Corrosion Evaluation With 3D Metrology

The 10th International Conference of the Slovenian Society for Non-Destructive Testing Application of Contemporary Non-Destructive Testing in Engineering

September 1-3, 2009, Ljubljana, Slovenia

SURFACE EVALUATION OF LASER SHOCK PROCESSED ALUMINIUM ALLOY AFTER PITTING CORROSION ATTACK

WITH OPTICAL 3-D METROLOGY METHOD

U. Trdan a, J.L. Ocaa b, J. Grum a

aUniversity of Ljubljana, Faculty of Mechanical Engineering, Laboratory for Materials Testing

and Heat Treatment, Akereva 6, 1000 Ljubljana, Slovenia b Centro Lser U.P.M., Carretera de Valencia km. 7,300, 28031 Madrid, Spain

[email protected], [email protected], [email protected],

ABSTRACT The corrosion behaviour of the AlSi1MgMn laser-shock-processed aluminium alloy was investigated in 3.5% NaCl water solution at ph7 and 23C. Potentiodynamic polarization tests were conducted at a scanning rate of 10mV/s, starting at 2000mVSCE. Corrosion plots showed pitting corrosion attack. Surface conditions after corrosion tests at different pulse density levels were evaluated with optical microscopy, scanning electron microscopy (SEM), EDS analysis and innovative non-destructive 3D surface metrology method. The SEM analysis showed a large presence of chloride and oxide products near the pits. The non-destructive 3D metrology method proved to be an extremely perspective method to determine surface conditions in a way the traditional microscopy could not. The results of the research confirm its use in detail analysis of the pits formed at the surface as a result of electrochemical corrosion attack. The tests confirm that a greater pulse density of the surface treated guarantees a smaller number of pits per surface unit and is reflected also by the movement of the pitting potential towards bigger corrosion resistance. Key words: Laser shock processing, pitting corrosion, potentiodynamic polarization tests, 3D optical metrology, aluminium alloy. 1. Introduction

The 6xxx series aluminium alloys are frequently used in industrial application due to their low density, favourable mechanical properties, and excellent corrosion resistance. In the presence of chloride ions with a very deteriorating action, the protective property of a passive/oxide film at the surface of specimens made of aluminium alloys is drastically reduced, which results in big corrosion damages of small surface pits. This a type of corrosion is referred to as pitting corrosion [1],[2]. To reduce surface damage on the parts exposed to work under demanding conditions, an adequate surface treatment and protection is essential. It increases corrosion or wear resistance and prolongs service life. Laser-shock processing (LSP) is an innovative surface treatment based on plasma generation at the moment of the interaction of laser light with a workpiece material. LSP produces shock impact waves and elastoplastic shifts, which considerably increases

densification of dislocations in the thin surface layer reflecting mainly the high gradient compressive residual stresses [3]. Numerous studies [4] confirmed the applicability of LSP, particularly in terms of improved material resistance to corrosion and stress corrosion cracking (SCC). Furthermore, the microscopic analysis confirmed the presence of cracks in the as-delivered specimens whereas in the LSP-treated specimens no cracks were found. 2. Experiment The experiment performed included an analysis of the LSP treatment conditions impact on the pitting corrosion level in the presence of seawater. In the first stage, the corrosion attack was evaluated by means of the pitting potential and then with the 3D optical measurement of the treated surface. The 3D measurement has a crucial role in checking and controlling the properties and the function engineering parts. While such measurements have been made using tactile devices for several decades, there has recently been a strong shift towards optical non-destructive 3D metrology devices [5]. For the experimental work, aluminium alloy AlSi1MgMn (ENAW6082) in the precipitation-hardened state T-651 was chosen. The specific precipitation heat-treatment of the alloy produces matrix and phase precipitates (Mg2Si), i.e. intermetallic phases, which means higher material hardness and strength. The alloys chemical composition is given in Table 1. For experimental purposes, cylindrical specimens were cut from a round drawn rod of 40 mm to uniform discs of 20 mm in height. A suitable specimen preparation guarantees an average value of the average surface roughness parameter Sa of 0.55 m .

Table 1: Chemical composition of AlSi1MgMn / ENAW6082.

