results, cracking, Scherpereel et al. [70] compared the effect of laser shock processing on the stress corrosion cracking behaviour of two stainless steels: an austenitic (AISI 316L) and a martensitic (Z12 CNDV 12,02). In a NaCl 0.01M+NaSO, 0.01M solution, open circuit and polsr- ization techniques were used to determine the metals” electrochemical parameters such as free and pitting potentials, and passive current densities. Laser pening found to be more important in the case of 316L than for the martensitic stainless steel. The pitting potentials, of the steel were not modified by the laser pening, but the free potentials were shifted to anodic values and the passive current densities reduced. Laser peening was investigated as a way to improve the stress corrosion cracking resistance of thermally sen- sitized (620°C for 24 h) type 304 stainless steel, cold worked 20% to simulate neutron irradiation damage [71]. A ereviced bent beam specimen was used to induce 41% tensile strain on the surface with the specimens subsequently corroded in water at 288°C in an autoclave for 500 h. The laser pening was found to be remarkably ‘more effective than conventional peening to increase the stress corrosion cracking (SCC) resistance of the ther- mally sensitized type 304 stainless steel, 5.4. Hardening and strengthening Laser shock processing can produce increases in metal surface hardness over the entire region of the laser imadiated area, The magnitude of surface hardening depends on the laser shock processing conditions, alloy type, and microstructure of the alloys. Moreover, it was demonstrated that laser-shock hardening can be used effectively to harden the weld zones of some alloys, such as welded 5086-H32 and 6061-T6 aluminum alloys [10,66], and 18Ni(250) maraging steel [65]. Laser peen- ing increased the yield strength of welded S086-H32 aluminum alloy to the level of the parent material and increased by 50% the tensile yield strength of the welds for 6061-6 aluminum alloy [10]. The significant strengthening of weld zones in both alloys was noted as due to the higher density dislocations induced by laser peening Experiments with aluminum alloys showed that the hardness properties of the non-heat treatable (5086-H32) and overaged (2024-13, 7075-T73) alloys were signifi- cantly improved as compared with the unshocked properties [11]. It was observed that a threshold of laser- induced pressure needed to be exceeded before much change in hardness occurred in the heat-treated alloys [13]. For an underaged alloy such as 2024-7351, this threshold pressure was approximately 2.5 GPa (Fig, 13(a)) while for an overaged 2024-T851, the threshold for change was 7.5 GPa (Fig. 13(b)). Little improvement in the hardness properties of the peak aged aluminum alloys (2024-T8, 7075-T6, and 6061-T6) was noted by some researchers for laser shock pressures of less than approximately 5 GPa. However, for a 6 GPa peak press- ure, changes in hardness were noted in peak aged 6061- 6 aluminum using a UMIS nano-indentor system [66]. A study of 304 stainless steel also shows that hardness increases with increasing number of shots, and further increases are still possible with more shots using a peak shock pressure of 4.9 GPa, as seen in Fig. 14, The increase in hardness was reported to be caused by an increase in the dislocation density with increasing laser shock repetitions. Below the surface, the shock hardening effect decreases with increasing distance from the surface. Peyre et al. [58] compared the effects of laser pening and shot pening on the surface hardness for 7075 and A356 aluminum alloys. As shown in Fig. 15, shot pe ing results in twice the surface hardness increase as com- pared with laser pening. This was attributed to the longer application of pressure in shot pening, which results in greater dislocation generation and motion. The number of slip planes activated by multiaxial surface loading in shot peening may affect the hardness. How- ever, some fundamentals concerning these mechanisms need to be further investigated. As discussed in a pre- vious section, this lack of a fundamental understandingC5. Monross etal. International Journal of Fatigue 24 (2002) 1021-1036 tos 180) (ey 5” ‘or 170) wo % of 2 4% i oe0 3m ‘sop ° isp ® Fiyer Plate Shocking +40) ° 5 10 15 Peak Pressure [GPa] Fig 13. Dependence of average surface hardness on peak shook Pressures, comparing laser shocking and flyer plate shocking, The Pulse durations in ns are shown beside each ata point (13). (a) 2024+ & Hardness [Kh] 8 200) 0 5 10 Number of Laser Shocks, Fig. 4, Increase of surface hardness for 304 stainless steel with increasing number of laser shots [20]. 