materials Article

Pitting Corrosion Behaviour of New Corrosion-Resistant Reinforcement Bars in Chloride-Containing Concrete Pore Solution Jin-yang Jiang 1,2, * 1

2 3

*

ID

, Yao Liu 1,2 , Hong-yan Chu 1,2 , Danqian Wang 1,2 , Han Ma 3 and Wei Sun 1,2

School of Materials Science and Engineering, Southeast University, Nanjing 211189, China; [email protected] (Y.L.); [email protected] (H.-y.C.); [email protected] (D.W.); [email protected] (W.S.) Jiangsu Key Laboratory of Construction Materials, Nanjing 211189, China Research Institute of Jiangsu Shasteel Iron and Steel, Zhangjiagang 215625, China; [email protected] Correspondence: [email protected]; Tel.: +86-025-5209-0667

Received: 8 July 2017; Accepted: 1 August 2017; Published: 4 August 2017

Abstract: In this study, the pitting behaviour of a new corrosion-resistant alloy steel (CR) is compared to that of low-carbon steel (LC) in a simulated concrete pore solution with a chloride concentration of 5 mol/L. The electrochemical behaviour of the bars was characterised using linear polarisation resistance (LPR) and electrochemical impedance spectroscopy (EIS). The pitting profiles were detected by reflective digital holographic microscopy (DHM), scanning electron microscopy (SEM), and the chemical components produced in the pitting process were analysed by X-ray energy dispersive spectroscopy (EDS). The results show that the CR bars have a higher resistance to pitting corrosion than the LC bars. This is primarily because of the periodic occurrence of metastable pitting during pitting development. Compared to the pitting process in the LC bars, the pitting depth grows slowly in the CR bars, which greatly reduces the risk of pitting. The possible reason for this result is that the capability of the CR bars to heal the passivation film helps to restore the metastable pits to the passivation state. Keywords: corrosion-resistant alloy steel bars; simulated concrete pore solution; digital holographic microscope; pitting corrosion

1. Introduction Pitting corrosion is one of the most widely occurring forms of localised corrosion in metal. It usually occurs in a series of corrosive environments where chloride is the most common corrosive ion [1,2]. Pitting generally occurs on the metal surface, and may cause perforation or stress corrosion. The pits are nucleated on a microscopic scale and are covered by the corrosion product; therefore, pitting is one of the more destructive and undetectable forms of corrosion in metals [3]. There are many factors that affect pitting corrosion. Bautista et al. [4] investigated the corrosion behavior of different kinds of steel bars subjected to various aggressive conditions. Major factors included the composition of the steel bars and the external conditions, such as the chemical composition of the environment and the pH. Pitting corrosion in metal occurs in three stages [5]: pitting nucleation, metastable pitting, and steady-state pitting. In recent years, the development of electrochemical technology has provided new methods for pitting corrosion research. Williams [6] found a higher anode current density at the boundary of inclusions using a scanning electrochemical microscope (SECM). Vuillemin et al. [7] observed the current ripple around the pits with the scanning vibrating electrode technique (SVET). Ameri et al. [8] studied the pitting corrosion behaviour of 316 stainless steel in a sodium chloride solution via zeta sets and constant potential polarisation. Materials 2017, 10, 903; doi:10.3390/ma10080903

www.mdpi.com/journal/materials

Materials 2017, 10, 903

2 of 21

They found that maintaining the stability of the surface potential of the pitting product is important to the stable development of pitting corrosion. In addition to electrochemical measurements, Kocijan et al. [9] and Natishan et al. [10] conducted several studies on the surface features of pitting corrosion; however, the interaction between chloride ions and passivation films remains unclear. Ernst and Newman [11] experimentally studied two-dimensional (2D) growth at the edge of a stainless steel sheet, and proposed a semi-quantitative model for the lace-like pit cover. Laycock et al. [12] proposed 2D models to predict pitting corrosion growth under the influence of surface roughness, concentration of chloride in solution, and potential. Their results were in good agreement with experiments. Vignal et al. [13] studied the electrochemical behaviour of pitting in stainless steel using secondary ion mass spectrometry (SIMS), X-ray photoelectron spectroscopy (XPS), and SVET techniques. The experimental results were consistent with those of a numerical simulation. At present, many studies have investigated the electrochemical process and topography profiles of pitting corrosion, but the analysis of pitting morphology is limited to 2D plane effects. The three-dimensional (3D) development process of pitting is yet to be investigated. In this study, the pitting behaviour of low-carbon steel (LC) is compared to that of a new corrosion-resistant alloy steel (CR) in a simulated concrete pore solution with a chloride concentration of 5 mol/L. The electrochemical behaviour of the steel bars is characterised using electrochemical measurements. The 2D morphological changes in the pitting are observed using scanning electron microscopy (SEM). The 3D morphology of the pits is analysed using reflective digital holographic microscopy (DHM). 2. Experiment 2.1. Raw Materials Comparative studies are conducted using new Cr10Mo1 corrosion-resistant alloy steel (CR) bars and low-carbon steel (LC) bars. The specific composition of the steel is shown in Table 1. Table 1. Chemical composition of the experimental steel (percentage). Chemical Composition

C

LC CR

0.22 0.01

Si

Mn

0.53 1.44 0.487 1.49

Cr

Cu

Ni

Al

Mo

10.36

-

-

-

1.162

LC, low-carbon steel; CR, corrosion-resistant alloy steel.

