- Email: [email protected]
The microstructures and corrosion properties of polycrystalline copper induced by high-current pulsed electron beam
Applied Surface Science 294 (2014) 9–14
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
The microstructures and corrosion properties of polycrystalline copper induced by high-current pulsed electron beam Zaiqiang Zhang a , Shengzhi Yang a , Peng Lv a , Yan Li a , Xiaotong Wang a , Xiuli Hou a , Qingfeng Guan a,b,∗ a b
College of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, China Key Laboratory of Materials Modification by Laser, Ion and Electron Beams of Ministry of Education, Dalian University of Technology, Dalian 116024, China
a r t i c l e
i n f o
Article history: Received 14 June 2013 Received in revised form 18 December 2013 Accepted 31 December 2013 Available online 8 January 2014 Keywords: High current pulsed electron beam (HCPEB) cp Copper Craters Microstructure Corrosion resistance
a b s t r a c t In order to investigate the corrosion mechanism of pure metal materials containing little impurities, polycrystalline commercial pure (cp) copper was irradiated by high-current pulsed electron beam (HCPEB). The surface microstructures of irradiated samples are characterized by using optical microscopy and transmission electron microscopy (TEM). The corrosion resistance is also investigated by using polarization curves of seawater corrosion and electrode impedance spectroscopy (EIS). The experimental results indicate that the corrosion resistance of cp copper irradiated by 10 pulses is remarkably improved comparing with the original sample. TEM observations suggest that large amount of supersaturated vacancy defects are produced when the material surface is subjected to the HCPEB irradiation. Furthermore, the agglomerations of the vacancy defects cause the formation of the vacancy cluster defects, such as vacancy dislocation loops, the stacking fault tetrahedra (SFTs) and voids. It is suggested that the structural defects on the irradiated surface have some relationships with the corrosion resistance’s improvement of the material. © 2014 Elsevier B.V. All rights reserved.
1. Introduction High-current pulsed electron beam (HCPEB) has drawn widely attention for its great application potential and advantages of being simple, reliable, effective, and low energy consumption [1–9]. Under the action of HCPEB, a high power density (106 –3 × 107 W/cm2 ) is deposited only in a very thin layer (up to several micrometers) within a short time (a few microseconds), and such pulsed electron irradiation produces extremely fast heating and cooling of the surface and induces thermal stresses. As a result, abundant metastable surface microstructures or phase structures, such as supersaturated solid solution [1], ultra-fine grain [3], abundant defect structures, and nanostructures [5,10] are formed within the irradiated surface layer. In general, craters are inevitably formed on the irradiated surface when metals and alloys are irradiated by intense-pulsed energetic beams, which are believed to deteriorate corrosion resistance of irradiated materials [1,2]. However, many researchers reported that evident improvement of corrosion resistance had been achieved due to intense-pulsed energetic beams irradiation. Zou et al. believed that the improvement of corrosion resistance of irradiated materials was mainly attributed to the surface
purification effect [11,12]. Based on this mechanism, special selective surface contained impurities firstly melted or sublimated in the process of crater eruption. The mixing in the eruption process and the subsequent solute trapping effect during fast solidification process resulted in the homogenization of elements in the melted layer. Both crater eruption and composition homogenization contribute to the selective surface purification effect. As a result, the corrosion resistance of materials is significantly improved. Whereas, the influence of the microstructure evolutions induced by HCPEB irradiation on corrosion resistance of the irradiated materials was less involved in the mechanism presented by Zou et al. As we know, the surface properties of irradiated materials are determined by the final structure states. However, limited amount of works has concentrated on detailed microstructural characterization of HCPEB irradiated surface. In this paper, we applied HCPEB technique to irradiate cp copper which almost contained no impurities. We present a detailed investigation on the microstructures of cp copper samples irradiated by HCPEB. The relationships between corrosion resistance and microstructures, especially defects structure, were explored. The detailed mechanism for the improvement in corrosion resistance of cp copper irradiated by HCPEB was investigated. 2. Experimental
∗ Corresponding author at: College of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, China. Tel.: +8651188790083. E-mail address: [email protected] (Q. Guan). 0169-4332/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2013.12.178
Our experiments were conducted on HCPEB equipment (Nadezhda-2 type). It produced a pulsed (0.5–3 s), low energy
10
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
(10–40 keV), high current (102 –103 A/cm2 ), and high-aperture (∼30 cm2 ) electron beam. Vacuum conditions during electron beam irradiation also can be supplied. The commercial pure (about 99.9%) copper was selected as the target material. Specimens were machined with a size of 10 mm in length, 10 mm in width, and 6 mm in height. All the samples were grounded with sandpapers and polished with diamond paste. Prior to HCPEB treatment, the specimens were ultrasonically cleaned in acetone. The polished surfaces of samples were irradiated at room temperature with 5 and 10 pulses, respectively. The HCPEB bombardments were carried out under the following conditions: the electron energy 27 keV, the current pulse duration 1.5 s, the energy density about 4 J/cm2 , and the vacuum 10−5 Torr. In order to obtain more information about microstructure, especially defects structure, the thin foils for transmission electron microscope (TEM) observation were prepared using mechanical pre-thinning, dimpling, and jet electrolytical thinning from the substrate side. The TEM examinations were conducted with JEM2100 transmission electron microscope, which was operated at 200 kV and the foils for TEM observation is about 100–300 nm in depth. The microstructure changes of the surface layer were also characterized by using a LEICA DM-2500 M optical microscope, JSM-5600F scanning electron microscope. Corrosion tests were carried out using the conventional three-electrode cell containing the work electrode, a saturated calomel electrode (SCE) as the reference electrode, and a platinum sheet as the counter electrode. The electrolyte solution was simulated sea water and its chemical components (wt.