- Email: [email protected]
Effect of high-phosphorus electroless nickel coating on fatigue life of Al–Cu–Mg–Fe–Ni alloy
Scripta Materialia 57 (2007) 783–786 www.elsevier.com/locate/scriptamat
Effect of high-phosphorus electroless nickel coating on fatigue life of Al–Cu–Mg–Fe–Ni alloy B. Lonyuk,* I. Apachitei and J. Duszczyk Department of Materials Science and Engineering, Delft University of Technology, Mekelweg 2, Delft, 2628 CD, The Netherlands Received 22 March 2007; revised 26 April 2007; accepted 8 May 2007 Available online 25 July 2007
The fatigue behaviour of Al–Cu–Mg–Fe–Ni alloy was evaluated in four different conditions: uncoated, after second zincating pre-treatment, coated with high phosphorus electroless nickel layer and coated with NiP and heat treated for hydrogen release. The results of the fatigue test indicated an increased fatigue life of the coated aluminium alloy up to 150%. This improvement was associated with the higher strength of the coating as compared to the substrate and with the development of compressive residual stresses in the coating during deposition. 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Aluminium alloy; Coating; Electroless nickel; Fatigue; Fracture
Electroless nickel plating is an effective method to increase the corrosion and wear resistance of structural materials such as steel and aluminium alloys [1,2]. One of the major parameters that influence the properties of the electroless nickel–phosphorus (NiP) coating is phosphorus content. For instance, NiP coatings with high phosphorus content (>8 wt.% P) have improved corrosion resistance and compressive stresses as compare to low- and medium-phosphorus containing coatings [2]. Very often NiP coated materials undergo cycling loading, which can induce fatigue fracture. The effect of the high-phosphorus electroless nickel coatings on fatigue behaviour of steel substrates has been investigated by several authors [3–6]. It was found that the ratio substrate/coating strength, coating thickness, coating internal residual stresses as well as the heat treatments applied for hydrogen release or coating hardening are among the factors which affect the fatigue properties of the NiP coated steel substrates. For example, Puchi et al. [3] have reported an increase in the fatigue life of two low strength (tensile strengths in the range of 220– 440 MPa) carbon steels (i.e. AISI 1010 and 1045) coated with about 20–22 lm electroless Ni–10 wt.% P coating. The increase in the fatigue life was higher in the case of AISI 1010 as compared to the AISI 1045. However, * Corresponding author. Tel.: +31 15 2782224; fax: +31 15 2786730; e-mail: [email protected]
deposition of high-phosphorus electroless NiP coatings on high strength steels proved to be detrimental to their fatigue properties. Thus, Wu et al. [4] reported a reduction in the fatigue limit of about 39% for the quenched and tempered 30CrMoA steel coated with a 10 wt.% P electroless nickel coating. Berrı´os et al. [5] have studied the effect of NiP layer thickness on fatigue properties of the AISI 1045 plain carbon steel. It was found that the deposition of a 7 lm NiP layer did not affect the fatigue life of the coated steel whereas with the increase of layer thickness up to 37 lm the fatigue life decreases. On the other hand, Garce´s and co-workers [6] reported that when 12–14 wt.% P is deposited on a quenched and tempered AISI 4340 (CrNiMo) steel, the fatigue life can decrease by almost 92% when the coating/substrate system is heat treated using a two steps heat treatment, i.e. 1 h at 200 C followed by 1 h at 400 C. In a recent work, Puchi-Cabrera et al. [7] showed that approximately 18 wt.% P electroless nickel coating can improve significantly the fatigue and the corrosion-fatigue performance of 7075-T6 aluminium alloy. According to authors the higher mechanical properties of the NiP coating in comparison with the aluminium substrate and its very good adhesion contributed to better fatigue performance of the coated system. A few studies on the medium-phosphorus deposits (5–8 wt.% P) indicated that NiP coating can either increase [8] or decrease [9] the fatigue strength of aluminium alloys. There are no reported studies in the open literature with respect to fatigue behaviour of electroless NiP deposits on
1359-6462/$ - see front matter 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2007.05.015
784
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