There are no standards for the selection of components for the simulated concrete pore solutions. The primary influencing factors affecting corrosion are the chemical composition of the steel, the solution contents, and the pH [14]. As reported in the literature, a simulated solution can be roughly divided into one of two categories; i.e., it is either a highly alkaline solution or a saturated calcium hydroxide solution. Most highly alkaline solutions contain K+ and Na+ ions, and their pH values are greater than 13. The pH of the saturated calcium hydroxide solutions is generally between 12.5 and 12.6. Studies have shown that compared to the saturated calcium hydroxide solution, the high-alkaline solution is more similar to the actual concrete pore solution [15] Therefore, the simulated concrete pore solution used in this study is a highly alkaline solution with a pH of 13.6. In addition, 0.003 mol/L CaSO4 was also added to the highly alkaline solution, so as to make the highly alkaline solution more similar to the actual concrete pore solution. The etching solution used in this study is obtained by adding a 5 mol/L (5 M) sodium chloride solution to the above-mentioned simulated concrete pore solution. In order to accelerate the corrosion of the CR steel, a high concentration of chloride ion in the simulated pore solution was used in this paper. Details are listed in Table 2. The etching solution is prepared with vacuum filtration, using a 2.5 µm filter to ensure that the solution is free from suspensions.

Materials 2017, 10, 903

3 of 21

Table 2. Corrosive soaking solutions for Steel. Corrosive Solution

KOH

NaOH

Ca(OH)2

NaCl

Concentration (mol/L)

0.6

0.2

0.03

5

2.2. Measurement Methods 2.2.1. Electrochemical Measurements The size of the steel bars used for testing was Φ20 mm × 10 mm (diameter is 20 mm, length is 10 mm). The bottoms of the bars were used as the testing surfaces. Each test surface was polished stepwise with increasing grades (#200, #600, #1000, and #2000) of SiC sandpaper, and the polishing time was about 5 min for each kind of sandpaper. After washing with deionised water, the surface was polished to a mirror finish with 0.25 µm diamond polishing liquid. The bars were then cleaned, and the surface impurities were removed with alcohol. The samples were dried with a hair dryer and immediately placed into the etching tank. The bars were immersed in a simulated concrete pore solution with a 13.6 pH and 5 M chloride concentration for 0 h, 24 h, 60 h, 120 h, 200 h, and 240 h. The electrochemical test was then performed with the PARSTAT 4000 electrochemical workstation. A three-electrode measurement system was used; i.e., the reinforced bars were set as the working electrode, the saturated calomel electrode was set as the reference electrode, and the platinum electrode was the auxiliary electrode. The test was performed at room temperature (25 ± 1 ◦ C). All electrochemical tests were performed after the open potentials working electrode was stabilised. Linear Polarisation Resistance Method The linear polarisation resistance method is a common method for measuring the corrosion rate of steel bars [16]. In this study, the polarisation resistance of the bar surface is measured by adding a small polarisation potential near the corrosion potential. The corrosion current [17] is then calculated as follows: icorr = B/R p (1) where Rp is the polarisation resistance (kΩ·cm2 ), icorr is the corrosion current density (µA/cm2 ), and B is the Stern-Geary [18] constant (mV). For reinforcement bars in concrete, if the bars are corroded, B is 26 mV; in the passivated state, B is 52 mV. The range of the linear polarisation resistance is ±10 mV vs. Ecorr at a scanning speed of 10 mV/min. Electrochemical Impedance Spectroscopy Electrochemical impedance spectroscopy (EIS) is a frequency domain-based nondestructive test method, and it has been widely used to characterise the corrosion behaviour of steel bars [19,20]. A sinusoidal voltage excitation signal with a disturbance amplitude of 10 mV is used. The frequency range is 104 ~10−2 Hz. The kinetic information in the electrochemical process is obtained by measuring the periodic response of the corrosion system. 2.2.2. Micro-Measurements Steel bars measuring 10 mm × 10 mm × 5 mm were used in these tests. The surface was polished stepwise with increasing grades (#200, #600, #1000, and #2000) of SiC sandpaper. After washing with deionised water, the surface was polished to a mirror finish with a 0.25 µm diamond polishing liquid. The bars were then cleaned, and the surface impurities were removed with alcohol. The samples were dried with a hair dryer and immediately immersed into a simulated 5 M concrete pore solution for 0 h, 24 h, 60 h, 100 h, 120 h, and 200 h.