%) was shown in Table 1. Standard potentiodynamic polarization and electrochemical impedance spectroscopy measurements were conducted after samples exposed to simulated sea water under open circuit potential for 30 min (at room temperature, about 25 ◦ C). Normally, after 30 min immersion in simulated sea water, a fairly stable potential could be achieved, the potentiodynamic polarization was carried out at a scan rate of 0.333 mV/s. The EIS spectra were obtained over the frequency (f) range 10−2 –105 Hz at the open circuit potential with an AC excitation amplitude of
Table 1 Composition of simulated sea water (wt.%). Compound
Concentration (g/L)
Compound
Concentration (g/L)
NaCl MgCl2 Na2 SO4 CaCl2 KCl
24.53 5.20 4.09 1.16 0.695
NaHCO3 KBr H3 BO3 SrCl2 NaF
0.201 0.101 0.027 0.025 0.003
10 mV. The size of effective testing area exposed to simulated sea water was 5 mm × 5 mm with other non-working surface covered with epoxy resin. 3. Results and discussion 3.1. Surface morphology and microstructure Fig. 1 shows the surface morphologies of both 5-pulsed and 10pulsed cp copper samples. For the irradiated samples, the surface becomes to be much rougher than original samples and the craters formed on the surface while their sizes range from several micrometers to tens of micrometers. Usually, there is a black “dimple” on the center of crater, as show in Fig. 1(a) and (b). They are formed by the eruptions of melting pool and remain until the last. As some researchers’ study, they are sensitive sites for the pitting corrosion, which was easily to be attacked by anions [12]. Compared with Fig. 1(a) and (b), one can see that the crater density of 5pulsed sample is obviously higher than that of 10-pulsed sample, indicating that the crater density decreases with the pulse number increase in range of 5–10. This result can be attributed to the polishing effect to the material surface of HCPEB irradiation. The preceding formed craters would be fused or eliminated by subsequent pulses [13]. From previous studies, such morphology is the result of local sub-layer melting and eruption through the solid outer surface [2,14]. Surface impurities, second phase particles, and
Fig. 1. Surface images of cp copper after HCPEB treatment (a), (c) 5 pulses, (b), (d) 10 pulses.
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
Fig. 2. TEM images of dislocation cells within the surface layer of cp copper after HCPEB irradiation (a) 5 pulses, (b) 10 pulses.
Fig. 3. TEM images of vacancy defects after 5(a–c) and 10(d–f) pulses irradiation of HCPEB.
11
12
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
various structural defects are likely to serve as nucleation sites for the eruption events that lead to the formation of craters. Based on this conclusion, some researchers believed that crater eruption events contributed to the selective surface purification effect and improved the corrosion resistance of irradiated materials [12]. Dense deformation bands are visible under SEM observations, as show in Fig. 1(c) and (d). The formation of this structure indicates the inhomogeneous deformations in the subsurface layer induced by the HCPEB irradiation and they also can be regarded as a witness mark of the structural defects introduced by HCPEB irradiation. Fig. 2 shows the accumulation of dislocations after HCPEB irradiation. Deformation-induced dislocations are scattered in some regions of the sample with 5-pulsed irradiation, as show in Fig. 2(a). In contrast, Fig. 2(b) reveals that dislocations have a tendency to develop into well-defined cell walls at 10-pulsed irradiation. These dislocation configurations were more frequently observed in conventionally deformed metals and alloys. Applied stresses produces cross-slip giving rise to tangled dislocations or cell structures which form as a minimum energy configuration when the dislocation density becomes large in response to large deformation or large strains [15], indicating that big stresses has been induced by HCPEB irradiation on the surface of materials. The density of dislocations in the boundaries of dislocation cells is much higher while there are only few, even no dislocations, in the center of dislocation cells. The sizes of these dislocation cells in the 5-pulsed and 10-pulsed samples are about 0.2 m and 1.2 m, respectively, which means the density of dislocations of 5-pulsed samples is little higher than that of 10-pulsed samples. Besides the structure defects mentioned above, very high density of small defects were observed in the irradiated samples with sizes about 2–20 nm.TEM observation along the [1 1 0] direction, as show in Fig. 3(a), indicates that dislocation loops around the grain boundary can be observed clearly (as indicated by the arrows). Fig. 3(b) illuminates that voids are present in the regions with many extinction contours. Especially, stacking fault tetrahedra (SFT) defects [16], a typical vacancy defect in faced central cubic (fcc) metal, with clear triangular images (arrowed in Fig. 3(c)) are also observed. For 10-pulsed samples, these three defects including dislocation loops, voids, and SFT were still observed, as show in Fig. 3(d–f), respectively. Comparing these defects with that in 5-pulsed samples after the analysis of dozens of TEM images of the small defects in various observed regions, there is no evident difference in kind, density, distribution, and size. The formation of a mass of vacancy defect clusters suggests that supersaturating vacancies can be introduced by HCPEB irradiation. It corresponds to Prgrebnjak et al. result detected by a positron annihilation method [17]. They demonstrated that a high density of non-equilibrium vacancies (up to 10−3 ) was formed in the near-surface of pure iron irradiated by HCPEB. The density of non-equilibrium vacancies is much higher than that of the situation at room temperature.