2618 aluminium substrates. In the present investigation the fatigue behaviour of the electroless high-phosphorus nickel deposit on aluminium 2618-T61 substrate has been studied. The substrate used in the present study was a wrought aluminium alloy 2618 supplied as extruded bars with a diameter of 75 mm. The chemical composition of this alloy is (in wt.%) 2.35 Cu, 1.58 Mg, 1.07 Fe, 1.07 Ni, 0.17 Si, 0.052 Zn, 0.012 Cr, 0.075 Ti and Al balance. The material was further extruded to 22 mm diameter and heat treated to T61 temper condition (solutioning at 530 C, water quenching and artificial aging at 200 C for 20 h). To be used for axial fatigue testing the cylindrical specimens with a gauge diameter of 8 mm, a gauge length of 10 mm and a 50 mm fillet radius were subsequently machined and then polished to a surface roughness of Ra = 0.2 lm. The deposition of the electroless NiP coatings was carried out using a proprietary electroless nickel solution for high phosphorus content. The deposition process was conducted at 88 ± 2 C and solution pH of 4.8 ± 0.1. Prior to electroless deposition all specimens underwent a double zincating pre-treatment that included 5 min degreasing at 60 C, 2 min etching, 1 min first desmutting, 30 s first zincating, 30 s second desmutting, and 15 s second zincating. The coating thickness was measured by optical microscopy and was found to be 17 ± 1 lm. A part of the coated specimens was subsequently heat-treated at 190 C for 1.5 h to release hydrogen entrapped in the coating during electroless nickel deposition. After the heat treatment, the specimens were slowly cooled down in the furnace to minimize induced internal stresses. The chemical composition of electroless NiP coating was determined by X-ray fluorescence (XRF) using a Philips PW1480 equipment. The phase composition of electroless NiP coatings has been studied by X-ray diffraction (XRD) technique. The measurements were performed on a Bruker-AXS type D8 Advance series 2 diffractometer, equipped with diffracted beam graphite monochromator, using Cu Ka radiation. The diffractometer scans were performed in h–2h geometry with a range of 15–1202h, a stepsize of 0.12h and a counting time of 2 s per step. The fatigue tests were performed using a 100 kN servohydraulic uniaxial Schenk testing machine under the load control condition and with a test frequency of 25 Hz in ambient air. Fully reversible push–pull (stress ratio, R = 1) sinusoidal load cycles were used. The tests were carried out according to ASTM E4689-90. The specimens were cycled at constant amplitude until failure or until at least 107 load cycles were reached. The fatigue mechanisms were studied by examining the fracture surfaces with scanning electron microscopy (SEM). The analysis was conducted on a JEOL JSM6400F microscope. The chemical and phase composition of the electroless NiP coating has been studied using X-ray analyses. The X-ray fluorescence analysis indicates that the average phosphorus content was 13.2 wt.%. The coating has a high phosphorus content and therefore its asdeposited structure is expected to be amorphous. The
structure of the NiP coating was investigated by X-ray diffraction analysis (Fig. 1). A large diffraction peak at 2h = 44.5 followed by a second one at about 2h = 81 appears to be very similar to that reported for amorphous NiP coatings [10]. After annealing at 190 C for 1 h the X-ray diffraction structure of NiP coating remained unchanged. The hardness of the coating (both in as-deposited and annealed conditions), measured with a Vickers diamond indenter under 100 g load, was about 550 HV0.1 [10]. This is more than three times higher than that of substrate (160 HV0.1). The maximum alternating stress as a function of number of cycles to failure (S–N curve) for the uncoated substrate, the substrate after double zincating pre-treatment, the substrate coated by electroless NiP in asdeposited and in heat-treated conditions is presented in Figure 2. From the Figure 2 it is followed that the double zincating pre-treatment decreases the fatigue life of the aluminium substrate. The magnitude of the reduction increases with the alternating stress decrease. For instance, at an alternating stress of 300 MPa the effect of the pre-treatment is almost negligible, whereas at 200 MPa the reduction in fatigue life reaches 75%. Moreover, it can be observed from the Figure 2 that high-phosphorus electroless nickel deposit onto 2618T61 substrate gives rise to the increase of the fatigue life of the substrate in relation to the uncoated aluminium alloy. The magnitude of the improvement depends on the applied alternating stress. Thus, at alternating stress >300 MPa the curves tend to converge. At alternating stress of 250 MPa the increase in fatigue life reaches about 100%, whereas at 200 MPa it reaches approximately 150%. At alternating stress <200 MPa the magnitude of the increase of fatigue life tends to decrease. The hydrogen release heat treatment at 190 C for 1.5 h did not change the fatigue behaviour of the electroless NiP coated samples, indicating improvement over the uncoated alloy. Figures 3 and 4 represent the SEM micrographs of the fracture surface of the uncoated samples. Subsurface fatigue crack initiation sites, located at different depths ranging from near zero up to 4 mm were detected. Figure 3 shows the fracture surface of the sample tested at 250 MPa and fractured after 105 loading cycles. The fracture process started close to the specimen surface and was dominated by the propagation of a single crack. The radial lines on the fracture surface indicate the origin of the fatigue crack, pointed out by the arrow in the
Figure 1. X-ray diffraction patterns of the substrate and electroless NiP deposit in as-deposited state and after heat treatment at 190 C for 1.5 h.