Element Al Mg Si Mn Fe Cr Zn Ti Cu wt. % bal. 0.9 0.8 0.8 0.5 0.25 0.2 0.14 0.1

The LSP treatment of the specimens was performed with a Q-switched Nd:YAG laser with a wavelength of =1.064 m. The laser pulse was well focused with the Gaussian distribution and a power density of 10.75 GW/cm2. Two levels of pulse density 900 pulses/cm2 and 2500 pulses/cm2 were chosen, laser pulse duration of 10 ns being uniform with a repetition of 10 Hz.The specimen was clamped in a movable computer-aided x-y table and submerged in water. At the interaction of the laser light with the material surface, high-energy plasma is generated due to a high temperature at an extremely small specimen surface area. The potentiodynamic polarisation tests were conducted with Voltalab 21 potentiostat/galvanostat, Radiometer Analytical. Prior to the corrosion tests, the specimen was cleaned by using ethanol and demineralized water. The specimens (working electrodes) with a surface area of 0.5 cm2 (d=8 mm) were exposed to the 3.5% NaCl water solution in the CEC/TH corrosion cell. A Pt electrode was used as a counter electrode, whereas the XR110 Calomel electrode with saturated KCl was used as a reference electrode. The data was established with a scan rate of a potential of 10 mV/s, the initial testing potential being -2000 mVSCE. The electrode potential was increased up to -500 mVSCE. The final analysis of the materials surface after the electrochemical corrosion tests was performed by applying the NDT 3D surface metrology method using the Infinite Focus (Alicona Imaging GmbH) system. The main feature of the system are high-resolution measurements even of complex surfaces, measuring surfaces with steep flanks up to 80 and strong varying roughness [6]. Focus-Variation combines a small focus depth of the optical system with vertical scanning to provide topographical and colour information obtained from focus variation [7]. In surface condition analysis, the choice of parameters was identical for all specimens, i.e. a magnification of 20x, seven million captured points, each point size being 438 nm.

3. Results 3.1. SEM and EDS analysis Microstructure analyses were carried out with the JEOL JXA-8600M scanning electron microscope. The SEM microstructure of the AlSi1MgMn aluminium alloy is given in Figure 1. Kellers reagent was used for the etching (94 mL water + 3 mL nitric acid + 2 mL hydrochloric acid + 1 mL hydrofluoric acid).

Table 2: EDS results for the chemical composition of the

participates (in wt.%).

Area Al Mg Si Mn Fe Zn Cu A bal. - - 0,35 0,24 - - B bal. - - 0,49 0,42 0,63 0,32 C bal. 0,40 0,55 0,46 0,21 - -

Fig. 1: Microstructure of the AlSi1MgMn alloy.

Figure 1 shows a basic aluminium matrix in which fine precipitates are homogeneously distributed. The latter were formed due to cold deformation and precipitation annealing. The size of the precipitates was between 0.5 and 1.0 m (point A, B), whereas the size of the larger precipitates was approximately 5 m (point C). The chemical composition of the precipitates is given in Table 2 and it was established by using three energy dispersive spectroscopy - EDS spectrographs for microchemical analysis. 3.2. Surface analysis prior to corrosion testing The examination of the selected surface and the testing of the LSP homogeneity was performed by using the 3D surface metrology method that enables a comprehensive surface analysis. In comparison with a tactile measurement device, optical measurement not only prevents touching and damaging the surface but also enables measurement of the entire area, instead of only surface profiles, and is typically much faster for detailed measurements of large areas [8]. The measurement of the surface conditions was carried out on a chosen area of 300 m x 300 m with a unified cut off wavelength c=200 m. Figure 2 shows the correlation coefficient model of the specimen before the LSP, i.e. in its basic condition, and of the LSP-treated specimen with an overlapping density of 900 pulses/cm2. The preferential direction of the surface texture is clearly visible. The non-treated specimen after cutting (Figure 2a) reveals a strong trend of the surface texture in the cutting direction whereas the LSP specimen (Figure 2b) shows no evident surface direction. We may therefore conclude that the surface of LSP-treated specimens was homogeneous.

Table 3: Parameter values of auto-correlation analysis.