Fig. 4. The polarization curve of cp copper in simulated seawater pre-and posttreatment by HCPEB. Table 2 The corrosion data of the treated and untreated pure copper samples. Samples
Ecorr (mV)
icorr (A cm−2 )
Untreated 5 pulses 10 pulses
−200.2 −209.1 −169.4
0.3514 0.7139 0.1166
samples show a higher corrosion resistance, with the lower icorr and the higher Ecorr . Fig. 6 shows the EIS of cp copper in simulated seawater preand post-treatment by HCPEB, which displays the relationship among impedance magnitude |Z|, phase angle and frequency f. In Fig. 6(a), the impedance magnitude of different frequency can be directly read out. There is no relationship between impedance and frequency in high frequency region but reveals the electrolyte impedance between samples and the reference electrode. At the low frequency limit, the impedance is attributed to the polarization resistance of the sample in the electrolyte [18]. The polarization impedance valued at the place of 0.01 Hz clearly shows that 10pulsed sample has increased by 2–2.5 times comparing to both untreated sample and 5-pulsed sample. In Fig. 6(b), the phase angles of 10-pulsed sample are evidently bigger than initial
3.2. Corrosion resistance The polarization curves, displayed in Fig. 4, were measured in simulated sea water to compare the diversity of corrosion resistance among the initial and irradiated samples. The plots were analyzed by Tafel extrapolation, and the parameters such as corrosion current density (icorr ) and corrosion potential (Ecorr ) were calculated and tabulated in Table 2. Fig. 5 shows the variations of icorr and Ecorr of cp copper. The corrosion current density has increased significantly with the increase of the pulses number up to 5, followed by a decreasing trend. The corrosion current decreases by ∼6 times for the sample irradiated with 10 pulses compared to 5 pulses. The corrosion potential of original and 5-pulsed samples has little difference while it reaches a higher level in 10-pulsed samples. It is clear that the 10-pused
Fig. 5. Variation of corrosion current density (icorr ) and corrosion potential (Ecorr ).
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
13
Fig. 6. EIS of cp copper in simulated seawater pre-and post-treatment by HCPEB (a) impedance and frequency, (b) phase angle and frequency, (c) Nyquist plot.
sample in the low and middle frequency scopes. Log|Z| has a linear relation with log (f) and the phase angle is approximately close to 90◦ in the middle frequency scope which exhibit that the passive layer formed on the HCPEB-modified surface has a predominantly near capacitive behavior [19]. It is demonstrated that the irradiated samples are in a passive state by the time. In addition, Fig. 6(b) also reveals that the near capacitive behavior is observed over a much wider frequency range for the irradiated samples comparing to the initial sample, suggesting that the passive film on the irradiated sample surfaces can maintain its characteristic response over a longer period of time. Fig. 6(c) is the Nyquist plot, one can conclude that the spectra of all the samples exhibit capacitance resistance arcs with big diameter lying in the first quadrant. The plot of untreated sample is restricted in the range Z < 15,000 and Z < 40,000 . Comparatively, the range increases drastically when the sample surface is applied 10 pulses of HCPEB treatment. This indicates that the passive layers on the HCPEB-modified samples have better corrosion resistance in the simulated sea water compared to the untreated sample. Furthermore, the diameter of the semi-circle increases, together with an increase in polarization resistance (Rp ), when the number of HCPEB pulses increases from 5 to 10. 4. Discussion The research of Zou et al. showed that the corrosion resistance of 316 stainless steel samples after 10-pulsed HCPEB irradiation were greatly improved. They believed that the formation of craters and
the associated with MnS inclusions eruptions induced by HCPEB irradiation together with the MnS dissolution in the melt lead to a decrease in the amount of inclusions on the surface of 316 L stainless steel, viz. the so-called “self purification” effect [11,12]. They mainly attributed the improved performances to the elimination of inclusions at the near surface, disappearance of craters, and formation of a homogeneous and smooth protective surface introduced by solidification of the modified melted layer. Our experiment results of HCPEB-irradiated cp copper samples were very similar with theirs in corrosion resistance. However, the inclusions are negligible in cp copper, so the mechanism of improved corrosion resistance should be different. In our opinion, the influence of microstructures induced by HCPEB irradiation on corrosion resistance should be considered to modify this corrosion mechanism. When a cp copper sample is immerged in the simulated sea water, the matrix will be oxidized to form a passive film composed of Cu2 O by contact with dissolved oxygen at the surface region, which can prevent anionic species (F− , SO4 2− , Cl− etc.) passing through the surface oxidation layer and delay the corrosion process. It was assumed that the passive layer in surface region of initial sample was thinner because of the low diffusing capacity of oxygen. Such a thin passive film is vulnerable to be broken by the corrosive anions of solution, which leads to poor corrosion resistance of initial samples. For the irradiated samples, as referred to in the last section, a mass of microstructural defects especially supersaturated vacancy (vacancy-cluster) defects were introduced within the near surface layer due to HCPEB irradiation. These structural defects supply plenty of pathways and positions for the adsorption
14
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
and entrance of dissolved O2− , promoting the formation of thicker and denser passive layer. However, overlarge density of craters (like 5-pulsed samples) will damage the continuity and integrity of oxide layer, which can badly affect the ability of oxide layer to prevent the corrosive anions and bring the corrosion resistance down. For 10-pulsed sample, surface passive film is still compact and thick. On the other hand, low density of craters was obtained on the irradiated surface, meaning the evident reduction of sensitive sites for the pitting corrosion. Therefore, the passive film induced by the 10 pulses of HCPEB treatment offers a good corrosion protection due to high destiny of structural defects and low density of craters.