140] 2 ster Pooning F327 a Ls caWar?, Gas Ovaiay) 2 Set Pomning F15208, 1204 LP sawar?, Gas Otay) 110} 100| 0 100 200 300 400 500 600 700 Depth [ur] Fig. 15. Vickers hardness measurements with a 25 g load on A3S6- ‘6 and 7075-17351 aluminum alloys teated by laser peening ad shot peening respectively (58) was most clearly seen with the unexpected property changes in the study of a laser pened 6061-T6 weld specimen [66]. For thin 2024-T351 aluminum alloy specimens, a hardness peak at mid-thickness was produced by laser peening from both sides simultaneously [13]. Thus this split-beam shocking procedure produced a more uniform through-thickness hardening in thinner sections than that in the case of a shot from one side only. 6. Applications of laser pening Since the development of the laser peening process, the strong interest in its commercialization can be seen by the number of patents issued on this process. The first two key patents [72,73] were issued in 1974 and 1983, respectively, not long after the benefits of the process ‘were first identified. In the period from 1996 to 2001, the General Electric Company alone received a minimum of 23 US patents based on laser peening. The increased compressive residual stress depth pro- duced by laser peening can significantly improve proper- ties and control the development and growth of surface cracks [74]. Many of the proposed applications of laser peening aim at increasing fatigue life and fatigue strength of structures as well as strengthening thin sec- tions [59,68]. Laser pening of braze repaired [75] tur- bine components and weld repaired [76] turbine compo- nents have been patented due to the clear improvement in properties. A unique advantage of the laser pening process is that the laser pulse beam can be adjusted and controlled in real time [77]. Through computer controlled laser application systems, the energy per pulse can be meas- ured and recorded for each location on the component being laser peened. If the applied laser pulse was below1034 ©S Montross eta. / Intemational Journal of Fatigue 24 (2002) 1021-1036 the specified energy, it can redone at that time rather than after the part has failed. Regions inaccessible to shot peening, such as small fillets and notches, can still be treated by laser peening [68,78]. If the location can bbe seen, it can be laser peened [79,80]. However, no data, have been found in the literature showing experimental results and possible benefits. The majority of the appli- cations presented have been proof of principle to encour- age investment for application research. The laser beam spot shape can be changed to suit the application. A laser beam with a square profile instead of a round one allows dense, uniform packing of the laser spots, Furthermore, the process is clean, and work- piece surface quality is essentially unaffected, especially for steel components. The potential of laser peening includes the possibility of ditect integration into manu- facturing production lines with a high degree of auto- ‘mation [77] The aerospace industry is leading the integration of methods to apply laser peening to many aetospace pro- ducts, such as turbine blades and rotor components [78,81], discs, gear shafts (82] and bearing components [83]. Laser peening could also be used to treat fastener holes in aircraft skins and to refurbish fastener holes in old aireraft in which cracks, not discemible by inspec- tion, have initiated, General Electric Aircraft Engines in the USA treated the leading edges of turbine fan blades [81] in FIO1-GE-102 turbine for the Rockwell B-IB bomber by laser peening in 1997, which enhanced fan blade durability and resistance to foreign object damage (FOD) without harming the surface finish [81]. Prot tion of turbine engine components against FOD [40] is a key priority of the US Ait Force, In addition, it was reported that laser pening would be applied to treat engines used in the Lockheed Martin F-16C/D [69] The applications of laser peening can be anticipated to expand from the current field of high value, low vol- ‘ume parts such as hip implants and biomedical compo- nents to higher volume components such as automobile parts, industrial equipment, and tooling in the near future as high power, Q-switched laser systems become more available [68,84]. 