No LSP 900 pulses/cm2 2500 pulses/cm2 Parameter Significant directed

texture of surface Surface without

preferred direction Surface without

preferred direction Sal [m] 8.763 24.668 30.981

Str [/] 0.056 0.640 0.792

(a)

(b)

Fig. 2: Auto correlation model of the specimen in the initial state (a) and LSP treated specimen

with 900 pulses/cm2 (b). The results of the analysis given in Table 3 confirmed a big difference between the treatment parameters. The highest parameter value, the auto-correlation length of Sal=30.981 m, was calculated for the LSP-treated specimen with 2500 pulses/cm2, which means that the surface is dominated by low frequencies. For the non-treated specimen, the calculated value was Sal=8.763 m, meaning that the surface is dominated by high frequencies. The parameter texture aspect ratio Str, determining the influence of the surface texture as expected increases with a higher overlap pulse level achieving its highest level of Str=0.792 at the rate of 2500 pulses/cm2.

Table 4: Surface roughness values of the specimens.

Surface roughness

No LSP 900 pulses/cm2

2500 pulses/cm2

Sa [m] 0.520 2.392 4.125 Sq [m] 0.653 3.039 5.115 Sp [m] 2.009 8.134 16.955

The surface roughness results for individual specimens are given in Table 4 according to treatment type. The condition of the surface layer was determined by using three parameters: average height of the selected area - Sa, root-mean-square height - Sq, and maximum peak height of the selected area - Sp. It can be concluded from the results that greater pulse density leads to greater surface roughness. The increase in surface roughness is a consequence of numerous laser-beam interactions with the specimen surface due to overlapping of tracks and cumulative action of shock waves which produces microplastic shifts in the surface layer. 3.3. Potentiodynamic polarization plots Figure 3 shows the potentiodynamic polarisation curves of aluminium-alloy specimens measured prior to and after the LSP. The tests were performed by using a medium of 3.5 % NaCl water solution with a pH7 and a temperature of 23 C. Prior to potentiodynamics scans, all specimens were immersed into the medium for 1h at constant open-circuit potential. A difference can be noted by electrochemical testing as early as in the initial potential. The highest initial potential at a current density of -15 mA/cm2 was recorded in the specimen in its basic condition, measuring -1814 mV, followed by the specimen treated with 900 pulses/cm2 with a value of -1831 mV and the specimen treated with the highest overlapping rate of 2500 pulses/cm2 with a value of 1886 mV.

-2

-1,6

-1,2

-0,8

-0,4

-15 -10 -5 0 5Current density [mA/cm2]

Pote

ntia

l [V

SCE]

No LSP900 pulses/cm22500 pulses/cm2

Fig. 3: Potentiodynamic polarization curves.

Table 5: Pitting potentials from corrosion polarization tests.

Pulse density[pulses/cm2]

Epitt [mVSCE]

Epitt [mVSCE]

No LSP - 782 0 900 - 720 + 62

2500 - 662 + 120

It can be inferred from the variation of polarization curves that the pitting potential - Epitt increases with an increasing overlap-pulse density level. Table 5 shows the results obtained in potentiodynamic polarization tests. Specimen treated with 900 pulses/cm2 showed an increase in the pitting potential of +62 mV. The specimens treated with 2500 pulses/cm2 showed an increase in the pitting potential of +120 mV compared to the same material in the as delivered state. 3.4. Surface analysis after corrosion tests To validate the improved post-LSP corrosion resistance even further, the surface condition of the specimens was checked by using the 3D surface metrology method. The specimen surface was cleaned in nitric acid (HNO3) for 120 seconds in accordance with the ASTM standard [9] for the preparation of specimens after the corrosion test. The site of the selected surface analysis area of individual specimens was approximately 1320 m x 980 m, with a vertical resolution of 100 nm. The example of measurement areas in real colour of two specimens are shown in Figure 4.

(a) (b)

Fig. 4: Measurement areas of individual specimen. Specimen without LSP treatment (a) and

LSP-treated specimen with 2500 pulses/cm2 (b). Figure 5 represents 3D images in the pseudo colour of the surface of different specimens after the electrochemical corrosion tests (ECT). It can be estimated from the surface representation of the given aluminium alloy that most corrosion damage (pits) occurs on the surface of the as-delivered specimens. The surface recordings of the LSP-treated specimens, on the other hand, reveal less pits at a higher pulse density. The furrowed surface of the LSP-treated specimens is only a reflection of the laser pulse interaction with the specimen surface and the resulting elasto-plastic movement of the atomic planes. The specimen surface condition was validated further by measuring surface roughness and by comparing the results thus obtained with initial roughness before the electrochemical corrosion tests for each individual treatment factor.