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
785
Figure 4. SEM micrograph of the fracture surface of uncoated aluminium substrate fractured at 170 MPa after 6 · 106 loading cycles: area A shows the location of crack initiation site; arrows indicate the direction of the crack propagation.
In general, in the absence of the defects, a fatigue crack is initiated at the specimen surface and the extrusion–intrusion mechanism explains this behaviour [11]. However, a relatively large pre-existing defects in the specimen may introduce higher stress concentration than the surface flaws, especially at lower cyclic stress since the extrusion–intrusion mechanism becomes less active. Umezawa and Nagai [12] reviewed several works, which describe occurrence of the subsurface (internal) fatigue crack initiation. Subsurface crack initiation in high-cycle fatigue is usually associated with the nonmetallic inclusions, pores or grain boundaries. The weak grain boundaries in 2618 alloy may be responsible for the shift in the crack initiation site from the surface to its interior, as observed in the present work for uncoated samples. Figure 5 illustrates an example of the fatigue crack initiation site typically observed in fractured double zincating pre-treated samples. A fatigue crack initiates from an etch pit in a form of V-shape flaw of about 10 lm deep. Figure 6 presents a crack initiation site on the fracture surface typically observed in the samples coated with electroless NiP in as-deposited state and fractured at 300 MPa. The fracture process was dominated by the single crack propagation of which origin can be well defined by the convergence of the fracture lines. It can be seen that the fatigue crack is associated with a defect from the pre-treatment filled with NiP deposit. A closer analysis of the deposit allows to determine the fatigue striations within it, which indicates that the fatigue crack developed in the coating and was subsequently transferred to the substrate. At alternating stress <250 MPa the crack nucleation site in the coated NiP samples shifts to the specimen interior. Figure 7 depicts the general fracture surface of the samples fractured at 200 MPa. As for the uncoated specimens, it can be observed that the fracture process originates into the specimen interior and propagates towards the specimen outer surface. The welldefined shear leaps, indicating final fracture, were observed at the specimen edges. A more detailed observation at the specimen core revealed a highly faceted area, which allows to identify the fracture mechanism as an intergranular fracture. The results, shown in Figure 2, indicate that the fatigue performance of the aluminium alloy under investigation is significantly affected by the surface treatments
picture. A number of crystallographic facet-like structures formed at the specimen surface was observed at the crack initiation site. A grain boundary corner could be identified, which define the grain boundaries as crack initiation sites. At alternating stress <180 MPa the fatigue crack initiation site shifts from the near specimen surface to its interior. Figure 4 depicts the fracture surface of uncoated sample tested at 170 MPa and fractured after 6 · 106 loading cycles. It is observed that the fracture process occurs as a consequence of propagation of the crack nucleated at the specimen centre. This area is highly faceted and the grain boundary corners are well distinguished.
Figure 5. SEM micrograph of the fracture surface of pre-treated sample fractured at 200 MPa after 2 · 105 loading cycles. The arrows in the figure indicate the direction of the crack propagation.
Figure 2. Number of cycles prior to failure as a function of stress amplitude.
Figure 3. SEM micrograph of the fracture surface of uncoated aluminium substrate fractured at 250 MPa after 105 loading cycles: area A shows the location of crack initiation site; arrows indicate the direction of the crack propagation.
786
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
Figure 6. SEM micrograph showing example of the fracture surface of coated with electroless NiP deposit specimen fractured at 300 MPa after 2 · 104 loading cycles. The arrows in the figures indicate the direction of the crack propagation.