Chinese Natural Science Foundation (Grant No. U1233111, 50671042), Open project of key Laboratory of Materials Modification by Laser, Ion and Electron Beams (Dalian University of Technology), Ministry of Education (DP1051102).
5. Conclusions
References
(1) HCPEB irradiation significantly changes the microstructures of cp copper. After HCPEB irradiation, the irradiated surface locally melted and then craters were formed in these regions. The crater density of 5-pulsed sample is obviously higher than that of 10-pulses. Besides, very abundant defect structures like dislocations were formed within the near surface irradiated. At the same time, the formation of supersaturated vacancies and three types of vacancy cluster defects (dislocation loops, SFTs, and voids) were also demonstrated after both 5- and 10-pulsed HCPEB irradiation. (2) HCPEB irradiation improves corrosion resistance of cp copper significantly. The corrosion resistance of 10-pulsed samples is better than original samples. For overlarge density of craters can bring the corrosion resistance of 5-pulsed samples down in spite of thicker and denser passive film. (3) The various structural defects introduced by HCPEB irradiation can play some role in the mechanisms of corrosion. These
[1] D.I. Proskurovsky, V.P. Rothtein, G.E. Ozur, et al., Surf. Coat. Technol. 125 (2000) 49. [2] K.M. Zhang, J.X. Zou, T. Grosdidier, et al., Surf. Coat. Technol. 201 (2006) 1393. [3] J.X. Zou, T. Grosdidier, K.M. Zhang, et al., J. Alloys Compd. 434 (2007) 707. [4] J.X. Zou, T. Grosdidier, K.M. Zhang, et al., Appl. Surf. Sci. 255 (2009) 4758. [5] X.D. Zhang, S.Z. Hao, X.N. Lin, et al., Appl. Surf. Sci. 257 (2011) 5899. [6] S.Z. Hao, P.S. Wu, J.X. Zou, et al., Appl. Surf. Sci. 253 (2007) 5349. [7] S.Z. Hao, X.D. Zhang, X.X. Mei, et al., Mater. Lett. 62 (2008) 414. [8] A.F. Shubin, V.P. Rotshtein, D.I. Proskurovskii, Sov. Phys. J. 17 (1974) 942. [9] V.P. Rotshtein, L.S. Bushnev, D.I. Proskurovskii, Sov. Phys. J. 18 (1975) 404. [10] D.M. Follstaedt, W.R. Wampler, Appl. Phys. Lett. 38 (1980) 140. [11] J.X. Zou, K.M. Zhang, C. Dong, et al., Appl. Phys. Lett. 89 (2006) 041913. [12] J.X. Zou, K.M. Zhang, S.Z. Hao, et al., Thin Solid Films 519 (2010) 1404. [13] V.A. Shulov, N.A. Nochovaya, Nucl. Instrum. Methods Phys. Res. 148 (1995) 154. [14] Q.F. Guan, Q. Zhang, C. Dong, ISIJ Int. 48 (2008) 235. [15] N. Hansen, Metall. Mater. Trans. A 32 (2001) 2917. [16] M. Kiritani, Mater. Chem. Phys. 50 (1997) 133. [17] A.D. Pogrebnjak, A.D. Mikhaliov, N.A. Progrebnjak Jr., et al., Phys. Lett. A 241 (1998) 357. [18] A. Norlin, J. Pan, C. Leygraf, Biomol. Eng. 19 (2002) 67. [19] A. Szummer, M. Janik-Czachor, S. Hofmann, Mater. Chem. Phys. 34 (1993) 181.
