Effect of surface mechanical attrition treatment on corrosion resistance of commercial pure titanium

    Effect of surface mechanical attrition treatment on corrosion resistance of commercial pure titanium Tianlin Fu, Zhaolin Zhan, Ling Zhang, Yanrong Yang, Zhong Liu, Jianxiong Liu, Li Li, Xiaohua Yu PII: DOI: Reference:

S0257-8972(15)30215-2 doi: 10.1016/j.surfcoat.2015.08.041 SCT 20515

To appear in:

Surface & Coatings Technology

Received date: Revised date: Accepted date:

17 March 2015 20 August 2015 21 August 2015

Please cite this article as: Tianlin Fu, Zhaolin Zhan, Ling Zhang, Yanrong Yang, Zhong Liu, Jianxiong Liu, Li Li, Xiaohua Yu, Effect of surface mechanical attrition treatment on corrosion resistance of commercial pure titanium, Surface & Coatings Technology (2015), doi: 10.1016/j.surfcoat.2015.08.041

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Effect of surface mechanical attrition treatment on corrosion resistance of commercial pure titanium 1

IP

T

Tianlin Fu, Zhaolin Zhan , Ling Zhang, Yanrong Yang, Zhong Liu, Jianxiong Liu, Li Li, Xiaohua Yu

SC R

Faculty of Material Science and Engineering, Kunming University of Science and Technology, No.68, Wenchang Road, Kunming 650093, China

Abstract:

NU

We report on the effects of surface mechanical attrition treatment on the

MA

corrosion behavior of commercial pure titanium. The corrosion resistance before and after treatment were investigated by studying potentiodynamic polarization curves

D

and by electrochemical impedance spectroscopy. The potentiodynamic polarization

TE

curves for a sample treated by surface mechanical attrition and for an untreated

CE P

commercial pure titanium sample at room temperature showed that the corrosion potential of the former ranged from -1.11 to -1.06 V, whereas that of the latter was

AC

-1.23 V. The corrosion current density for the treated sample ranged from -5.02×10-4 to -4.98×10-4A.cm-2, and that for the untreated sample was -4.56×10-4A.cm-2. A comparison of current densities at the same polarized potential showed a significant reduction in dissolution current of the treated sample. This indicates a lower corrosion rate for the sample treated by surface mechanical attrition. Surface mechanical attrition treatment was therefore confirmed to have beneficial impacts on corrosion behavior in 3.5 wt.% NaCl solution. Key words: A. Titanium; B. EIS; B. Polarization; C. Repassivation, C. Passive films

1

Corresponding author, E-mail address: [email protected], TEL: +86 0871 65109212

ACCEPTED MANUSCRIPT

1. Introduction

IP

T

Titanium and its alloys have proven to be important engineering alloys for

because of their high corrosion resistance

[1-5]

SC R

advanced applications in aerospace, automotive, chemical and biomedical industries . Titanium has a remarkable corrosion

behavior because of the very stable oxide film formed on its surface[6]. However, this

NU

film is unstable in reducing or complexing media, such as hydrochloric acid and

MA

solutions that contain chloride ions, which limits its industrial application [7-9]. In general, passive films formed on titanium and its alloys consist mainly of

D

amorphous titanium dioxide. The formation of a stable passive oxide film on titanium

TE

and its alloys may reduce their corrosion[10, 11]. A high-surface grain boundary density

CE P

is beneficial for the formation of a passive film[12]. Most material failures occur on the surface and therefore, controlling the surface

AC

properties can effectively improve the overall material behavior[13]. Many efforts have focused on surface hardening to improve corrosion resistance. Grain refining has been shown to be an effective method to improve surface properties [12-15]. Surface mechanical attrition treatment (SMAT) can be used to apply repeated impacts on a sample surface and refine grains on the material surface

[16, 17]

. But few

investigations have been carried out to predict the effect of SMAT on corrosion behavior when using a 3.5 wt.% NaCl solution (the equivalent concentration of sea water). Very few of these have investigated the influence of processing time on corrosion protection. Also, none have evaluated the corrosion protection mechanisms

ACCEPTED MANUSCRIPT and characterized the passive layers formed on material surfaces in 3.5 wt.% NaCl solution. The purpose of this work was to study the microstructure and properties of

IP

T

refined grains on the surface of commercial pure titanium as induced by SMAT.

SC R

Attention was focused primarily on evaluating the corrosion protection mechanisms. The effect of SMAT treatment on the corrosion behavior was also investigated and analyzed. Samples were processed under different SMAT conditions to form different

NU

refined grain surface layers. Electrochemical measurements were conducted in 3.5

MA

wt.% NaCl solution to investigate the corrosion behavior of commercial pure titanium before and after treatment. Commercial pure titanium was analyzed by X-ray

D

diffractometry (XRD) for phase analysis and calculation of grain size. Scanning

TE

electron microscopy (SEM) was carried out to characterize the surface microstructure.

titanium.