Materials 2017, 10, 903 Materials 2017, 10, 903

4 of 21 4 of 21

To perform micro-measurements, the samples werewas removed fromfrom the corrosive solution, the ethanol for 30the min for ultrasonic cleaning. The sample removed the solution and rinsed sample surface was washed with deionised water, and the sample was placed in anhydrous ethanol forThe with anhydrous ethanol, ultrasonically cleaned for another 30 min, then dried with a hair dryer. 30 min for ultrasonic cleaning. sample the solution and rinsed with anhydrous samples were placed onto aThe watch glasswas andremoved dried in from an oven. ethanol, ultrasonically cleaned for another 30 min, then dried with a hair dryer. The samples were placed onto aDigital watch glass and dried in an oven. Reflective Holographic Microscopy

A Lyncee (Lyncee Tec SA, Lausanne, Vaud, Switzerland) digital holographic microscope, a Reflective Digital Tec Holographic Microscopy type of reflective digital holographic microscope, with a view field of 5 mm, was used in this A Lyncee Tec Tecalong SA, Lausanne, Vaud, Switzerland) digital holographic microscope, experiment. The (Lyncee resolution the x- and y-axes is >300 nm, and the resolution along the z-axis is a type of reflective digital holographic microscope, with a view field of 5 mm, was used in this >0.6 nm. The surfaces of the samples etched for different durations was viewed with a lens of 20× experiment. The resolution along the xand y-axes is >300 nm, and the resolution along the z-axis magnification. The 3D morphology of the pitting corrosion in the same area was measured afteristhe >0.6 nm. The surfaces of the samples etchedexposure. for different was was viewed with amicroscopically. lens of 20× sample endured various times of corrosive The durations pitting process analysed magnification. The 3D morphology of the pitting corrosion in the same area was measured after the sample endured various times of corrosive Scanning Electron Microscopy (SEM) exposure. The pitting process was analysed microscopically. The scanning electron microscope used in this experiment is a field emission scanning electron Scanning Electron Microscopy (SEM) microscope (SIRION-100, FEI, Eindhoven, Noord-Brabant, The Netherlands). During the test, the The scanning electron microscope used in this experiment is a field emission scanning electron surface profiles of the samples were viewed under 10,000× and 30,000× magnifications. Elemental microscope (SIRION-100, FEI, Eindhoven, Noord-Brabant, The Netherlands). During the test, the analyses were performed both inside and outside of the pitting corrosion areas. surface profiles of the samples were viewed under 10,000× and 30,000× magnifications. Elemental analyses were performed both inside and outside of the pitting corrosion areas. 3. Experiments and Discussion 3. Experiments and Discussion 3.1. Electrochemical Test

100

CR LC

Rp/kΩ·cm2

80 60 40 20 0 0

40

80

120

160

Time/h (a)

200

240

icorr/μA·cm-2

3.1. Electrochemical Test 3.1.1. Linear Polarisation Curve 3.1.1. Linear Polarisation Curve The data in Figure 1 show that, after 24 h, the corrosion current density of the LC bars is as high asThe 20 μA·cm ; however, for that, the corrosion steel bars, it is only approximately data in−2Figure 1 show after 24 h,resistant the corrosion current density of the LC bars 0.2 is asμA·cm high −2. This the corrosion ofcorrosion both types of barssteel afterbars, beingit immersed in the corrosive solution as 20 µAindicates ·cm−2 ; however, for the resistant is only approximately 0.2 µA ·cm−2 .for −2 24indicates h. The corrosion current of the LC bars 2.3 μA·cmin after 240 h, and continues This the corrosion of density both types of bars afterreaches being immersed the corrosive solution for to −2 after On the contrary, the corrosion current the CR bars increases slowly, only reaching 24 increase. h. The corrosion current density of the LC barsdensity reachesof 2.3 µA ·cm 240 h, and continues to −2 approximately 1.0 μA·cm after 240 h, and remains afterwards. The decrease of icorrreaching is because increase. On the contrary, the corrosion current density stable of the CR bars increases slowly, only the material1.0 in µA solution to creating protective coating ondecrease the surface of is the LC bars, approximately ·cm−2 contributed after 240 h, and remainsastable afterwards. The of icorr because the in gradual increase of icorr to forcreating the LC bars after 120 h was on due tosurface the pitting corrosion the LC thewhile material solution contributed a protective coating the of the LC bars,ofwhile It should be of noted theLC current density cannot used corrosion to study the corrosion rate thebars. gradual increase icorr that for the bars after 120numbers h was due to thebe pitting of the LC bars. of the CR bars when pitting is seen. It should be noted that the current density numbers cannot be used to study the corrosion rate of the CR bars when pitting is seen. 22 20 18 16 14 12 10 8 6 4 2 0

CR LC

icorr=1.0 icorr=0.5

0

40

80

120

160

200

240

Time/h (b)

Figure 1. Linear polarisation resistance corrosion current density of two types of steel bars Figure 1. Linear polarisation resistance (a) (a) andand corrosion current density (b) (b) of two types of steel bars at different corrosive exposure times. at different corrosive exposure times.