structural defects supply plenty of pathways and positions for the adsorption and entrance of dissolved O2− , promoting the formation of thicker and denser passive layer, which delay the corrosion process. Acknowledgements
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
The microstructures and corrosion properties of polycrystalline copper induced by high-current pulsed electron beam Zaiqiang Zhang a , Shengzhi Yang a , Peng Lv a , Yan Li a , Xiaotong Wang a , Xiuli Hou a , Qingfeng Guan a,b,∗ a b
College of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, China Key Laboratory of Materials Modification by Laser, Ion and Electron Beams of Ministry of Education, Dalian University of Technology, Dalian 116024, China
a r t i c l e
i n f o
Article history: Received 14 June 2013 Received in revised form 18 December 2013 Accepted 31 December 2013 Available online 8 January 2014 Keywords: High current pulsed electron beam (HCPEB) cp Copper Craters Microstructure Corrosion resistance
a b s t r a c t In order to investigate the corrosion mechanism of pure metal materials containing little impurities, polycrystalline commercial pure (cp) copper was irradiated by high-current pulsed electron beam (HCPEB). The surface microstructures of irradiated samples are characterized by using optical microscopy and transmission electron microscopy (TEM). The corrosion resistance is also investigated by using polarization curves of seawater corrosion and electrode impedance spectroscopy (EIS). The experimental results indicate that the corrosion resistance of cp copper irradiated by 10 pulses is remarkably improved comparing with the original sample. TEM observations suggest that large amount of supersaturated vacancy defects are produced when the material surface is subjected to the HCPEB irradiation. Furthermore, the agglomerations of the vacancy defects cause the formation of the vacancy cluster defects, such as vacancy dislocation loops, the stacking fault tetrahedra (SFTs) and voids. It is suggested that the structural defects on the irradiated surface have some relationships with the corrosion resistance’s improvement of the material. © 2014 Elsevier B.V. All rights reserved.
1. Introduction High-current pulsed electron beam (HCPEB) has drawn widely attention for its great application potential and advantages of being simple, reliable, effective, and low energy consumption [1–9]. Under the action of HCPEB, a high power density (106 –3 × 107 W/cm2 ) is deposited only in a very thin layer (up to several micrometers) within a short time (a few microseconds), and such pulsed electron irradiation produces extremely fast heating and cooling of the surface and induces thermal stresses. As a result, abundant metastable surface microstructures or phase structures, such as supersaturated solid solution [1], ultra-fine grain [3], abundant defect structures, and nanostructures [5,10] are formed within the irradiated surface layer. In general, craters are inevitably formed on the irradiated surface when metals and alloys are irradiated by intense-pulsed energetic beams, which are believed to deteriorate corrosion resistance of irradiated materials [1,2]. However, many researchers reported that evident improvement of corrosion resistance had been achieved due to intense-pulsed energetic beams irradiation. Zou et al. believed that the improvement of corrosion resistance of irradiated materials was mainly attributed to the surface
purification effect [11,12]. Based on this mechanism, special selective surface contained impurities firstly melted or sublimated in the process of crater eruption. The mixing in the eruption process and the subsequent solute trapping effect during fast solidification process resulted in the homogenization of elements in the melted layer. Both crater eruption and composition homogenization contribute to the selective surface purification effect. As a result, the corrosion resistance of materials is significantly improved. Whereas, the influence of the microstructure evolutions induced by HCPEB irradiation on corrosion resistance of the irradiated materials was less involved in the mechanism presented by Zou et al. As we know, the surface properties of irradiated materials are determined by the final structure states. However, limited amount of works has concentrated on detailed microstructural characterization of HCPEB irradiated surface. In this paper, we applied HCPEB technique to irradiate cp copper which almost contained no impurities. We present a detailed investigation on the microstructures of cp copper samples irradiated by HCPEB. The relationships between corrosion resistance and microstructures, especially defects structure, were explored. The detailed mechanism for the improvement in corrosion resistance of cp copper irradiated by HCPEB was investigated. 2. Experimental
∗ Corresponding author at: College of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, China. Tel.: +8651188790083. E-mail address: [email protected] (Q. Guan). 0169-4332/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2013.12.178
Our experiments were conducted on HCPEB equipment (Nadezhda-2 type). It produced a pulsed (0.5–3 s), low energy
10
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
(10–40 keV), high current (102 –103 A/cm2 ), and high-aperture (∼30 cm2 ) electron beam. Vacuum conditions during electron beam irradiation also can be supplied. The commercial pure (about 99.9%) copper was selected as the target material. Specimens were machined with a size of 10 mm in length, 10 mm in width, and 6 mm in height. All the samples were grounded with sandpapers and polished with diamond paste. Prior to HCPEB treatment, the specimens were ultrasonically cleaned in acetone. The polished surfaces of samples were irradiated at room temperature with 5 and 10 pulses, respectively. The HCPEB bombardments were carried out under the following conditions: the electron energy 27 keV, the current pulse duration 1.5 s, the energy density about 4 J/cm2 , and the vacuum 10−5 Torr. In order to obtain more information about microstructure, especially defects structure, the thin foils for transmission electron microscope (TEM) observation were prepared using mechanical pre-thinning, dimpling, and jet electrolytical thinning from the substrate side. The TEM examinations were conducted with JEM2100 transmission electron microscope, which was operated at 200 kV and the foils for TEM observation is about 100–300 nm in depth. The microstructure changes of the surface layer were also characterized by using a LEICA DM-2500 M optical microscope, JSM-5600F scanning electron microscope. Corrosion tests were carried out using the conventional three-electrode cell containing the work electrode, a saturated calomel electrode (SCE) as the reference electrode, and a platinum sheet as the counter electrode. The electrolyte solution was simulated sea water and its chemical components (wt.