Figure 7. SEM micrograph of the fracture surface of coated with electroless NiP deposit specimen fractured at 200 MPa after 2 · 106 loading cycles: area A shows the location of crack initiation site; the arrows in the figure indicate the direction of the crack propagation.
applied. The low fatigue strength of the specimens after double zincating pre-treatment is clearly associated with the pits formation on the aluminium substrate during the etching process. Those pits act as stress risers and determine an early fatigue crack initiation. The beneficial effect of the electroless NiP coating on the fatigue behaviour of the aluminium alloy is associated with the retardation of the fatigue crack initiation. At alternating stress >250 MPa fatigue cracks are nucleated at the outer surface of the coating and propagate through the coating towards the substrate. The higher strength of the coating as compared to the substrate material, measured as microhardness, is probably one of the factors responsible for the better fatigue resistance of the coated alloy. Another factor, which may affect the fatigue performance, is a compressive state of the internal residual stresses in the coating. The residual stresses are developed in electroless nickel coating presumably due to the difference in the thermal expansion coefficient between the deposit and the substrate during cooling from the deposition temperature (e.g. 88 C) to the ambient temperature. According to Parker and Shah [13] for as-deposited high phosphorus electroless Ni coatings on aluminium substrates the residual internal stresses are expected to be compressive. At alternating stress <250 MPa due to the weakened grain boundaries of the material under investigation the subsurface crack nucleation mechanism becomes favourable. Although, these stresses are higher than that needed for the nucleation of the fatigue crack in uncoated samples. Based on the SEM analysis it can also be suggested that the refining of the grain size of the aluminium alloy and an improvement of the pre-treatment process in order to avoid severe etching could be added to the bene-
ficial effect of the NiP deposit on the fatigue performance. The fatigue performance of a wrought aluminium alloy 2618-T61 coated with a high-phosphorus (13.2 wt.% P) electroless nickel deposit was evaluated. In addition, the effect of the double zincating pre-treatment and a heat treatment for hydrogen release was also studied. The double zincating pre-treatment has been shown to reduce the fatigue life of the aluminium substrate. This is associated with the pit formation on the aluminium substrate during etching process. The pits act as stress risers leading to an early fatigue crack initiation. The subsequent electroless nickel coating with 13.2 wt.% P gives rise to an increase in the fatigue life of the aluminium substrate up to about 150% at a stress amplitude of 200 MPa. The improvement in the fatigue behaviour of the coated aluminium alloy can be associated with the higher strength of the coating per se as compared to the substrate and with the compressive residual stresses induced to the coating during deposition process. The microscopic observations of the fracture surface of the coated samples revealed that at the alternating stress >250 MPa the crack initiates at the coating surface and is associated with the etch pits formed during pre-treatment and filled with NiP deposit during coating. At alternating stress <250 MPa the fatigue crack initiates in the specimen interior by intergranular mechanism. Such a shift in the location of the crack nucleation site from the specimen surface to the specimen interior is observed also for uncoated samples at an alternating stress of 180 MPa. This shift is related to the weak grain boundaries of the alloy under investigation. The heat treatment for hydrogen release did not affect the fatigue performance of the coated material. [1] R. Weil, K. Parker, in: G.O. Mallory, J.B. Hajdu (Eds.), Electroless Plating: Fundamentals and Applications, American Electroplaters and Surface Finishers Society, Orlando, Florida, USA, 1990, p. 111. [2] W. Riedel, Electroless Nickel Plating, ASM International, Metals Park, OH, 1991, p. 320. [3] E.S. Puchi, M.H. Staia, H. Hintermann, A. Pertuz, J. Chitty, Thin Solid Films 290–291 (1996) 370. [4] Y. Wu, Y. Zhang, M. Yao, Plat. Surf. Finish. (1995) 83. [5] J.A. Berrios, M.H. Staia, E.C. Hernandez, H. Hintermann, E.S. Puchi, Surf. Coat. Technol. 108–109 (1998) 466. [6] Y. Garces, H. Sanchez, J. Berrios, A. Pertuz, J. Chitty, H. Hintermann, E.S. Puchi, Thin Solid Films 355–356 (1999) 487. [7] E.S. Puchi-Cabrera, C. Villalobos-Gutierrez, I. Irausquin, J. La Barbera-Sosa, G. Mesmacque, Int. J. Fatigue doi:10.1016/j.iifatigue.2005.12.005. [8] G.V. Karpenko, V.I. Pokhmurskii, V.B. Dalisov, S.I. Rusin, V.S. Zamikovskii, Ya.P. Brodyak, Prot. Met. 8 (1973) 333. [9] Z. Hongzhe, H. Hui, in: H. Kitagawa, T. Tanaka (Eds.), Fatigue 90, Proceedings of the Fourth International Conference on Fatigue and Fracture, MCEP Birmingham, UK, 1990, p. 985. [10] I. Apachitei, F.D. Tichelaar, J. Duszczyk, L. Katgerman, Surf. Coat. Technol. 149 (2002) 263. [11] G. Dieter, D. Bacon, Mechanical Metallurgy, McGrawHill Book Company, New York, USA, 1988. [12] O. Umezawa, K. Nagai, ISIJ Int. 37 (1997) 1170. [13] K. Parker, H. Shah, Plating 58 (1971) 230.