Increase of the pitting potential (Epit) with higher pulse density level

Increase of potential Epass

(a)

(b)

(c)

Fig. 5: Digital elevation model of the specimens after the electrochemical tests in pseudo colour. (a) initial state, (b) LSP treated specimen with 900 pulses/cm2 and (c) LSP with 2500 pulses/cm2. The results prior to and after the LSP-treatment are given in Table 6. Upon comparison of the roughness values Sa and Sq before and after the corrosion testing, considerable differences were determined with regard to treatment conditions. The highest roughness increase was thus recorded in the non-treated specimen, where roughness values Sa increased by 111 %, whereas roughness values Sq went up as far as 131.4% compared to the roughness of the specimen immediately after the metallographic cut. As expected, roughness increase of the LSP-treated specimens was minimal and was even reduced as the pulse overlapping rate went up.

Table 6: Surface roughness comparison before and after ECT.

Prior to ECT After ECT Comparison Specimen Sa [m] Sq [m] Sa [m] Sq [m] Sa [%] Sq [%]

Comment

No LSP 0.520 0.653 1.097 1.511 111 131.4 Not acceptable 900 pulses/cm2 2.392 3.039 2.612 3.287 9.20 8.16 Very good

2500 pulses/cm2 4.123 5.115 4.180 5.251 1.38 2.66 Excellent In the specimen treated with pulse rate of 900 pulses/cm2, the roughness value Sa increased by 9.20 %, while the Sq value only went up by 8.16 % compared to the specimen values before the ECT. At a higher overlapping rate of 2500 pulses/cm2, the increase in roughness values was even less, i.e. Sa=1.38% and Sq=2.66 %. Hence the roughness increase Sa at an overlapping rate of 2500 pulses/cm2 is 83.45 smaller than in the non-treated specimen without LSP. A bigger resistance of the LSP specimens against pitting corrosion depends on the transformation of the aluminium oxide due to the surface ablation during plasma generation. According to literature [2], it a temperature of over 350 C to generate a transformation of an amorphous aluminium oxide (Al2O3) with a density of 3.4 g/cm3 into -Al2O3 (corundum) with a density of 3.98 g/cm3 that exhibits a bigger resistance against a pitting corrosion attack. 4. Conclusion The results of the research performed yield the following conclusions:

The correlation coefficient points to a strong preferential trend of the non-treated specimens surface texture whereas the LSP-treated specimens show no evident surface texture orientation.

Potentiodynamic polarization tests confirmed a growth of the pitting potential with a higher pulse density: a 62 mV higher pitting potential at 900 pulses/cm2 and a 120 mV higher pitting potential at 2500 pulses/cm2 compared to the non-treated specimen.

Surface analysis after the electrochemical corrosion tests with a 3D surface metrology method revealed most corrosion damage in the specimen in its basic condition whereas in the specimens treated with a higher pulse density the number of pits is reduced.

A comparison between the roughness values Sa and Sq prior to and after the corrosion testing confirms big differences with regard to laser treatment conditions. In the non-treated specimen, the Sa values increased by 111 % while the Sq values increased as high as by 131.4 % compared to the specimen in its basic condition after the metallographic cut. In the laser-treated specimens, on the other hand, the increase in roughness was minimal and even reduces as the pulse overlapping rate increases. The difference in roughness increase Sa between the specimen treated with 2500 pulses/cm2 and the non-treated specimen is a ratio of 83.45.

5. References [1] Ezuber H. et. al.: Materials Design, 2007, doi:10.1016/j.matdes.2007.01.021. [2] ASM Metals Handbooks, Volume 13: Corrosion, ASM INTERNATIONAL, 2004, USA. [3] Grum, J., Trdan, U., Hill, M.R.: Materials Science Forum Vol. 589, 2008, pp 379-384. [4] Sano, Y., et. al. (2007): Proc. of ICONE 15, Agoya, Japan. [5] Jiang X. et. al: Proc. of the Royal Society, Vol. 463, No. 2085, 2007, pp. 2071-2099. [6] Danzl R., Helmli F., Scherer S.: Proc. 11th Int. Conf. on Metrology and Properties of

Engineering Surfaces, 2007, 41-46. [7] ISO 25178-6: Geometrical product specifications (GPS) -- Surface texture: Areal -- Part 6:

Classification of methods for measuring surface texture, Draft. [8] IFM Manual, IFM G4 3.2 Demo EN 29.04.2009, Alicona GmbH. [9] ASTM Standard G1-03, ASTM Int., 2003, West Conshohocken, USA.