CE P

For comparison, the same measurements were also conducted on commercial pure

AC

2. Experimental

The material used in this study was commercial pure titanium with the following nominal composition (wt. %): C 0.02, Fe 0.10, O 0.15, N 0.02, H 0.0011 and a balance of titanium. The surfaces of rectangular specimens (100×100×2mm) were ground using 360-1200# SiC abrasive papers; polished using 2.5 μm diamond pastes until the surfaces reached a highly reflective level; cleaned ultrasonically with acetone, alcohol, and distilled water; and dried in air. SMAT may cause severe plastic deformation of the material surface. In SMAT,

ACCEPTED MANUSCRIPT the sample surface to be treated is peened with a large number of impacts over a short time using a high frequency system (50 kHz). SMAT was conducted for 15, 30, and

IP

T

45 min using a large quantity of 8-mm-diameter GCr15 balls and a vibration

SC R

amplitude of 50 kHz. After SMAT, the sample surface was electropolished slightly in 5% HCl + C2H5OH solution at room temperature to remove the impurities and surface contaminants, and to eliminate the effects of contamination on corrosion

NU

performance[18]. The final samples were termed SMATx, where x = 15, 30 or 45 min.

MA

Microstructures were examined using a G2 pro scanning electron microscope. X-ray diffraction (XRD) analysis of the surface layer was carried out using a Philips

D

PW 1830/00 diffractometer. X-ray photoelectron (XPS) analysis was performed on a

TE

PHI5500 system using AlKα radiation.

CE P

Electrochemical experiments were carried out in three-necked flasks using an electrochemical potentiostat model PARSTAT 4000 (Princeton Applied Research, NJ,

AC

USA) to identify the corrosion behavior. A saturated calomel electrode (SCE) was used as the reference electrode, the samples as the working electrode, and a platinum coil as the counter electrode. The temperature of the three-necked flask was maintained at 27 ± 0.1°C. Potentiodynamic polarization tests were initiated after 1 h immersion of the samples in a 3.5 wt.% NaCl solution, when a stable open circuit potential can be obtained at a potential scanning speed of 3 mV/s from -2000 mV to 1000 mV. Electrode impedance spectroscopy (EIS) tests were conducted at an open circuit potential value to characterize the corrosion behavior of these samples. The electrode response was analyzed from ~0.01 Hz to 100 kHz using a 10 mV amplitude

ACCEPTED MANUSCRIPT alternating current voltage signal. The EIS tests were recorded at the open circuit potential developed by the samples after 600 s of immersion in the 3.5 wt.% NaCl

IP

T

solution. All electrochemical tests were repeated three times to ensure reproducibility

SC R

of the measurements, and reproducibility of the data is expressed in terms of the corresponding standard deviation. 3. Results

NU

3.1 Potentiodynamic polarization curve studies

MA

To investigate the corrosion behavior before and after SMAT, potentiodynamic polarization tests were carried out on commercial pure titanium samples, in 3.5 wt.%

D

NaCl solution at pH = 6.0 (Fig. 1).

TE

On the cathodic branch, the main reaction is hydrogen evolution reaction. The

CE P

chemical equation may be[19]:

2H++2e-→H2↑

AC

On the anodic branch, the curves describe typical behavior for metals that undergo passivation when exposed to a certain level of electric current because of the formation of a passive oxide film over the sample surface, which inhibits corrosion evolution. The formation of the passive film can be represented by the following chemical reactions[20]: Ti→Ti3+ + 3e-; 2Ti3+ + 3H2O→Ti2O3 + 6H+; Ti3+→Ti4+ + e-; Ti4+ + H2O→TiO2 + 4H+;

ACCEPTED MANUSCRIPT The series of reactions produces Ti3+/Ti4+ ions, which form a stable passive film of TiO2 and Ti2O3. The TiO2/Ti2O3 film protects the material from further corrosion.

IP

T

The presence of aggressive ions, such as chloride, sulfate and fluoride in solution,

SC R

accelerates the anodic process[21]. According to Liu, Cl- can migrate across the passive oxide film in parallel with oxide ions[22], as represented by the following chemical reaction:

NU

TixOy + 2Cl-→TixOyCl2 + 2e-

MA

A wide passive region from -1.0 to -0.25 VSCE with current density ~1 μA.cm-2 was observed after a short active dissolution for the SMAT samples. In contrast, a

D

narrow passive region existed for the untreated samples from -1.10 to -1.15 VSCE.