7. Conclusions Laser shock processing, also known as laser peening, can produce a compressive residual stress layer more than 1 mm deep in commercially available aluminum alloys, and has been shown to significantly improve fatigue performances. Morcover, laser peening has been shown to harden the surface and improve the mechanical properties of some structural metal components such as commercially available carbon steels, stainless stecls, cast irons, aluminum alloys, titanium alloys, and nickel- based super-alloys. However, more research is required. Early work in this field by American researchers in the 1970s to early 1980s focused on laser physics, shock wave generation, and initial investigations applying technology. This work was later repeated by French researchers in the late 1980s to 1990s, looking a bit further into the laser phys- ies and shock wave generation and also looking at apply- ing the technology. “Proof of principle” of the laser peening has been the focus of both groups with more systematic engineering research and analysis needed for applying the technology with respect to fatigue. Little has been done with respect to stress corrosion cracking or erosion/corrosion which are major metallurgical prob- Jems. Many applications, such as “on the fly” laser peen- ing or laser peening in corners and small fillets or places not readily accessible to shot peening have been patented for commercial reasons. Experimental data from such applications validating these applications is lacking, ‘The metallurgical physics of the laser pecning process has not been deeply looked at, Recent nano-indentation analysis of laser peened metal identified a number of phenomena previously missed that could affect the behaviour of the metal. Past research on explosively driven shock waves in metals was found to be useful source of information to explain such phenomena. aascr peening with smaller spots with overlapping is important due to various difficulties in making lasers with pulse energies in the range of 50-200 J and a rep- atition rate of 1 Hz, all in an economical manner, Initial ‘work was done by French researchers, and some proof of principle work was accomplished. But more system- atic work is needed addressing the effect of the degree of overlap, the use of planar or spherical shock waves, and in particular, the effect of gaps between laser spots due to laser misfire on the resulting mechanical and met- allurgical properties. Material type and heat treatment conditions should be also considered in process optimization. Prior to the nano-indentation analysis of the laser peened 6061-T6 weldment, the effect of heat treatment conditions on shock wave propagation was not clear. More process ‘modeling is also needed to understand the residual stress fields generated by laser peening, in particular over- lapped spots and repeated shocking of a selected are ‘A major problem for this field of research is the ‘tremendous commercial interest in applying this tech- nology as seen by the large number of patents produced by General Electric for turbine blade applications. Because of this commercial interest, the existing basic science and process experience is either buried within the various companies or ignored by a focus on commer- ally applicable empirical results. ‘The research com- ‘munity’s limited understanding is dangerous because if a laser pened component does fail catastrophically, what really occurred will be unknown. Few independentCS Montross eta. / Intemational Journal of Fatigue 24 (2002) 1021-1036 10! people will have the background able to analyze what happened let alone provide a preventative solution. Acknowledgements ‘The authors are grateful to the support from the Air- frames and Engines Division of the Aeronautical & Maritime Research Laboratory in Defence Science and Technology Organisation. The authors would like the acknowledge the comments from the referees which helped improve this paper. References [i] Askar yan CA, Moror EM, Pressure on evaporation