Effect of high-phosphorus electroless nickel coating on fatigue life of Al–Cu–Mg–Fe–Ni alloy B. Lonyuk,* I. Apachitei and J. Duszczyk Department of Materials Science and Engineering, Delft University of Technology, Mekelweg 2, Delft, 2628 CD, The Netherlands Received 22 March 2007; revised 26 April 2007; accepted 8 May 2007 Available online 25 July 2007
The fatigue behaviour of Al–Cu–Mg–Fe–Ni alloy was evaluated in four different conditions: uncoated, after second zincating pre-treatment, coated with high phosphorus electroless nickel layer and coated with NiP and heat treated for hydrogen release. The results of the fatigue test indicated an increased fatigue life of the coated aluminium alloy up to 150%. This improvement was associated with the higher strength of the coating as compared to the substrate and with the development of compressive residual stresses in the coating during deposition. 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Aluminium alloy; Coating; Electroless nickel; Fatigue; Fracture
Electroless nickel plating is an effective method to increase the corrosion and wear resistance of structural materials such as steel and aluminium alloys [1,2]. One of the major parameters that influence the properties of the electroless nickel–phosphorus (NiP) coating is phosphorus content. For instance, NiP coatings with high phosphorus content (>8 wt.% P) have improved corrosion resistance and compressive stresses as compare to low- and medium-phosphorus containing coatings [2]. Very often NiP coated materials undergo cycling loading, which can induce fatigue fracture. The effect of the high-phosphorus electroless nickel coatings on fatigue behaviour of steel substrates has been investigated by several authors [3–6]. It was found that the ratio substrate/coating strength, coating thickness, coating internal residual stresses as well as the heat treatments applied for hydrogen release or coating hardening are among the factors which affect the fatigue properties of the NiP coated steel substrates. For example, Puchi et al. [3] have reported an increase in the fatigue life of two low strength (tensile strengths in the range of 220– 440 MPa) carbon steels (i.e. AISI 1010 and 1045) coated with about 20–22 lm electroless Ni–10 wt.% P coating. The increase in the fatigue life was higher in the case of AISI 1010 as compared to the AISI 1045. However, * Corresponding author. Tel.: +31 15 2782224; fax: +31 15 2786730; e-mail: [email protected]
deposition of high-phosphorus electroless NiP coatings on high strength steels proved to be detrimental to their fatigue properties. Thus, Wu et al. [4] reported a reduction in the fatigue limit of about 39% for the quenched and tempered 30CrMoA steel coated with a 10 wt.% P electroless nickel coating. Berrı´os et al. [5] have studied the effect of NiP layer thickness on fatigue properties of the AISI 1045 plain carbon steel. It was found that the deposition of a 7 lm NiP layer did not affect the fatigue life of the coated steel whereas with the increase of layer thickness up to 37 lm the fatigue life decreases. On the other hand, Garce´s and co-workers [6] reported that when 12–14 wt.% P is deposited on a quenched and tempered AISI 4340 (CrNiMo) steel, the fatigue life can decrease by almost 92% when the coating/substrate system is heat treated using a two steps heat treatment, i.e. 1 h at 200 C followed by 1 h at 400 C. In a recent work, Puchi-Cabrera et al. [7] showed that approximately 18 wt.% P electroless nickel coating can improve significantly the fatigue and the corrosion-fatigue performance of 7075-T6 aluminium alloy. According to authors the higher mechanical properties of the NiP coating in comparison with the aluminium substrate and its very good adhesion contributed to better fatigue performance of the coated system. A few studies on the medium-phosphorus deposits (5–8 wt.% P) indicated that NiP coating can either increase [8] or decrease [9] the fatigue strength of aluminium alloys. There are no reported studies in the open literature with respect to fatigue behaviour of electroless NiP deposits on
1359-6462/$ - see front matter 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2007.05.015
784
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
2618 aluminium substrates. In the present investigation the fatigue behaviour of the electroless high-phosphorus nickel deposit on aluminium 2618-T61 substrate has been studied. The substrate used in the present study was a wrought aluminium alloy 2618 supplied as extruded bars with a diameter of 75 mm. The chemical composition of this alloy is (in wt.