%) was shown in Table 1. Standard potentiodynamic polarization and electrochemical impedance spectroscopy measurements were conducted after samples exposed to simulated sea water under open circuit potential for 30 min (at room temperature, about 25 ◦ C). Normally, after 30 min immersion in simulated sea water, a fairly stable potential could be achieved, the potentiodynamic polarization was carried out at a scan rate of 0.333 mV/s. The EIS spectra were obtained over the frequency (f) range 10−2 –105 Hz at the open circuit potential with an AC excitation amplitude of
Table 1 Composition of simulated sea water (wt.%). Compound
Concentration (g/L)
Compound
Concentration (g/L)
NaCl MgCl2 Na2 SO4 CaCl2 KCl
24.53 5.20 4.09 1.16 0.695
NaHCO3 KBr H3 BO3 SrCl2 NaF
0.201 0.101 0.027 0.025 0.003
10 mV. The size of effective testing area exposed to simulated sea water was 5 mm × 5 mm with other non-working surface covered with epoxy resin. 3. Results and discussion 3.1. Surface morphology and microstructure Fig. 1 shows the surface morphologies of both 5-pulsed and 10pulsed cp copper samples. For the irradiated samples, the surface becomes to be much rougher than original samples and the craters formed on the surface while their sizes range from several micrometers to tens of micrometers. Usually, there is a black “dimple” on the center of crater, as show in Fig. 1(a) and (b). They are formed by the eruptions of melting pool and remain until the last. As some researchers’ study, they are sensitive sites for the pitting corrosion, which was easily to be attacked by anions [12]. Compared with Fig. 1(a) and (b), one can see that the crater density of 5pulsed sample is obviously higher than that of 10-pulsed sample, indicating that the crater density decreases with the pulse number increase in range of 5–10. This result can be attributed to the polishing effect to the material surface of HCPEB irradiation. The preceding formed craters would be fused or eliminated by subsequent pulses [13]. From previous studies, such morphology is the result of local sub-layer melting and eruption through the solid outer surface [2,14]. Surface impurities, second phase particles, and
Fig. 1. Surface images of cp copper after HCPEB treatment (a), (c) 5 pulses, (b), (d) 10 pulses.
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
Fig. 2. TEM images of dislocation cells within the surface layer of cp copper after HCPEB irradiation (a) 5 pulses, (b) 10 pulses.
Fig. 3. TEM images of vacancy defects after 5(a–c) and 10(d–f) pulses irradiation of HCPEB.
11
12
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
various structural defects are likely to serve as nucleation sites for the eruption events that lead to the formation of craters. Based on this conclusion, some researchers believed that crater eruption events contributed to the selective surface purification effect and improved the corrosion resistance of irradiated materials [12]. Dense deformation bands are visible under SEM observations, as show in Fig. 1(c) and (d). The formation of this structure indicates the inhomogeneous deformations in the subsurface layer induced by the HCPEB irradiation and they also can be regarded as a witness mark of the structural defects introduced by HCPEB irradiation. Fig. 2 shows the accumulation of dislocations after HCPEB irradiation. Deformation-induced dislocations are scattered in some regions of the sample with 5-pulsed irradiation, as show in Fig. 2(a). In contrast, Fig. 2(b) reveals that dislocations have a tendency to develop into well-defined cell walls at 10-pulsed irradiation. These dislocation configurations were more frequently observed in conventionally deformed metals and alloys. Applied stresses produces cross-slip giving rise to tangled dislocations or cell structures which form as a minimum energy configuration when the dislocation density becomes large in response to large deformation or large strains [15], indicating that big stresses has been induced by HCPEB irradiation on the surface of materials. The density of dislocations in the boundaries of dislocation cells is much higher while there are only few, even no dislocations, in the center of dislocation cells. The sizes of these dislocation cells in the 5-pulsed and 10-pulsed samples are about 0.2 m and 1.2 m, respectively, which means the density of dislocations of 5-pulsed samples is little higher than that of 10-pulsed samples. Besides the structure defects mentioned above, very high density of small defects were observed in the irradiated samples with sizes about 2–20 nm.TEM observation along the [1 1 0] direction, as show in Fig. 3(a), indicates that dislocation loops around the grain boundary can be observed clearly (as indicated by the arrows). Fig. 3(b) illuminates that voids are present in the regions with many extinction contours. Especially, stacking fault tetrahedra (SFT) defects [16], a typical vacancy defect in faced central cubic (fcc) metal, with clear triangular images (arrowed in Fig. 3(c)) are also observed. For 10-pulsed samples, these three defects including dislocation loops, voids, and SFT were still observed, as show in Fig. 3(d–f), respectively. Comparing these defects with that in 5-pulsed samples after the analysis of dozens of TEM images of the small defects in various observed regions, there is no evident difference in kind, density, distribution, and size. The formation of a mass of vacancy defect clusters suggests that supersaturating vacancies can be introduced by HCPEB irradiation. It corresponds to Prgrebnjak et al. result detected by a positron annihilation method [17]. They demonstrated that a high density of non-equilibrium vacancies (up to 10−3 ) was formed in the near-surface of pure iron irradiated by HCPEB. The density of non-equilibrium vacancies is much higher than that of the situation at room temperature.