of mater ina tation beam, Joumal of Experimenta and Theoretical Physics Late 1963: 16:1638-44. [2] White RM. Plastic wave generation by electon bombardment or elecromagnatie wave absorption. Jounal of Applied Physics 19638:2123-4 [5] Gregg DW, Thoms SI. Momentum transfer produced by focused laser gint pulses, Journal of Applied Physics 196627.2787-9, [8] Steen CH, York CM. Lase-taduced “blow-afl” phenomenon Applied Physics Levers 1968 12:369-71 [5] Andesolm NC. Lsergenerted stress waves. Applied Physics eters 1970; 16:113-8, (6) Jones ED. Uifast aser-induod stress waves i stds, Appi Physics Latere 1971 1833-5. [7] Farand BP, Clauer'AH, Jung RG, Wileox BA. Quantitative assessment of later-induced stress waves generated at confined surface. Appi Phyies Leters 1974 25431-3 (8) Farad BP, Clauer AH. Use of ase generated socks to improve the propetes of mals and alloys. Indusval Applications of High Power Laser Technology 197686:112-. [9] Fairand BP, Cluce AK, Wilcox BA. Pulsed laser induced defo: mation in ap Fe Wi Pet Sally. Metallurgical Transactions A 19TTNA:119-25, (U0) Faiand BP, Claver AH, Wileox BA. Laser shock hardening of veld zones in aluminum alloys. Metallurg! Transactions A I9T6AATSTI-G (01) Faiund BP, Clauer AIL Laser generated sess waves: their characteristics and their effec lo materials. Presened at Pre ‘American Inst. of Physics Conon ‘Taser Solid Interaction and Laser Processing’; 1978 (02] Fairind BP, Claicr AHL. Applications of lscr-induced stress waves, Presented at Lasers in ModersIndasey Seminae 197, 113] Claver AH, Fatand BP. Intracton of aser-indueed stress waves ‘with metal, Presented at Proc. ASM Conference Applications of Lasers in Materials Processing: Weshington DC; 197. (04) Faro R, Fournier J Ballard P, Devaux D, Viront J. Physical study ofaseeqroduced plasma in confined year. Journal of Appl Physics 1990,08:775-84, {15] Orta J, AL, Peney CM, Jones MG, Exkson CE. US Patent 21; General Bese Company, Schenectaly, New York; 1990, [16] Devaux D, Fabbro R, Toller L, Barmicki E. Generation of shock ‘waves by later induced plasms in confined geometry. Journal of| Applied Physies 1993:74:2268-73 Gerland M, Hallouin M, Presles HIN, Comparison of two new surface treatment processes, laser induced shock waves and pri= ‘mary explosive: application to fatigue behaviour, Materials Science and Engineering 1992;156A:175-82. 07) [18] Takayanagi N, US Patent 5,571,575; Toyota Jidosha Kabushiki Kaisha (Aichi, JP), 1996, [19] Peyre P, Fabbro , Bethe L, Dubouchet C. Laser shock pro- cessing of materials, physical processes involved and examples of applications. Journal of Laser Applications 1996; 135-41 [20] Claucr AH, Holbrook JH, Fairand BP. Fffocts of laser induced, shock waves on metals. In! Meyers MA, Murr LE, editors. Shock waves and high-sran-rate phenomena in metals. New York: Ple- ‘num Publishing Corporation; 1981. p. 675-702. [2] Borgstom RP, Laser shock processing. 1993;105:49.51 [22] Clauer AH. Laser shock peening for fatigue resistance. In: Gre ory JK, Rack HJ, Fylon D, editors, Surface performance of titanium” Warrendale (PA): TMS; 1996, p. 217-30, [23] O*Keofe JD, Skeen CH. Laserinduced sress-wave and impulse augmentation. Applied Physies Letters 1972;21-464-6, [4] Hoffman CG, Laser-aepet interactions, Journal of Applied Phys jes 1974,45:2125-8, [25] Yang LC. Stress waves generated in thin metalic films by a Q ssvtched ruby laser. Journal of Applied Physics 1974:45:2601-1, [26] Romain JP, Covet F, Hallouin M, Fabbro , Faal B, Pepin H, Laser shock experiments at presres above 100 Mbar. Physica 1986;139,1408:595-598 [27] Ling P, Wight CA. Laser-generated shock waves in thin films of ‘energetic materials, Journal of Applied Physics 1995;78:7022-5, [28] Couture S, de Resseduier M, Hallouin M, Romain JP, Bauer F. Shock profile induced by short laser pulses. Journl of Applied Physics 1996;79:9538-49, [29] Fairand BP, Wileox BA, Gallagher WI, Williams DN. Laser shock-induced mierostractural and mechanical property changes ‘in 7073 aluminum, Joural of Applicd Physics 1972;43:3802-6, [50] Montross CS, Florea V, Bolger JA. Laserinduced shock wave generation and shock wave enhancement in basalt. Intemational Journal of Rock Mechanics and Mining Sciences 1999;36:849- 5. [51] Montross C8, Florea V, Swain MY. Influence of coatings on sub- surface mechanzal properties of