TE

Corrosion resistance behavior is exhibited in the region in which the samples are

CE P

passive. The current increases for SMAT samples at a potential up to -0.25 VSCE, according to Oliveira et al. and Tavares et al., which indicates that the passive film is 23]

. However, the current range is set rapidly as per the metal’s

AC

breaking[1,

repassivation. For the SMAT samples, the SMAT15 sample translated directly into the passive region from the Tafel region from -1.10 to 1 VSCE, and exhibited a typical self-passivation characterization. However, the SMAT30 and 45 experience several active–passive transitions from -1.10 to 0.14 VSCE and then tend to passivation again. This indicates the formation of a passive film that is not sufficiently protective under immersion into the electrolyte, and in the active–passive region the corrosion rate increased[24]. According to Geetha et al.[25], the phenomenon of metal repassivation also plays an important role in the alloy’s corrosion behavior. Titanium alloys show a

ACCEPTED MANUSCRIPT faster repassivation phenomenon than that from stainless steel. This provides excellent corrosion resistance for SMAT30 and 45 samples

[26-28]

. In the passive

IP

T

region, the SMAT sample passive current densities were lower than for the untreated

SC R

sample. The decrease in current density resulted because the passivating oxide film dominated over the dissolution rate on the bare surface

[29-32]

. By comparing the

current densities at the same polarized potential, a significant reduction in dissolution

NU

current of the SMAT sample can be observed. This represents a reduction in SMAT

MA

sample corrosion rate. A comparison of the SMAT samples shows that the SMAT15 sample has no active–passive transitions, and that the current density always

D

maintains a lower value. However, the SMAT30 and 45 samples have active–passive

TE

transitions, and in this region, the current density increased. This phenomenon may

CE P

occur because the SMAT processing time is too long and the substrate surface has many more micro-cracks, which reduces the corrosion resistance.

AC

The corrosion potential (Ecorr), corrosion current density (icorr), and anodic and cathodic Tafel slopes (βa and βc) of the materials were determined by Tafel analysis of anodic and cathodic branches of the polarization curves using EC-Lab software (see Table 1). Data show that all samples have a low current density and high corrosion potential. Compared with the corrosion potential (Ecorr) of the untreated sample (approximately -1.25 V), the Ecorr increased to approximately -1.05 V after SMAT. This increase represents the achievement of a more noble electrode potential, and indicates an improved corrosion resistance of the pure titanium following SMAT treatment. The corrosion current density is characterized as an important parameter

ACCEPTED MANUSCRIPT for evaluating the corrosion reaction kinetics[33]. The metal corrosion rate is normally proportional to the corrosion current density as measured by Tafel plots and the lower

IP

T

current density implies a lower corrosion rate.

SC R

In addition, the polarization resistance (P.R.) of the Tafel curves was calculated using the Stern-Geary equation.

 C   2.303   C    icorr

NU

P.R. 

The polarization resistance for untreated sample is 72.6 Ω.cm2, and the

MA

polarization resistances for SMAT15, 30, and 45 samples are 221.6, 227.8, and 229.4

D

Ω.cm2, respectively. The results showed that the polarization resistance of SMAT

TE

samples is much higher than that on the untreated sample surface. 3.2 Electrode impedance spectroscopy tests

CE P

Electrochemical impedance spectroscopy (EIS) was used to obtain information on the characteristics of the passive films and the nature of the electrochemical processes

AC

at the interface. From potentiodynamic polarization curves, we find that SMAT samples exhibit similar electrochemical performance. We chose untreated and SMAT30 samples for the EIS tests. The SMAT sample data and untreated sample measured after 600 s of immersion in the 3.5 wt.% NaCl solution at 27°C is presented in Figs 2. The Nyquist plots for SMAT30 and the untreated sample exhibit an incomplete and capacitive-like semicircle. This phenomenon is related to the charge transfer reaction from the sample surface to the electrolyte through the double electrochemical layer[34]. An increase in semicircle indicates an increase in film stability and a

ACCEPTED MANUSCRIPT decrease in semicircle diameter indicates a decrease in passive film resistance. A similar trend was observed in the impedance spectra of some Ti alloys, such as

IP

T

Ti-6Al-4V[12] and Ti45Zr38Al17[35]. The semicircular diameter of the SMAT30

SC R

sample is much larger than the untreated sample, and demonstrates a nobler electrochemical behavior.

The corresponding Bode magnitude plots have three characteristic regions and

NU

are presented in Fig. 2. A flat portion exits in the high- and middle-frequency range

MA

(10–105 Hz), which has a slope that is approximately equal to 0. This results from the response of electrolyte resistance. In the low frequency ranges (10-2–101 Hz), the

D

impedance spectra display a linear slope of approximately -1, which is characteristic

TE

of a typical passive film presented on the surface and is a response of the passive film.