Materials 2017, 10, 903 Materials 2017, 10, 903

5 of 21 5 of 21

3.1.2. Electrochemical Impedance Spectroscopy 3.1.2. Electrochemical Impedance Spectroscopy The data in Figure 2a show that before the samples have soaked for 60 h, the radius of the capacitive loop of the CR gradually, indicating the corrosion The data in Figure 2a steel showincreases that before the samples havethat soaked for 60 h, resistance the radiusincreases of the with soaking time. After 24 h of soaking, the high-frequency band is still a capacitive loop. The low capacitive loop of the CR steel increases gradually, indicating that the corrosion resistance increases frequency changes into athe straight line with a band slope is ofstill 45°,ai.e., the Warburg with soakingband, time.however, After 24 h of soaking, high-frequency capacitive loop.impedance The low with characteristics of diffusion. Thisaindicates thatwith the acorrosion of◦the is controlled by the frequency band, however, changes into straight line slope of 45 , i.e.,system the Warburg impedance diffusion of the oxygen process. After immersion for 60 h, the radius of the capacitive loop in the CR with characteristics of diffusion. This indicates that the corrosion of the system is controlled by the steel bars to decrease gradually, and thusfor the60 corrosion resistance decreases gradually. diffusion of begins the oxygen process. After immersion h, the radius of thealso capacitive loop in the All of the above demonstrate that the passivation film is formed in the CR steel bars even when the CR steel bars begins to decrease gradually, and thus the corrosion resistance also decreases gradually. corrosive ions the critical concentration The pitting Allconcentration of the above of demonstrate thatreaches the passivation film is formed in of thechloride CR steelions bars[21]. even when the is nucleated after approximately 24 h ofthe immersion, and the pitting corrosion starts develop concentration of corrosive ions reaches critical concentration of chloride ions [21].to The pittingafter is 60 h of immersion. nucleated after approximately 24 h of immersion, and the pitting corrosion starts to develop after 60 h of immersion.

-Zim/(Ω·cm2)

60k 50k

16.0k

(a)

LC-0 h LC-24 h LC-60 h LC-120 h LC-200 h LC-240 h

14.0k 12.0k 2

CR-0 h CR-24 h CR-60 h CR-120 h CR-200 h CR-240 h

-Zim/(Ω·cm )

70k

40k 30k 20k 10k

10.0k

(b)

8.0k 6.0k 4.0k 2.0k

0

0.0

0.0

20.0k

40.0k

60.0k

Zre/(Ω·cm2)

80.0k

100.0k

0.0

5.0k

10.0k

15.0k

20.0k

25.0k

Zre/(Ω·cm2)

Figure 2. Nyquist plot at different corrosive duration times: (a) the CR bars; and (b) the LC bars.

Figure 2. Nyquist plot at different corrosive duration times: (a) the CR bars; and (b) the LC bars.