laser peeved 2011-7 aluni- Production (52) Bere, ato Rs Pere Ps Barc E. Wavelngthdepen- dence of laser shock wave generation in the water confinement regime, Journal of Applied Physics 1999;85:7552-5, [53] Dane CB, Hackel LA, Daly J, Harrison J. Sho: peeaing with lasers. Advanced Materials and Procestes 1998;153:37-8 [54] Ashley $. Powerful laser means beter peening. Mechanical Engineering 1998;120:12, [35] Johnson JN, Rhode RW. Dynamic deformation twinning in shock ded iron, Journal of Applied Paysics 1971;42:4171-82. [36] Zhang H, Yu C. Laser shock processing of 2024-762 aluminum alloy. Materials Science and Engineering 1998;2874:322-. [7] Clouer AH, Dulaney JL, Rice RC, Koucky JR. Laser shock pro essing for treating fstener holes in ging atest. In: Adu SN, Haris CE, Hoggard A, Mille N, Sampath SG, editors, Durability of metal aireraft structures. Atlanta (GA): Atlata Technology Publications; 1992, p. 350-61 [58] Grevey D, Maitiedy L, Vannes AB. Laser shock on a TRIP. alloy: mechanical and metallurgical consequences. Journal of Materials Science 1992:27:2110-6, [59] Banas G, Elbayed-Ali HE, Lawzence FV, Rigsbee JM, Laser shock-induced mechanical and microstructural modification of welded maraging steel Journal of Applied Physies 1990,672380-4, [Mo] Ruschaw 31, John R, Thompson SR, Nicholas T. Fatigue crack nucleation and growth rate behaviour of laser shock pened titanium. Intemational Journal of Fatigue 1999;21:8199-S209 [at] Cluer AH, Keucky JR Laser shock processing increases the fatigue life of metal pars. Materials and Processing 1991:6:5, [42] Cottet F, Marty L, Hallouin M, Romain P, Viemont J, Febbro1036 €S. Montross eta. / International Journal of Fatigue 24 (2002) 1021-1036 R, Fara B. Two-timensional study of shock breakout atthe rear face of laser iradated metalic targets. Journal of Applied Phys= ies 19m 644747, [43] Covtet F, Boustie M. Spallation studies in aluminum targets using shock waves indueed by laser irradiation at various pulse dur ations. Journal of Applied Physics 1989;66:4067-73, [44] de Resseguier T, Couturier 8, David J, Nirat G. Spallation of ‘metal targets subjected to intense laser shocks, Journal of Applied Physics 1997:82:2617-32, [45] Dubouchet C. PRD dissertation in Metallurgy. Orsey (France): Universite d’Oreay; 1995, [46] Ballard P. Contraintesresidvlles induites par impact rapide. Application au ehoe-laser.In- Metallurgy. Paris (France): Feole Polytechnique France; 1991, p. 217. [47] Ballard P, Fournier J, Fabbro R, Frelat J. Residual stresses induced by laser shocks, Joumal de Physique IV 1991;1:487-94, [4] Ballard P, Frelat J, Rougier Y, Girardot D. Pasic strain and residual stestflelds induced by @ homogecous impact of the boundary of an clastic half space. European Journal of Mech anies, A’Solids 1995;14:100S-16 [49] Braised W, Brockman R, Finite element simulation of laser shock peening. International Journal of Fatigue 1999;21:719-24, [50] Ding K, Ye L, Monioss CS. Finite element simulation of mul- Uiple laser shock peening of 3SCD4 stecl, Materials Processing ‘Technology, submited for publication [51] Chu JP, Rigsbee JM, Banas G, Lawrence FV, Fsayed-Ali HE, fects of laser shock processing on the microstructure and su face mechanical properties of Hadlield manganese stel. Metal- Turgiea and Materials Transsetins A 1995,26A:1507-17. [52] Peyre P, Berthe L, Scherperssl X, Fabbro R. Laser-shock pro- ‘essing of aluminum coated S5Ci stl in water-confinement regime, characterization and application to high-eyele fatigue ‘behaviour, Joural of Materials Science 1998;33:1421-8, [53] Zhang ¥, Zang 8, Zang X, Cai 1, Yang J, Ren N. Tnvestigation ‘of the surface qualities of laser shock processed zones and the effet on fatigue life of aluminum alloy, Surface and Coatings “Technology 19979251049. [54] Cai L, Zhang ¥. Study of laser parameters optimum of laser shocking against fatigue and fracture of metal. Chinese Journal of Lasers 1996;23A:1117-20, [55] Moniross CS, Florea V. Unpublished work, Redstone Mining, Sydney (Australia, 1997 [56] Fabbro R, Peyre P, Berthe L, Sherperel X. Physics and appli ‘ations of laser-shock processing. Joumal of Laser Applications 1998;10:265-79, [57] Bolger JA, Montross CS, Rode AV. Shock waves in basalt rock generated with igh powered lasers in a confined geometry. Jou nal of Applied Physics 1999;86:5461-6, [ss] Peyre