CE P

In the low frequency ranges (10-2–101 Hz), |Z| finally reaches a certain value because of the contribution of passive film resistance, and the SMAT30 sample shows a much

AC

higher |Z| (almost 60000 Ω.cm2) than the untreated sample (almost 10000 Ω.cm2). This means that the SMAT process can generate a more effective barrier layer and a passive film on its surface during immersion. Bode phase plots show that SMAT30 and the untreated sample have three characteristic regions. In the high-frequency region, the phase angle drops towards 0 with the response of electrolyte resistance; in the middle-frequency region, phase angle peaks exist, which indicate the interaction of one time constant and the formation of a duplex film; and in the low-frequency region, the phase angle decreases to a lower value because of the passive film.

ACCEPTED MANUSCRIPT Analysis of the EIS measurements is done by fitting the impedance data to an equivalent circuit. For the untreated sample, the corresponding equivalent circuit in

IP

T

this immersion system can be modeled by Randle’s circuit (Rsol(RtcCdl)) shown in Fig.

SC R

3[35], which assumes that the corrosion of the passive metal is hindered by an oxide film that acts as a barrier-type compact layer. Rtc and Cdl describe properties of the passive film formed on the sample surface, namely, the charge transfer resistance

NU

corresponding to the resistance to electron transfer during electrochemical reaction

MA

and capacitance of the oxide film, respectively, whereas Rsol represents the solution resistance of the 3.5 wt.% NaCl solution. Excellent agreement between the

D

experimental and model values was obtained. The impedance parameters determined

TE

from fitting the equivalent circuit to the EIS data are given in Table 2.

CE P

The Rtc values are strongly dependent on the passive film characteristic and corrosion resistance of the materials. A higher Rtc value implies good corrosion

AC

resistance. A decrease in Rtc value reflects the protective efficiency decreases of the passive layer[34]. The Rtc values increase after SMAT treatment and reach 197 kΩ. This increase is associated with the formation of a passive layer in 3.5 wt.% NaCl solution on the surface of commercial pure titanium, which can improve the corrosion properties. The high reactivity of titanium with oxygen ensures the formation of the passive film that protects the surface of the treated sample, as proven in other studies. 3.3 Characteristics of treated samples Figure 4 shows the surface microstructure of SMAT samples. It has demonstrated that SMAT method would lead to the refinement of the grains on surface layer of

ACCEPTED MANUSCRIPT metals [36-39]. The corrosion resistance of titanium is influenced by the environment and temperature to which it is exposed, and by its microstructure and chemical

IP

T

composition. If the environment, temperature and chemical composition conditions

SC R

are fixed, the improvement in corrosion behavior of titanium is assumed to be a function of microstructure. The high grain boundary density was beneficial for the formation of a passive film, which accounts for the improvement in corrosion

NU

behavior.

MA

The wide-scan survey spectra of passive films formed on the untreated sample and SMAT 30 sample are shown in Fig. 5. As Fig. 5(a) shows, the native passive film

D

on the untreated sample is composed of O, Ti, C, and N. The peaks corresponding to

TE

the C 1s and N 1s are attributed to surface contamination of the specimens during the

CE P

experiments. Furthermore, the XPS profile analysis revealed that the passive film formed on the untreated sample was composed mainly of titanium oxides. The

AC

spectrum corresponding to the SMAT sample (Fig. 5(b)), reveals the presence of O, Ti, C, and N. As in the case of its pure titanium counterpart, the peaks of C 1s and N 1s are attributed to contamination of the surface. Analysis of the XPS profiles also confirmed that titanium and nickel oxides formed on the nickelized SMAT sample. By considering that oxygen is mainly involved in titanium oxide, and based on the molar concentrations of O and Ti elements, it is possible to give an average stoichiometry value regarding the passive oxide layer. Composition data of untreated sample (wt. %) is O: 34.4, Ti: 16.5 and SMAT sample is O: 45.5, Ti: 26.93 which show a noticeable enrichment of oxygen on treated

ACCEPTED MANUSCRIPT samples, indicating that SMAT treatment enhances the formation of the passive oxide layer, as expected.

IP

T

The surface morphologies of untreated and SMAT samples after being immersed

SC R

in 3.5 wt.% NaCl solution for 240 h are shown in Fig. 6. Comparing a with b, c, and d, shows that the SEM morphology of the SMAT samples after immersion differs from that of the untreated sample. After 10 days immersion, pits formed on the

NU

surface, and the size and area of corrosion pits on the surface of the untreated sample

MA

increased and are larger than those of the SMAT samples. Because of the SMAT process, samples exhibit a better corrosion resistance than untreated titanium.

D

According to electrochemical and immersion tests, the SMAT process promotes

TE

the corrosion resistance behavior of commercial pure titanium in 3.5 wt.% NaCl

CE P

solution. This is consistent with the corrosion potential, the corrosion current density, more noble passive region and Rtc results reported previously.’ It is due to the repeated

AC

impact provided by the GCr15 balls on its surface, resulting in a large amount of deformation in the surface during the SMAT process.