In the case of the LC steel bars, the capacitive loop is extremely small after 24 h immersion in Inas the case of LC steel bars, the capacitive loop is extremely small after 24inh the immersion chloride shown in the Figure 2b. This suggests the immediate corrosion of the LC bar chloridein chlorideThus, as shown Figure 2b. This thefilm immediate corrosion the stable LC barininchloride the chloride solution. it caninbe deduced that suggests the passive of LC bars cannotofstay or solution. Thus, it can be deduced that the passive film of LC bars cannot stay stable in chloride even that there is no passive film formed on the LC bar. After 24 h of immersion, the capacitive loopor even that there noexposure passive film on This the LC bar. 24by h of capacitive loop becomes larger asisthe timeformed increases. may beAfter caused theimmersion, continuousthe development of becomes larger as the exposure time increases. This may be caused by the continuous development pitting corrosion. The corrosion products continue to increase attaching to the interface between the of pitting corrosion. The pore corrosion products increase to thebetween interface between steel bars and the concrete solution. Theircontinue presenceto hinders theattaching ion transport the steel the steel bars and the concrete pore solution. Their presence hinders the ion transport between the bars and the pore solution. Compared to the CR steel bars, the polarisation resistance of the LC bars is steel bars and the pore solution. Compared to the CR steel bars, the polarisation resistance of the lower, indicating that the corrosion resistance of the LC steel bars is significantly lower than that of theLC bars is in lower, indicating that the corrosion resistance of the LC steel bars is significantly lower than CR steel a chloride environment. that of the CR steel in a chloride environment. 3.2. Micro Measurements and Analyses 3.2. Micro Measurements and Analyses 3.2.1. Digital Holographic Microscopy Analysis 3.2.1. Digital Holographic Microscopy Analysis Figure 3 shows the surface morphology and the depth of the pits in carbon steel bars at different corrosive exposure times. It shows that at the and startthe of depth the test h),pits theinsteel surface smooth and Figure 3 shows the surface morphology of(0 the carbon steelisbars at different stable; no surface irregularities are present. soaking fortest 24 h(0inh),the corrosive solution, the steel corrosive exposure times. It shows that atAfter the start of the the steel surface is smooth and surface obvious signs of roughness. There aresoaking some small on the surface,solution, and the the depth stable;shows no surface irregularities are present. After for 24peaks h in the corrosive steel ofsurface the pitsshows is 50 nm. Aftersigns 60 h of immersion, the surface is further maximum depth obvious roughness. There are some small roughened. peaks on theThe surface, and the depth ofof the pits is is 300 the60 soaking duration reaches 100 h, pit isroughened. visible (as The shown in Figure 3). the pits 50nm. nm.As After h of immersion, the surface is the further maximum depth Itsofdepth is measured There are tworeaches peaks at theh,bottom of visible the pit,(as indicating new3). the pits is 300 nm.at As350 thenm. soaking duration 100 the pit is shown inthat Figure pitting corrosion is nucleated, further damaging the peaks bottomatofthe thebottom pits. After soaking for 120 h, pits Its depth is measured at 350 nm. There are two of the pit, indicating thatare new prominently visible.isThe depth offurther the pitsdamaging is approximately 500 of nm, indicates that, for at this pitting corrosion nucleated, the bottom thewhich pits. After soaking 120 time, h, pits pitting corrosion is visible. developing along the vertical are prominently The depth of the pits isdirection. approximately 500 nm, which indicates that, at this time, pitting corrosion is developing along the vertical direction.

Materials 2017, 10, 903

6 of 21

Materials 2017, 10, 903

6 of 21

(a)

(b)

(c)

(d) Figure 3. Cont.

Materials 2017, 10, 903

7 of 21

Materials 2017, 10, 903

7 of 21

(e)

(f)

(g) Figure 3. Surface morphology and pitting depth in carbon steel bars at different corrosive exposure

Figure 3. Surface morphology and steelmorphology bars at different corrosive exposure times. (a) Surface morphology forpitting LC at 0depth and 24in h;carbon (b) Surface for LC at 60 and 100 h; Surface morphology for LCfor at 120 Pitting LC at 24 h; (e) Pitting depth foratLC 60 h;100 h; times.(c)(a) Surface morphology LCh;at(d) 0 and 24depth h; (b)for Surface morphology for LC 60atand (f) Pitting depth for LC and forfor LCLC at 120 h. h; (OPL: Optical depth path length). (c) Surface morphology for at LC100 ath; 120 h;(g) (d)Pitting Pittingdepth depth at 24 (e) Pitting for LC at 60 h; (f) Pitting depth for LC at 100 h; and (g) Pitting depth for LC at 120 h. (OPL: Optical path length). Figure 4 shows the surface topography, height, and depth of the pits in the corrosion resistant steel bars at different corrosive exposure times. The pitting depth of the CR bar increases with Figure 4 shows surface topography, height, and depth the pits in the immersion time,the which is similar to the LC bar. However, theofpitting depth of corrosion the CR barresistant is much steel bars at different corrosive exposure times. The pitting depth of the CR bar increases with immersion smaller than that of the LC bar before 100 h of immersion, while the pitting depth of the CR bar time,sharply which increases is similartoto300 thenm LC bar.120 However, the pitting depth it ofshould the CRbebar is much smaller after h of immersion. In addition, noted that the pitting than width of the bar is 100 significantly smaller than thatthe of the LC bar during immersion period. to that of the LC barCR before h of immersion, while pitting depth of the thewhole CR bar sharply increases These combined with the results the electrochemical show that the CR barsCR arebar is 300 nm after 120results, h of immersion. In addition, it of should be noted thattest, the pitting width of the passivated at 0 h. The pitting corrosion is nucleated at approximately 24 h, and starts to develop at significantly smaller than that of the LC bar during the whole immersion period. 60 h. According to the literature [22], under the attack of corrosive chloride ions, the passivation film These results, combined with the results of the electrochemical test, show that the CR bars are possibly breaks around the inclusions in the CR bars, thus creating pits with curvature radii of dozens passivated at 0 h. The pitting corrosion is nucleated at approximately 24 h, and starts to develop at of nanometres. The reason for this phenomenon is that these inclusions are favourable locations for 60 h. pitting According to the literature [22], under the attack of corrosive chloride ions, the passivation film initiation [23].

possibly breaks around the inclusions in the CR bars, thus creating pits with curvature radii of dozens of nanometres. The reason for this phenomenon is that these inclusions are favourable locations for pitting initiation [23].