%) 2.35 Cu, 1.58 Mg, 1.07 Fe, 1.07 Ni, 0.17 Si, 0.052 Zn, 0.012 Cr, 0.075 Ti and Al balance. The material was further extruded to 22 mm diameter and heat treated to T61 temper condition (solutioning at 530 C, water quenching and artificial aging at 200 C for 20 h). To be used for axial fatigue testing the cylindrical specimens with a gauge diameter of 8 mm, a gauge length of 10 mm and a 50 mm fillet radius were subsequently machined and then polished to a surface roughness of Ra = 0.2 lm. The deposition of the electroless NiP coatings was carried out using a proprietary electroless nickel solution for high phosphorus content. The deposition process was conducted at 88 ± 2 C and solution pH of 4.8 ± 0.1. Prior to electroless deposition all specimens underwent a double zincating pre-treatment that included 5 min degreasing at 60 C, 2 min etching, 1 min first desmutting, 30 s first zincating, 30 s second desmutting, and 15 s second zincating. The coating thickness was measured by optical microscopy and was found to be 17 ± 1 lm. A part of the coated specimens was subsequently heat-treated at 190 C for 1.5 h to release hydrogen entrapped in the coating during electroless nickel deposition. After the heat treatment, the specimens were slowly cooled down in the furnace to minimize induced internal stresses. The chemical composition of electroless NiP coating was determined by X-ray fluorescence (XRF) using a Philips PW1480 equipment. The phase composition of electroless NiP coatings has been studied by X-ray diffraction (XRD) technique. The measurements were performed on a Bruker-AXS type D8 Advance series 2 diffractometer, equipped with diffracted beam graphite monochromator, using Cu Ka radiation. The diffractometer scans were performed in h–2h geometry with a range of 15–1202h, a stepsize of 0.12h and a counting time of 2 s per step. The fatigue tests were performed using a 100 kN servohydraulic uniaxial Schenk testing machine under the load control condition and with a test frequency of 25 Hz in ambient air. Fully reversible push–pull (stress ratio, R = 1) sinusoidal load cycles were used. The tests were carried out according to ASTM E4689-90. The specimens were cycled at constant amplitude until failure or until at least 107 load cycles were reached. The fatigue mechanisms were studied by examining the fracture surfaces with scanning electron microscopy (SEM). The analysis was conducted on a JEOL JSM6400F microscope. The chemical and phase composition of the electroless NiP coating has been studied using X-ray analyses. The X-ray fluorescence analysis indicates that the average phosphorus content was 13.2 wt.%. The coating has a high phosphorus content and therefore its asdeposited structure is expected to be amorphous. The
structure of the NiP coating was investigated by X-ray diffraction analysis (Fig. 1). A large diffraction peak at 2h = 44.5 followed by a second one at about 2h = 81 appears to be very similar to that reported for amorphous NiP coatings [10]. After annealing at 190 C for 1 h the X-ray diffraction structure of NiP coating remained unchanged. The hardness of the coating (both in as-deposited and annealed conditions), measured with a Vickers diamond indenter under 100 g load, was about 550 HV0.1 [10]. This is more than three times higher than that of substrate (160 HV0.1). The maximum alternating stress as a function of number of cycles to failure (S–N curve) for the uncoated substrate, the substrate after double zincating pre-treatment, the substrate coated by electroless NiP in asdeposited and in heat-treated conditions is presented in Figure 2. From the Figure 2 it is followed that the double zincating pre-treatment decreases the fatigue life of the aluminium substrate. The magnitude of the reduction increases with the alternating stress decrease. For instance, at an alternating stress of 300 MPa the effect of the pre-treatment is almost negligible, whereas at 200 MPa the reduction in fatigue life reaches 75%. Moreover, it can be observed from the Figure 2 that high-phosphorus electroless nickel deposit onto 2618T61 substrate gives rise to the increase of the fatigue life of the substrate in relation to the uncoated aluminium alloy. The magnitude of the improvement depends on the applied alternating stress. Thus, at alternating stress >300 MPa the curves tend to converge. At alternating stress of 250 MPa the increase in fatigue life reaches about 100%, whereas at 200 MPa it reaches approximately 150%. At alternating stress <200 MPa the magnitude of the increase of fatigue life tends to decrease. The hydrogen release heat treatment at 190 C for 1.5 h did not change the fatigue behaviour of the electroless NiP coated samples, indicating improvement over the uncoated alloy. Figures 3 and 4 represent the SEM micrographs of the fracture surface of the uncoated samples. Subsurface fatigue crack initiation sites, located at different depths ranging from near zero up to 4 mm were detected. Figure 3 shows the fracture surface of the sample tested at 250 MPa and fractured after 105 loading cycles. The fracture process started close to the specimen surface and was dominated by the propagation of a single crack. The radial lines on the fracture surface indicate the origin of the fatigue crack, pointed out by the arrow in the
Figure 1. X-ray diffraction patterns of the substrate and electroless NiP deposit in as-deposited state and after heat treatment at 190 C for 1.5 h.