4. Discussion In the SMAT process, severe plastic deformation causes a large number of dislocation deliveries and proliferation, which increased the number of vacancies, interstitial atoms, stacking faults, and other structural defects, and then increased the dislocation motion[40]. Because of the limited number of active slip systems at room temperature, twinning plays an important role in accommodating homogeneous

ACCEPTED MANUSCRIPT plastic deformation in commercial purity α-titanium[41]. Since the interfaces of deformation twins can also present obstacles to dislocation glide, the multiple impact

IP

T

nature of the shot peening process implies that dislocation pile-up at twin interfaces

SC R

will occur should the mechanical twins form early on in the peening process. Furthermore the SMAT process can also lead to an increase in measured levels of subsurface oxygen compared with untreated conditions. Thomas et al.[42, 43] has shown

NU

using SIMS how oxygen increases in shot-peened samples compared with non-SP.

MA

This is attributed to mechanical twinning and increasing diffusion pathways for oxygen, and the increased sub-surface oxygen content in SMAT samples can be

D

attributed to rapid diffusion along the mechanical twin boundaries and dislocation slip

TE

bands within the plastically deformed surface layer. It is believed that grain

CE P

refinement supplies a high density of nucleation sites for passive film formation[41, 42], which assists in the formation of a dense passive layer and a reduction in corrosion

AC

rate. The improved corrosion resistance behavior of the SMAT sample can be attributed to the rapid formation of a passive film on the grain refinement surface, which improves the chemical stability of the alloy by decreasing the diffusion of corrosive ions and the activity of charge carriers at the interface between solution and substrate [23, 25, 44, 45]. An accumulation of metal chloride at the metal/film interface may cause oxide film rupture, which leads to an initiation of pits. Through SMAT treatment, the passive film thickness increases. This makes it more difficult for chloride ions to cross the passive oxide film and increases the extent of corrosion.

ACCEPTED MANUSCRIPT 5. Conclusion SMAT process has a significant influence on the corrosion resistance of titanium.

IP

T

The corrosion potential of the SMAT samples ranged from -1.11 to -1.06 V, whereas

SC R

that of the untreated was -1.25 V. The corrosion current density for the SMAT sample ranged from -5.02×10-4 to -4.98×10-4A.cm-2, and that for the untreated sample was -4.56×10-4A.cm-2. A higher corrosion potential decreased the tendency for corrosion, a

NU

very large passive potential range provided a good anticorrosive resistance, and a low

MA

corrosion current density provided low corrosion rates. The increase in corrosion resistance is attributed to the formation of a dense

D

passive film on the SMAT samples. Through SMAT treatment, the more stable passive

TE

film formed on the surface of titanium which makes it more difficult for chloride ions

CE P

to cross the passive oxide film.

AC

Acknowledgements

This work was supported financially by the Natural Science Foundation of China (no. 51165016).

Reference [1] Oliveira VMCA, Aguiar C, Vazquez AM, et al. Improving corrosion resistance of Ti–6Al–4V alloy through plasma-assisted PVD deposited nitride coatings[J]. Corrosion Science, 2014,88: 317-327. [2] Pohrelyuk IM, Fedirko VM, Tkachuk OV, et al. Corrosion resistance of

ACCEPTED MANUSCRIPT Ti–6Al–4V alloy with nitride coatings in Ringer’s solution[J]. Corrosion Science, 2013,66: 392-398.

IP

T

[3] Ahn H, Lee D, Lee K-M, et al. Oxidation behavior and corrosion resistance of

SC R

Ti–10Ta–10Nb alloy[J]. Surface and Coatings Technology, 2008,202(22-23): 5784-5789.

[4] Chen S, Wu BH, Xie D, et al. The adhesion and corrosion resistance of Ti–O

NU

films on CoCrMo alloy fabricated by high power pulsed magnetron sputtering

MA

(HPPMS)[J]. Surface and Coatings Technology, 2014,252: 8-14. [5] da Silva LLG, Ueda M, Silva MM, et al. Corrosion behavior of Ti–6Al–4V alloy

D

treated by plasma immersion ion implantation process[J]. Surface and Coatings

TE

Technology, 2007,201(19-20): 8136-8139.

CE P

[6] Narayanan R, Seshadri SK. Point defect model and corrosion of anodic oxide coatings on Ti–6Al–4V[J]. Corrosion Science, 2008,50(6): 1521-1529.