Materials 2017, 10, 903

8 of 21

Materials 2017, 10, 903

8 of 21

(a)

(b)

(c)

(d) Figure 4. Cont.

Materials 2017, 10, 903

9 of 21

Materials 2017, 10, 903

9 of 21

(e)

(f)

(g) Figure 4. Surface morphology and height and depth of pits in corrosion resistant alloy steel bars at

Figure 4. Surface morphology height and depth of pits in steel bars at different corrosive exposure and times. (a) Surface morphology forcorrosion CR at 0 resistant and 24 h;alloy (b) Surface different corrosive times. Surface morphology andh;24 (b) Surface morphology morphology forexposure CR at 60 and 100(a) h; (c) Surface morphologyfor forCR CRatat0120 (d)h;Height and depth of pits for CR at 24 h; (e) Height and depth of pits for CR at 60 h; (f) Height and depth of pits for CR for CR at 60 and 100 h; (c) Surface morphology for CR at 120 h; (d) Height and depth of pitsatfor CR (g) Height and depth of pits h. Height and depth of pits for CR at 100 h; and at 24100 h; h; (e)and Height and depth of pits forfor CRCR at at 60120 h; (f) (g) Height and depth of pits for CR at 120 h. 3.2.2. Scanning Electron Microscopy Analysis

Figure 5Electron shows photos taken with a scanning electron microscope (10,000× magnification) of the 3.2.2. Scanning Microscopy Analysis LC samples at different corrosion immersion times. The data show that when the test starts (0 h), the Figure is5 shows photos taken with a scanning electron microscope (10,000 × magnification) surface flat without corrosion pits. After immersion for 24 h, round pits of nano-scale diameters of the LC samples at different times. datain show that when thefor test (0 h), the appear distributed oncorrosion the surfaceimmersion of the LC bars. AfterThe soaking corrosive solutions 60 starts h, regular surface is flat without pits. After immersion for 24the h, round pits of of nano-scale diameters circular holes of 1 corrosion μm in diameter appear on the surface; vast majority the uniformly distributed holeson of the nano-scale formed h of immersion and disappeared after 60 regular h appear distributed surfacediameter of the LC bars. after After24soaking in corrosive solutions for 60 h, of immersion. Afterinimmersion for 100 on h, large areas ofthe irregular pitting corrosion appear on the circular holes of 1 µm diameter appear the surface; vast majority of the uniformly distributed A large pit is created by the after connection and further growth of two adjacent pits.60 The bottoms holessurface. of nano-scale diameter formed 24 h of immersion and disappeared after h of immersion. of the pits have stepped contours. After immersion for 120 h, in addition to some large holes, small After immersion for 100 h, large areas of irregular pitting corrosion appear on the surface. A large holes of nano-scale diameters are distributed around the large holes. After immersion for 200 h, large

pit is created by the connection and further growth of two adjacent pits. The bottoms of the pits have stepped contours. After immersion for 120 h, in addition to some large holes, small holes of nano-scale diameters are distributed around the large holes. After immersion for 200 h, large holes

Materials 2017, 10, 903

10 of 21

Materials 2017, 10, 903

10 of 21

of irregular shapes appear on the surface. The percentage of large holes is high. The data show that after holes soaking for 120 shapes h in the corrosive pitting corrosion hasholes developed the surfaces of irregular appear on thesolution, surface. The percentage of large is high. across The data show soaking 120 h corrosion in the corrosive solution, corrosion has developed acrosscorrosion. the of thethat LCafter samples. Theforpitting kind for the CRpitting steel bar was steady-state pitting surfaces ofthe thereinforcement LC samples. The pitting corrosion kind for the CR steel bar was steady-state Additionally, bar generally experiences a high corrosion rate, which ispitting consistent Additionally, reinforcement bar generally experiences a high corrosion rate, which is with corrosion. the electrochemical testthe results. consistent with the electrochemical test results.

(a)

(b)

(c) Figure 5. Scanning electron microscope photos (10,000× magnification) of carbon steel bars at different

Figure 5. Scanning electron microscope photos (10,000× magnification) of carbon steel bars at different corrosive exposure times. (a) Micrograph of LC at 0 and 24 h; (b) Micrograph of LC at 60 and 100 h; corrosive exposure times. (a) Micrograph of LC at 0 and 24 h; (b) Micrograph of LC at 60 and 100 h; and (c) Micrograph of LC at 120 and 200 h. and (c) Micrograph of LC at 120 and 200 h.