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
785
Figure 4. SEM micrograph of the fracture surface of uncoated aluminium substrate fractured at 170 MPa after 6 · 106 loading cycles: area A shows the location of crack initiation site; arrows indicate the direction of the crack propagation.
In general, in the absence of the defects, a fatigue crack is initiated at the specimen surface and the extrusion–intrusion mechanism explains this behaviour [11]. However, a relatively large pre-existing defects in the specimen may introduce higher stress concentration than the surface flaws, especially at lower cyclic stress since the extrusion–intrusion mechanism becomes less active. Umezawa and Nagai [12] reviewed several works, which describe occurrence of the subsurface (internal) fatigue crack initiation. Subsurface crack initiation in high-cycle fatigue is usually associated with the nonmetallic inclusions, pores or grain boundaries. The weak grain boundaries in 2618 alloy may be responsible for the shift in the crack initiation site from the surface to its interior, as observed in the present work for uncoated samples. Figure 5 illustrates an example of the fatigue crack initiation site typically observed in fractured double zincating pre-treated samples. A fatigue crack initiates from an etch pit in a form of V-shape flaw of about 10 lm deep. Figure 6 presents a crack initiation site on the fracture surface typically observed in the samples coated with electroless NiP in as-deposited state and fractured at 300 MPa. The fracture process was dominated by the single crack propagation of which origin can be well defined by the convergence of the fracture lines. It can be seen that the fatigue crack is associated with a defect from the pre-treatment filled with NiP deposit. A closer analysis of the deposit allows to determine the fatigue striations within it, which indicates that the fatigue crack developed in the coating and was subsequently transferred to the substrate. At alternating stress <250 MPa the crack nucleation site in the coated NiP samples shifts to the specimen interior. Figure 7 depicts the general fracture surface of the samples fractured at 200 MPa. As for the uncoated specimens, it can be observed that the fracture process originates into the specimen interior and propagates towards the specimen outer surface. The welldefined shear leaps, indicating final fracture, were observed at the specimen edges. A more detailed observation at the specimen core revealed a highly faceted area, which allows to identify the fracture mechanism as an intergranular fracture. The results, shown in Figure 2, indicate that the fatigue performance of the aluminium alloy under investigation is significantly affected by the surface treatments
picture. A number of crystallographic facet-like structures formed at the specimen surface was observed at the crack initiation site. A grain boundary corner could be identified, which define the grain boundaries as crack initiation sites. At alternating stress <180 MPa the fatigue crack initiation site shifts from the near specimen surface to its interior. Figure 4 depicts the fracture surface of uncoated sample tested at 170 MPa and fractured after 6 · 106 loading cycles. It is observed that the fracture process occurs as a consequence of propagation of the crack nucleated at the specimen centre. This area is highly faceted and the grain boundary corners are well distinguished.
Figure 5. SEM micrograph of the fracture surface of pre-treated sample fractured at 200 MPa after 2 · 105 loading cycles. The arrows in the figure indicate the direction of the crack propagation.
Figure 2. Number of cycles prior to failure as a function of stress amplitude.
Figure 3. SEM micrograph of the fracture surface of uncoated aluminium substrate fractured at 250 MPa after 105 loading cycles: area A shows the location of crack initiation site; arrows indicate the direction of the crack propagation.
786
B. Lonyuk et al. / Scripta Materialia 57 (2007) 783–786
Figure 6. SEM micrograph showing example of the fracture surface of coated with electroless NiP deposit specimen fractured at 300 MPa after 2 · 104 loading cycles. The arrows in the figures indicate the direction of the crack propagation.
Figure 7. SEM micrograph of the fracture surface of coated with electroless NiP deposit specimen fractured at 200 MPa after 2 · 106 loading cycles: area A shows the location of crack initiation site; the arrows in the figure indicate the direction of the crack propagation.