AC

[7] Duarte LT, Biaggio SR, Rocha-Filho RC, et al. Surface characterization of oxides grown on the Ti–13Nb–13Zr alloy and their corrosion protection[J]. Corrosion Science, 2013,72: 35-40. [8] Fojt J, Joska L, Málek J. Corrosion behaviour of porous Ti–39Nb alloy for biomedical applications[J]. Corrosion Science, 2013,71: 78-83. [9] Li Y, Qu L, Wang F. The electrochemical corrosion behavior of TiN and (Ti,Al)N coatings in acid and salt solution[J]. Corrosion Science, 2003,45(7): 1367-1381. [10] Diamanti MV, Pedeferri MP. Effect of anodic oxidation parameters on the titanium oxides formation[J]. Corrosion Science, 2007,49(2): 939-948.

ACCEPTED MANUSCRIPT [11] Ferreira EA, Rocha-Filho RC, Biaggio SR, et al. Corrosion resistance of the Ti–50Zr at.% alloy after anodization in different acidic electrolytes[J]. Corrosion

IP

T

Science, 2010,52(12): 4058-4063.

SC R

[12] Jelliti S, Richard C, Retraint D, et al. Effect of surface nanocrystallization on the corrosion behavior of Ti–6Al–4V titanium alloy[J]. Surface and Coatings Technology, 2013,224: 82-87.

NU

[13] Ge L, Tian N, Lu Z, et al. Influence of the surface nanocrystallization on the gas

MA

nitriding of Ti–6Al–4V alloy[J]. Applied Surface Science, 2013,286: 412-416. [14] Chui P, Sun K, Sun C, et al. Effect of surface nanocrystallization induced by fast

D

multiple rotation rolling on hardness and corrosion behavior of 316L stainless

TE

steel[J]. Applied Surface Science, 2011,257(15): 6787-6791.

CE P

[15] Chen C, Liu X, Zhao X, et al. The Effects of Intermediate Layer and Surface Nanocrystallization on the Vacuum Diffusion Bonding of Commercially Pure

AC

Titanium[J]. Physics Procedia, 2012,25: 63-67. [16] Mao X-y, Li D-y, Wang Z-z, et al. Surface nanocrystallization by mechanical punching process for improving microstructure and properties of Cu-30Ni alloy[J]. Transactions of Nonferrous Metals Society of China, 2013,23(6): 1694-1700. [17] Wang Y, Yue W, She D, et al. Effects of surface nanocrystallization on tribological

properties

of

316L stainless

steel

under

MoDTC/ZDDP

lubrications[J]. Tribology International, 2014,79: 42-51. [18] Jin L, Cui W-f, Song X, et al. Effects of surface nanocrystallization on corrosion

ACCEPTED MANUSCRIPT resistance of β-type titanium alloy[J]. Transactions of Nonferrous Metals Society of China, 2014,24(8): 2529-2535.

IP

T

[19] Jiang Z, Dai X, Middleton H. Effect of silicon on corrosion resistance of Ti–Si

SC R

alloys[J]. Materials Science and Engineering: B, 2011,176(1): 79-86. [20] Heljo PS, Wolff K, Lahtonen K, et al. Anodic Oxidation of Ultra-Thin Ti Layers on ITO Substrates and their Application in Organic Electronic Memory

NU

Elements[J]. Electrochimica Acta, 2014,137: 91-98.

MA

[21] Escrivà-Cerdán C, Blasco-Tamarit E, García-García DM, et al. Effect of temperature on passive film formation of UNS N08031 Cr–Ni alloy in acid

contaminated

with

different

aggressive

anions[J].

D

phosphoric

TE

Electrochimica Acta, 2013,111: 552-561.

CE P

[22] Liua C, Leyland A, Bi Q, et al. Corrosion resistance of multi-layered plasmaassisted physical vapour deposition TiN and CrN coatings[J]. Surface and

AC

Coatings Technology, 2001:164-173. [23] Tavares AMG, Fernandes BS, Souza SA, et al. The addition of Si to the Ti–35Nb alloy and its effect on the corrosion resistance, when applied to biomedical materials[J]. Journal of Alloys and Compounds, 2014,591: 91-99. [24] Xie F, He X, Cao S, et al. Influence of pore characteristics on microstructure, mechanical properties and corrosion resistance of selective laser sintered porous Ti–Mo alloys for biomedical applications[J]. Electrochimica Acta, 2013,105: 121-129. [25] Geetha M, Singh AK, Asokamani R, et al. Ti based biomaterials, the ultimate

ACCEPTED MANUSCRIPT choice for orthopaedic implants – A review[J]. Progress in Materials Science, 2009,54(3): 397-425.

IP

T

[26] Guo X, Wang S, Gong J, et al. Characterization of highly corrosion-resistant

SC R

nanocrystalline Ni coating electrodeposited on Mg–Nd–Zn–Zr alloy from a eutectic-based ionic liquid[J]. Applied Surface Science, 2014,313: 711-719. [27] Li T, Liu L, Zhang B, et al. Direct observation of thin membrane passive film

NU

over the growing pit on sputtered nanocrystalline austenitic stainless steel

MA

film[J]. Electrochemistry Communications, 2015,52: 80-84. [28] Jang H, Park C, Kwon H. Photoelectrochemical analysis on the passive film

TE

3503-3508.