Materials 2017, 10, 903

11 of 21

The SEM photos (30,000× magnification) of the LC bars at different corrosive exposure times Materials 2017, 10, 903 11 of 21 (see Figure 6) show that the pitting corrosion starts as stepped pits, then develops layer by layer until reaching the of the pit. magnification) of the LC bars at different corrosive exposure times (see Thebottom SEM photos (30,000× The scanning microscope photos × magnification) in Figure showuntil the CR Figure 6) show electron that the pitting corrosion starts (10,000 as stepped pits, then develops layer by7layer samples at different corrosive exposure times. The data show that in the beginning (0 h), the surface is reaching the bottom of the pit. The scanning electron photos (10,000× magnification) in Figure 7 show the for CR 24 h, smooth compared to that of themicroscope LC bars, and there is no pitting corrosion. After immersion samples at different corrosive times. The dataappear show that beginning (0 h), surface a small number of corrosion pits ofexposure nano-scale diameters on in thethe surface, mostly inthe regular, round is smooth compared to that of the LC bars, and there is no pitting corrosion. After immersion for shapes. After immersion for 60 h, individual large pits appear; the largest diameter is approximately 24 h, a small number of corrosion pits of nano-scale diameters appear on the surface, mostly in 1 µm. However, the number of small pits, i.e., pits of nano-scale diameter, in the areas surrounding the regular, round shapes. After immersion for 60 h, individual large pits appear; the largest diameter is large pits decreases. This result suggests that the nano-scale pits are possibly re-passivation, resulting approximately 1 μm. However, the number of small pits, i.e., pits of nano-scale diameter, in the areas in thesurrounding remaining good corrosion resistance ofresult the CR bar, which is consistent the possibly electrochemical the large pits decreases. This suggests that the nano-scalewith pits are reresults. After immersion for 100 h, individual large corrosion pits appear on the surface. Some small passivation, resulting in the remaining good corrosion resistance of the CR bar, which is consistent pits, with i.e., pits of nano-scaleresults. diameter, appear. immersion for corrosion 120 h, some micron-scale the electrochemical Afteralso immersion for After 100 h, individual large pits appear on the surface. Some small pits, i.e., pitswidely of nano-scale diameter, also appear. Afterpits. immersion for 120 h, for corrosion pits appear, with some more distributed smaller nano-scale After immersion micron-scale corrosion pits appear, some more widely distributed smaller pits.it was 200 h,some the size of the pits remains the same;with however, the number of nano-scale pitsnano-scale is less than After immersion for 200 h, the size of the pits remains the same; however, the number of nano-scale at 120 h. Pistorius et al. [24] showed that the development of pitting corrosion includes pitting nuclei, pits is less thancorrosion, it was at 120 h. Pistoriuspitting et al. [24] showed that development of or pitting corrosion metastable pitting steady-state corrosion, and the growth stoppage, the re-passivation includes pitting nuclei, metastable pitting corrosion, steady-state pitting corrosion, and growth of pitting corrosion pits known as metastable pitting corrosion. Metastable pitting corrosion occurs stoppage, or the re-passivation of pitting corrosion pits known as metastable pitting corrosion. before the formation of steady-state pitting corrosion. However, the mechanism of the formation of Metastable pitting corrosion occurs before the formation of steady-state pitting corrosion. However, metastable pitting corrosion is still of unclear [25].pitting corrosion is still unclear [25]. the mechanism of the formation metastable

(a)

(b) Figure 6. Cont.

Materials 2017, 10, 903

12 of 21

Materials 2017, 10, 903

12 of 21

Materials 2017, 10, 903

12 of 21

(c) Figure 6. Scanning electron microscope photos (30,000× (c) magnification) of carbon steel bars at different Figure 6. Scanning electron photos (30,000 × magnification) of carbon different corrosive exposure times.microscope (a) Micrograph of LC at 0 and 24 h; (b) Micrograph of LCsteel at 60bars and at 100 h; Figure 6. Scanning electron microscope photos (30,000× magnification) of carbon steel bars at different corrosive exposure times. (a)atMicrograph and (c) Micrograph of LC 120 and 200 h.of LC at 0 and 24 h; (b) Micrograph of LC at 60 and 100 h; corrosive exposure times. (a) Micrograph of LC at 0 and 24 h; (b) Micrograph of LC at 60 and 100 h; and (c) Micrograph of LC at 120 and 200 h. and (c) Micrograph of LC at 120 and 200 h.

(a) (a)

(b) (b) Figure 7. Cont.