applied. The low fatigue strength of the specimens after double zincating pre-treatment is clearly associated with the pits formation on the aluminium substrate during the etching process. Those pits act as stress risers and determine an early fatigue crack initiation. The beneficial effect of the electroless NiP coating on the fatigue behaviour of the aluminium alloy is associated with the retardation of the fatigue crack initiation. At alternating stress >250 MPa fatigue cracks are nucleated at the outer surface of the coating and propagate through the coating towards the substrate. The higher strength of the coating as compared to the substrate material, measured as microhardness, is probably one of the factors responsible for the better fatigue resistance of the coated alloy. Another factor, which may affect the fatigue performance, is a compressive state of the internal residual stresses in the coating. The residual stresses are developed in electroless nickel coating presumably due to the difference in the thermal expansion coefficient between the deposit and the substrate during cooling from the deposition temperature (e.g. 88 C) to the ambient temperature. According to Parker and Shah [13] for as-deposited high phosphorus electroless Ni coatings on aluminium substrates the residual internal stresses are expected to be compressive. At alternating stress <250 MPa due to the weakened grain boundaries of the material under investigation the subsurface crack nucleation mechanism becomes favourable. Although, these stresses are higher than that needed for the nucleation of the fatigue crack in uncoated samples. Based on the SEM analysis it can also be suggested that the refining of the grain size of the aluminium alloy and an improvement of the pre-treatment process in order to avoid severe etching could be added to the bene-
ficial effect of the NiP deposit on the fatigue performance. The fatigue performance of a wrought aluminium alloy 2618-T61 coated with a high-phosphorus (13.2 wt.% P) electroless nickel deposit was evaluated. In addition, the effect of the double zincating pre-treatment and a heat treatment for hydrogen release was also studied. The double zincating pre-treatment has been shown to reduce the fatigue life of the aluminium substrate. This is associated with the pit formation on the aluminium substrate during etching process. The pits act as stress risers leading to an early fatigue crack initiation. The subsequent electroless nickel coating with 13.2 wt.% P gives rise to an increase in the fatigue life of the aluminium substrate up to about 150% at a stress amplitude of 200 MPa. The improvement in the fatigue behaviour of the coated aluminium alloy can be associated with the higher strength of the coating per se as compared to the substrate and with the compressive residual stresses induced to the coating during deposition process. The microscopic observations of the fracture surface of the coated samples revealed that at the alternating stress >250 MPa the crack initiates at the coating surface and is associated with the etch pits formed during pre-treatment and filled with NiP deposit during coating. At alternating stress <250 MPa the fatigue crack initiates in the specimen interior by intergranular mechanism. Such a shift in the location of the crack nucleation site from the specimen surface to the specimen interior is observed also for uncoated samples at an alternating stress of 180 MPa. This shift is related to the weak grain boundaries of the alloy under investigation. The heat treatment for hydrogen release did not affect the fatigue performance of the coated material. [1] R. Weil, K. Parker, in: G.O. Mallory, J.B. Hajdu (Eds.), Electroless Plating: Fundamentals and Applications, American Electroplaters and Surface Finishers Society, Orlando, Florida, USA, 1990, p. 111. [2] W. Riedel, Electroless Nickel Plating, ASM International, Metals Park, OH, 1991, p. 320. [3] E.S. Puchi, M.H. Staia, H. Hintermann, A. Pertuz, J. Chitty, Thin Solid Films 290–291 (1996) 370. [4] Y. Wu, Y. Zhang, M. Yao, Plat. Surf. Finish. (1995) 83. [5] J.A. Berrios, M.H. Staia, E.C. Hernandez, H. Hintermann, E.S. Puchi, Surf. Coat. Technol. 108–109 (1998) 466. [6] Y. Garces, H. Sanchez, J. Berrios, A. Pertuz, J. Chitty, H. Hintermann, E.S. Puchi, Thin Solid Films 355–356 (1999) 487. [7] E.S. Puchi-Cabrera, C. Villalobos-Gutierrez, I. Irausquin, J. La Barbera-Sosa, G. Mesmacque, Int. J. Fatigue doi:10.1016/j.iifatigue.2005.12.005. [8] G.V. Karpenko, V.I. Pokhmurskii, V.B. Dalisov, S.I. Rusin, V.S. Zamikovskii, Ya.P. Brodyak, Prot. Met. 8 (1973) 333. [9] Z. Hongzhe, H. Hui, in: H. Kitagawa, T. Tanaka (Eds.), Fatigue 90, Proceedings of the Fourth International Conference on Fatigue and Fracture, MCEP Birmingham, UK, 1990, p. 985. [10] I. Apachitei, F.D. Tichelaar, J. Duszczyk, L. Katgerman, Surf. Coat. Technol. 149 (2002) 263. [11] G. Dieter, D. Bacon, Mechanical Metallurgy, McGrawHill Book Company, New York, USA, 1988. [12] O. Umezawa, K. Nagai, ISIJ Int. 37 (1997) 1170. [13] K. Parker, H. Shah, Plating 58 (1971) 230.