D

formed on Ni in pH 8.5 buffer solution[J]. Electrochimica Acta, 2005,50(16-17):

CE P

[29] Fernández-Domene RM, Blasco-Tamarit E, García-García DM, et al. Repassivation of the damage generated by cavitation on UNS N08031 in a LiBr

AC

solution by means of electrochemical techniques and Confocal Laser Scanning Microscopy[J]. Corrosion Science, 2010,52(10): 3453-3464. [30] Fernández-Domene RM, Blasco-Tamarit E, García-García DM, et al. Thermogalvanic corrosion of Alloy 31 in different heavy brine LiBr solutions[J]. Corrosion Science, 2012,55: 40-53. [31] Leiva-García R, García-Antón J, Muñoz-Portero MJ. Contribution to the elucidation of corrosion initiation through confocal laser scanning microscopy (CLSM)[J]. Corrosion Science, 2010,52(6): 2133-2142. [32] Fernández-Domene RM, Blasco-Tamarit E, García-García DM, et al. Cavitation

ACCEPTED MANUSCRIPT corrosion and repassivation kinetics of titanium in a heavy brine LiBr solution

IP

Microscopy[J]. Electrochimica Acta, 2011,58: 264-275.

T

evaluated by using electrochemical techniques and Confocal Laser Scanning

SC R

[33] Handzlik P, Fitzner K. Corrosion resistance of Ti and Ti–Pd alloy in phosphate buffered saline solutions with and without H2O2 addition[J]. Transactions of Nonferrous Metals Society of China, 2013,23(3): 866-875.

NU

[34] Ningshen S, Kamachi Mudali U, Ramya S, et al. Corrosion behaviour of AISI

MA

type 304L stainless steel in nitric acid media containing oxidizing species[J]. Corrosion Science, 2011,53(1): 64-70.

D

[35] Jayaraj J, Ravi Shankar A, Kamachi Mudali U. Electrochemical and passive

TE

characterization of a beta type Ti45Zr38Al17 cast rod in nitric acid medium[J].

CE P

Electrochimica Acta, 2012,85: 210-219. [36] Tong WP, Tao NR, Wang ZB, et al. Nitriding iron at lower temperatures[J].

AC

Science, 2003, 299:686-688. [37] Jin L, Cui WF, Song X, et al. The formation mechanisms of surface nanocrystallites in β -type biomedical TiNbZrFe alloy by surfacemechanical attrition treatment. Applied Surface Science, 2015,347: 553-560. [38] Liu Y, Jin B, Li DJ et al. Wear behavior of nanocrystalline structured magnesium alloy induced by surface mechanical attritiontreatment. Surface and Coatings Technology, 2015, 261:219-226. [39] Guo S, Wang ZB, Wang LM, et al. Lower-temperature aluminizing behaviors of a ferritic–martensitic steel processed by means of surface mechanical attrition

ACCEPTED MANUSCRIPT treatmen. Surface and Coatings Technology, 2014, 258:329-336. [40] Li Y, Sun K, Liu P, et al. Surface nanocrystallization induced by fast multiple

IP

T

rotation rolling on Ti-6Al-4V and its effect on microstructure and properties[J].

SC R

Vacuum, 2014,101: 102-106.

[41] Gurao NP, Kapoor R, Suwas S. Deformation behaviour of commercially pure titanium at extreme strain rates[J]. Acta Materialia, 2011,59(9): 3431-3446.

NU

[42] Thomas M, Lindley T, Rugg D, et al. The effect of shot peening on the

MA

microstructure and properties of a near-alpha titanium alloy following high temperature exposure[J]. Acta Materialia, 2012,60(13-14): 5040-5048.

D

[43] Thomas M, Lindley T, Jackson M. The microstructural response of a peened

TE

near-α titanium alloy to thermal exposure[J]. Scripta Materialia, 2009,60(2):

CE P

108-111.

[44] Chen S, Xiong W, Yao Z, et al. Corrosion behavior of Ti(C,N)-Ni/Cr cermets in

AC

H2SO4 solution[J]. International Journal of Refractory Metals and Hard Materials, 2014,47: 139-144. [45] Shanaghi A, Chu PK, Sabour Rouhaghdam AR, et al. Structure and corrosion resistance of Ti/TiC coatings fabricated by plasma immersion ion implantation and deposition on nickel–titanium[J]. Surface and Coatings Technology, 2013,229: 151-155.