P, Fabbro R, Merrion P, Lieurale HP. Laser shock prom essing of aluminum alloys. Application to high eyele fatigue behaviour. Materials Science and Engineering 1996,4210:102— ie [9] Dane CB, Hackel LA, Daly J, Harrison J. Laser peening of met als—enabling laser technology. Advanced Materials ané Pro- ‘eases 1997;May:13-27, [60] Masse JE, Barreay G. Surface modification by laser induced shock waves, Surface Engineering 1995;11:131-2 [61] Montoss CS, Florea V, Brandt M, Swain MV. The effeet of pecning with small, high energy, Q-wvitched laser spots on sub- surface properties oF aluminum. Joural of Materials Science Let ters, submitted for publication, [62] Clauer AH, Waltrs CT, Ford SC. The effects of laser shock processing onthe fatigus properties of 73 aluminum. In: Lasers in materials processing. Metals Park (OH): American Society for Metals 1983. p. 7-22, [63] Thompson SD, Soe D, Lykins CD, Sampson PG. Laser shock peening vs shot peening, a damage tolerance investigation. In Gregory JK, Rack HJ, Ejlon D, editors. Surface performance of titanium. Warrendale (PA): The Minerals, Metals & Materials Society: 1997. p. 238-81 Geriand M, Hallouin M. Bfect of pressure onthe microstructure of an austenitic stainfest steel shock-losded by very short laser pulses, Journal of Materials Science 1994;29:345-5 [65] Banas G, Lawrence FV, Rigsbee JM. Laser shock hardening of Weld maraging set, In: Meguld SA, editor, Surface engineering, New York: Flsevier Applied Seiencs; 1990. p. 280-50, [66] Montross CS, Florea V, Brandt M, Swain MV. Subsurface properties of laser peoned 6061-76 Al weldments, Surface Engin ering 2000;16:11621 [67] Montoss CS, Brandt M, Swain MV. SelFlimiing. hardness changes in laser peened 6061-T6 aluminum, Surface Engineering 2001:17(6)477-82 [6s] Vaceari JA. Laser shocking extends fatigue life. American Machinist 1992,ulys62-4 [69] Brown AS. A shocking way to strengthen metal, In: Aerospace America, 1998, p. 21-3, [70] Seherpereel X, Peyre P, Fano R, Lederer G, Celli N, Most tations of mechanical and electrochemical properties of stainless steels surfaces by laser shock processing. Presented at Proceed ings of SPIE-Duropto Conference; Munich (Germany); 1997 [71] Obata M, Sano Y, Mukai N, Yoda M, Shims $, Kanno M, Eifet of laser peening on residual stress and stess corosion cracking for type 304 stainless scl. Presented at International Conference ton Shot Peening—3; 1999, [72] Malloei PJ, Fairand BP, US Patent 3,850,698; Industil acerals Limited; 197 [73] Cluuer AH, Fairand BP, Ford SC, Walters CT. US Patent 4,401,477; Battelle Development Corporation; 1983, [74] Mannava §, Cowvie WD. US Patent §,559,018; General Eletric Company (Cincinnati, OH); 1996 [75] Mannava S, Ferrigno SJ. US Patent 5,675,892; General Electric Company (Cincinnati, OH); 1997 [76] Ferrigno Si, Cowie WD, Mannava S, US Patent 5,845,057; Gen- eral Fletrc Company (Cincinati, OH); 1998. [77] Mannava S. US Patent 5,756,965; General Electric Company (Cincinnati, OH); 1998, [78] Mannava S, MeDanie! AE, Cowie WD, US Patent 5,492,487; General Electric Company (Cincinnati, OF); 1996, [79] Clauer AH, Dulaney JL, Toller SM. Canada Patent 2226444; LSP. Technologies: 1998, [80] Cisuer AH, Dulaney JL, Toller SM, Walters CT, Canada Patent 2226451; LSP Technologies; 1998 [1] Mannava 8, McDaniel AE, Cowie WD, Halla H, Rhoda JE, Gut Knecht JE.” US Patent 5,591,008, General Electric Company (Cincinnati, OH; 1997 [82] Ferrigno Si, MeAllster KG, Mannava S. US Patent 6,200,689; Cincinnati (OH); General Eleene Company, 2001, [83] Casareia DA, Cowie WD, Mannava 8. US Patent 5,584,586; Cin inna (OH); General Blecrie Company, 1986 [84] Clauer AH. New life for laser shock processing Industrial Laser Review 1996:March:7-9 [85] Peyre P, Bere L, Scherperel X, abbr R, Barticki . Experi~ rental study of laser driven shock waves in stainless secs, Joure nal of Applied Physics 1998:84:5985-92, [86] Banas G, Chu JP, Rigsbee IM, Lawrence FV, Hlsayed-Ali HE, Laser shock processing of low carbon steel. In: Aliabadi MH, Brebbie CA, editors, Computer methods and. experimental measurements for sures treatment effect, South Hampton (UK): Computational Mechanics Publications; 199. p. 171-84 [87] Berthe L, Faboro R, Peyre P,TolirL, Barticki F. Shock waves fiom a waterconfined laser-geneated plasma, Joursal of Applied Physies 1997,82:2826-32, (64)