Fig. 4. The polarization curve of cp copper in simulated seawater pre-and posttreatment by HCPEB. Table 2 The corrosion data of the treated and untreated pure copper samples. Samples
Ecorr (mV)
icorr (A cm−2 )
Untreated 5 pulses 10 pulses
−200.2 −209.1 −169.4
0.3514 0.7139 0.1166
samples show a higher corrosion resistance, with the lower icorr and the higher Ecorr . Fig. 6 shows the EIS of cp copper in simulated seawater preand post-treatment by HCPEB, which displays the relationship among impedance magnitude |Z|, phase angle and frequency f. In Fig. 6(a), the impedance magnitude of different frequency can be directly read out. There is no relationship between impedance and frequency in high frequency region but reveals the electrolyte impedance between samples and the reference electrode. At the low frequency limit, the impedance is attributed to the polarization resistance of the sample in the electrolyte [18]. The polarization impedance valued at the place of 0.01 Hz clearly shows that 10pulsed sample has increased by 2–2.5 times comparing to both untreated sample and 5-pulsed sample. In Fig. 6(b), the phase angles of 10-pulsed sample are evidently bigger than initial
3.2. Corrosion resistance The polarization curves, displayed in Fig. 4, were measured in simulated sea water to compare the diversity of corrosion resistance among the initial and irradiated samples. The plots were analyzed by Tafel extrapolation, and the parameters such as corrosion current density (icorr ) and corrosion potential (Ecorr ) were calculated and tabulated in Table 2. Fig. 5 shows the variations of icorr and Ecorr of cp copper. The corrosion current density has increased significantly with the increase of the pulses number up to 5, followed by a decreasing trend. The corrosion current decreases by ∼6 times for the sample irradiated with 10 pulses compared to 5 pulses. The corrosion potential of original and 5-pulsed samples has little difference while it reaches a higher level in 10-pulsed samples. It is clear that the 10-pused
Fig. 5. Variation of corrosion current density (icorr ) and corrosion potential (Ecorr ).
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
13
Fig. 6. EIS of cp copper in simulated seawater pre-and post-treatment by HCPEB (a) impedance and frequency, (b) phase angle and frequency, (c) Nyquist plot.
sample in the low and middle frequency scopes. Log|Z| has a linear relation with log (f) and the phase angle is approximately close to 90◦ in the middle frequency scope which exhibit that the passive layer formed on the HCPEB-modified surface has a predominantly near capacitive behavior [19]. It is demonstrated that the irradiated samples are in a passive state by the time. In addition, Fig. 6(b) also reveals that the near capacitive behavior is observed over a much wider frequency range for the irradiated samples comparing to the initial sample, suggesting that the passive film on the irradiated sample surfaces can maintain its characteristic response over a longer period of time. Fig. 6(c) is the Nyquist plot, one can conclude that the spectra of all the samples exhibit capacitance resistance arcs with big diameter lying in the first quadrant. The plot of untreated sample is restricted in the range Z < 15,000 and Z < 40,000 . Comparatively, the range increases drastically when the sample surface is applied 10 pulses of HCPEB treatment. This indicates that the passive layers on the HCPEB-modified samples have better corrosion resistance in the simulated sea water compared to the untreated sample. Furthermore, the diameter of the semi-circle increases, together with an increase in polarization resistance (Rp ), when the number of HCPEB pulses increases from 5 to 10. 4. Discussion The research of Zou et al. showed that the corrosion resistance of 316 stainless steel samples after 10-pulsed HCPEB irradiation were greatly improved. They believed that the formation of craters and
the associated with MnS inclusions eruptions induced by HCPEB irradiation together with the MnS dissolution in the melt lead to a decrease in the amount of inclusions on the surface of 316 L stainless steel, viz. the so-called “self purification” effect [11,12]. They mainly attributed the improved performances to the elimination of inclusions at the near surface, disappearance of craters, and formation of a homogeneous and smooth protective surface introduced by solidification of the modified melted layer. Our experiment results of HCPEB-irradiated cp copper samples were very similar with theirs in corrosion resistance. However, the inclusions are negligible in cp copper, so the mechanism of improved corrosion resistance should be different. In our opinion, the influence of microstructures induced by HCPEB irradiation on corrosion resistance should be considered to modify this corrosion mechanism. When a cp copper sample is immerged in the simulated sea water, the matrix will be oxidized to form a passive film composed of Cu2 O by contact with dissolved oxygen at the surface region, which can prevent anionic species (F− , SO4 2− , Cl− etc.) passing through the surface oxidation layer and delay the corrosion process. It was assumed that the passive layer in surface region of initial sample was thinner because of the low diffusing capacity of oxygen. Such a thin passive film is vulnerable to be broken by the corrosive anions of solution, which leads to poor corrosion resistance of initial samples. For the irradiated samples, as referred to in the last section, a mass of microstructural defects especially supersaturated vacancy (vacancy-cluster) defects were introduced within the near surface layer due to HCPEB irradiation. These structural defects supply plenty of pathways and positions for the adsorption
14
Z. Zhang et al. / Applied Surface Science 294 (2014) 9–14
and entrance of dissolved O2− , promoting the formation of thicker and denser passive layer. However, overlarge density of craters (like 5-pulsed samples) will damage the continuity and integrity of oxide layer, which can badly affect the ability of oxide layer to prevent the corrosive anions and bring the corrosion resistance down. For 10-pulsed sample, surface passive film is still compact and thick. On the other hand, low density of craters was obtained on the irradiated surface, meaning the evident reduction of sensitive sites for the pitting corrosion. Therefore, the passive film induced by the 10 pulses of HCPEB treatment offers a good corrosion protection due to high destiny of structural defects and low density of craters.