ACCEPTED MANUSCRIPT Tables Table 1 Corrosion parameters determined from potentiodynamic polarization curves

SC R

IP

T

measured for samples before and after SMAT in 3.5 wt.% NaCl solution

AC

CE P

TE

D

MA

NU

Table 2 EIS data simulation for untreated and SMAT30 samples

ACCEPTED MANUSCRIPT Figure captions Fig. 1 Potentiodynamic polarization curves for samples before and after SMAT after 1

SC R

IP

T

h immersion in 3.5 wt.% NaCl solution at 27°C. Scanning rate: 3 mV.s-1.

Fig. 2 Nyquist and Bode diagrams for untreated sample (a,c) and SMAT sample (b,d)

NU

after 600 s of immersion in 3.5 wt.% NaCl solution at 27°C, measured at Ecorr.

MA

Fig. 3 Equivalent circuits used to analyze impedance spectra.

TE

CE P

(d)SMAT45 samples

D

Fig. 4 The surface morphologies of (a) untreated, (b)SMAT15, (c)SMAT30 and

Fig. 5 XPS wide-scan spectrum obtained for the (a) pure titanium and (b) SMAT

AC

sample

Fig. 6 Surface morphologies of (a) untreated, (b) SMAT15, (c) SMAT30, (d) SMAT45 after immersion in 3.5 wt.% NaCl solution for 240 h.

ACCEPTED MANUSCRIPT Table 1 2

Ecorr, VSCE

icorr, A.cm-2

P.R. Ω.cm

Untreated

-1.25

-4.56×10-4

72.6

SMAT15

-1.06

-5.01×10-4

SMAT30

-1.11

SMAT45

-1.07

SC R

IP

T

Sample

221.6 227.8

-5.02×10-4

229.4

NU

-4.98×10-4

MA

Table 2 Rs (Ω)

Rtc×103 (Ω)

Cdl (μF)

Untreated

23.57

58.58

7.56×10-2

197

4.628×10-2

AC

24.68

TE

CE P

SMAT30

D

Sample

ACCEPTED MANUSCRIPT Fig. 1

-1

IP

Repassivation

-3

UT SMAT15 SMAT30 SMAT45

-4

-5 -1.5

-1.0

-0.5

0.0

NU

-2.0

SC R

Log i /(A.cm-2)

-2

CE P

TE

D

MA

E/(V vs. SCE)

AC

T

Breakdown of the passive film Passivation

0.5

1.0

ACCEPTED MANUSCRIPT

4

1.0x10

5

1.2x10

a

simulated

3

5

1.0x10

-Zlm  cm2

3

6.0x10

3

4.0x10

4

8.0x10

4

6.0x10

4

4.0x10

NU

3

2.0x10

4

2.0x10

3

Zre cm

8.0x10

2

c

80

untreated sample simulated

D

60

TE

40 20

-20 -40 -1

10

0

10

CE P

0

1

10

2

10

1.4x10

4

1.2x10

4

1.0x10

4

8.0x10

3

6.0x10

3

4.0x10

3

2.0x10

3

0.0

4

2.0x10

4

4

4.0x10

4

6.0x10

Zre  cm

8.0x10

5

1.0x10

2

4

7x10

d

80

SMAT sample 4 6x10 simulated

60

4

5x10 40

4

4x10

20

4

3x10

0

4

2x10

-20

4

1x10

-40

0 3

10

Frequence /Hz

AC

0.0

4

1.0x10

4

10

5

10

-1

10

0

10

1

10

2

10

3

10

Frequence /Hz

4

10

5

10

Zmod  cm2

3

6.0x10

-Phase angle /degree

3

4.0x10

Zmod  cm2

3

2.0x10

MA

0.0

-Phase angle /degree

SMAT sample simulated

SC R

8.0x10

-Zlm /cm2

b

untreated sample

IP

T

Fig. 2

ACCEPTED MANUSCRIPT Fig. 3

Rtc

substrate

AC

CE P

TE

D

MA

NU

Oxide layer

SC R

Cd1

IP

T

Rs

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig. 4

D

TE

CE P

AC 1000 800 600 400 200

Binding Energy / eV 0

1000

800

600 400

Binding Energy / eV

- Ti 3p

- C 1s

T

- O 1s

b - Ti 2p3

- Ti 2s

IP

- O KLL

counts / intensity

SC R

-Ti3p

-C1s

-O 1s

a

NU

MA

-Ti2p3

-Ti2s -O 1s

counts / intensity

ACCEPTED MANUSCRIPT

Fig. 5

200 0

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig. 6

ACCEPTED MANUSCRIPT

IP

T

Highlights

SMAT impacts passive film growth favorably by gaining grain boundaries.



A large passive potential range provided a good anticorrosive resistance by

SC R



SMAT.

After SMAT, a low corrosion current density provided low corrosion rates.



Ecorr increase and Icorr decrease after SMAT.



Rtc values for the SMAT sample were higher than those for the untreated sample.

AC

CE P

TE

D

MA

NU