Chinese Natural Science Foundation (Grant No. U1233111, 50671042), Open project of key Laboratory of Materials Modification by Laser, Ion and Electron Beams (Dalian University of Technology), Ministry of Education (DP1051102).
5. Conclusions
References
(1) HCPEB irradiation significantly changes the microstructures of cp copper. After HCPEB irradiation, the irradiated surface locally melted and then craters were formed in these regions. The crater density of 5-pulsed sample is obviously higher than that of 10-pulses. Besides, very abundant defect structures like dislocations were formed within the near surface irradiated. At the same time, the formation of supersaturated vacancies and three types of vacancy cluster defects (dislocation loops, SFTs, and voids) were also demonstrated after both 5- and 10-pulsed HCPEB irradiation. (2) HCPEB irradiation improves corrosion resistance of cp copper significantly. The corrosion resistance of 10-pulsed samples is better than original samples. For overlarge density of craters can bring the corrosion resistance of 5-pulsed samples down in spite of thicker and denser passive film. (3) The various structural defects introduced by HCPEB irradiation can play some role in the mechanisms of corrosion. These
[1] D.I. Proskurovsky, V.P. Rothtein, G.E. Ozur, et al., Surf. Coat. Technol. 125 (2000) 49. [2] K.M. Zhang, J.X. Zou, T. Grosdidier, et al., Surf. Coat. Technol. 201 (2006) 1393. [3] J.X. Zou, T. Grosdidier, K.M. Zhang, et al., J. Alloys Compd. 434 (2007) 707. [4] J.X. Zou, T. Grosdidier, K.M. Zhang, et al., Appl. Surf. Sci. 255 (2009) 4758. [5] X.D. Zhang, S.Z. Hao, X.N. Lin, et al., Appl. Surf. Sci. 257 (2011) 5899. [6] S.Z. Hao, P.S. Wu, J.X. Zou, et al., Appl. Surf. Sci. 253 (2007) 5349. [7] S.Z. Hao, X.D. Zhang, X.X. Mei, et al., Mater. Lett. 62 (2008) 414. [8] A.F. Shubin, V.P. Rotshtein, D.I. Proskurovskii, Sov. Phys. J. 17 (1974) 942. [9] V.P. Rotshtein, L.S. Bushnev, D.I. Proskurovskii, Sov. Phys. J. 18 (1975) 404. [10] D.M. Follstaedt, W.R. Wampler, Appl. Phys. Lett. 38 (1980) 140. [11] J.X. Zou, K.M. Zhang, C. Dong, et al., Appl. Phys. Lett. 89 (2006) 041913. [12] J.X. Zou, K.M. Zhang, S.Z. Hao, et al., Thin Solid Films 519 (2010) 1404. [13] V.A. Shulov, N.A. Nochovaya, Nucl. Instrum. Methods Phys. Res. 148 (1995) 154. [14] Q.F. Guan, Q. Zhang, C. Dong, ISIJ Int. 48 (2008) 235. [15] N. Hansen, Metall. Mater. Trans. A 32 (2001) 2917. [16] M. Kiritani, Mater. Chem. Phys. 50 (1997) 133. [17] A.D. Pogrebnjak, A.D. Mikhaliov, N.A. Progrebnjak Jr., et al., Phys. Lett. A 241 (1998) 357. [18] A. Norlin, J. Pan, C. Leygraf, Biomol. Eng. 19 (2002) 67. [19] A. Szummer, M. Janik-Czachor, S. Hofmann, Mater. Chem. Phys. 34 (1993) 181.
structural defects supply plenty of pathways and positions for the adsorption and entrance of dissolved O2− , promoting the formation of thicker and denser passive layer, which delay the corrosion process. Acknowledgements