Materials 2017, 10, 903

13 of 21

Materials 2017, 10, 903

13 of 21

(c) Figure 7. Scanning electron microscope photos (10,000× magnification) of corrosion-resistant alloy

Figure 7. bars Scanning electron microscope (10,000 × magnification) steel at different corrosive exposurephotos times. (a) Micrograph of CR at 0 andof24corrosion-resistant h; (b) Micrograph ofalloy steel bars exposure times. Micrograph CR atat 60different and 100 h;corrosive and (c) Micrograph of CR at(a) 120 and 200 h. of CR at 0 and 24 h; (b) Micrograph of CR at 60 and 100 h; and (c) Micrograph of CR at 120 and 200 h. The results show that in the simulated concrete pore solution with a Cl− concentration of 5 mol/L, before the immersion duration reaches 200 h, the maximum diameter of the corrosion pits in the CR The results show that in the simulated concrete pore solution with a Cl− concentration of 5 mol/L, samples is 1 μm. The steel bar passes through the following periodic process under chloride attack. before thepassivation immersion duration reaches h, the diameter of nano-scale the corrosion pits in the CR The film is roughened and 200 the pits aremaximum nucleated, which creates corrosion pits. samples is of 1 µm. The steel pitting bar passes through following periodicmetastable process under attack. Some the nano-scale corrosion pits the are unstable, exhibiting pittingchloride corrosion. The passivation film the pits and are nucleated, creates corrosion Some portion of is theroughened pits is thenand passivated disappears, which while the rest nano-scale of the nano-scale pits pits. further into pitting micron-scale pits. A dynamic balance exhibiting is maintained when the pitting pits grow to Somedevelop of the nano-scale corrosion pits are unstable, metastable corrosion. approximately 1 μm. During the entire periodic process, in the initial period, i.e., during the Some portion of the pits is then passivated and disappears, while the rest of the nano-scale pits develop immersion period of 24 h, 60 and 100 balance h, there are few pitting nuclei pits. further into micron-scale pits. A hdynamic is maintained when or themetastable pits growcorrosion to approximately This is because, by this time, the surface of the CR steel is still passivated because it is not prone to 1 µm. During the entire periodic process, in the initial period, i.e., during the immersion period of pitting corrosion. During the period of 100 h, 120 h and 200 h, the quantity of metastable pitting 24 h, 60 h and 100 h, there are few pitting nuclei or metastable corrosion pits. This is because, by this corrosion increases since the accumulation of chloride ions on the surface of the CR bar reaches a time,certain the surface of the CR steel is still passivated because it is not prone to pitting corrosion. During level. This is consistent with the results from the macro electrochemical test. The corrosion the period 120 h and 200they h, the quantity of metastable corrosion since the rate of of the100 CR h, bars is high after have been immersed for 120 h.pitting The corrosion rateincreases of the CR steel accumulation of120 chloride ions on theofsurface theduring CR bar reaches a certain level.the This is consistent decreases at h. The adsorption chlorideof ions pitting nucleation damages passivation with the from electrochemical test. The corrosion of the CR is highpores after they film,results resulting in the macro formation of nano-scale pores. Under certainrate conditions, thebars nano-scale to grow, for and120 theh. pitting corrosion enters development forming have continue been immersed The corrosion rate ofthe thepitting CR steel decreasesstage, at 120 h. Themicronadsorption scale corrosion pits. pitting nucleation damages the passivation film, resulting in the formation of chloride ions during The scanning electron microscope photos magnification) of the to CRgrow, bars at different of nano-scale pores. Under certain conditions, the(30,000× nano-scale pores continue and the pitting corrosive exposure times (see Figure 8) show that, in CR steel, the shape of the pitting corrosion pits corrosion enters the pitting development stage, forming micron-scale corrosion pits. is primarily a regular circle. Each individual pit is independent of the others. Pits generally extend to The scanning electron microscope photos (30,000× magnification) of the CR bars at different the bottom. This is different from the evolving pattern of pitting corrosion in the LC bars.

corrosive exposure times (see Figure 8) show that, in CR steel, the shape of the pitting corrosion pits is primarily a regular circle. Each individual pit is independent of the others. Pits generally extend to the bottom. This is different from the evolving pattern of pitting corrosion in the LC bars.

Materials 2017, 10, 903

14 of 21

Materials 2017, 10, 903

14 of 21

(a)

(b)

(c) Figure 8. Scanning electron microscope photos (30,000× magnification) of corrosion-resistant alloy Figure 8. Scanning electron microscope photos (30,000× magnification) of corrosion-resistant alloy steel bars at different corrosive exposure times. (a) Micrograph of CR at 0 and 24 h; (b) Micrograph of steel bars at different corrosive exposure times. (a) Micrograph of CR at 0 and 24 h; (b) Micrograph of CR at 60 and 100 h; and (c) Micrograph of CR at 120 and 200 h. CR at 60 and 100 h; and (c) Micrograph of CR at 120 and 200 h.

Materials 2017, 10, 903

15 of 21

Materials 2017, 10, 903

15 of 21

3.2.3. Analysis Analysis of of the the Pitting Pitting Corrosion Corrosion Process 3.2.3. Process in in Corrosion-Resistant Corrosion-Resistant Alloy Alloy Steel Steel Bars Bars To analyse process To analyse the the effect effect of of the the chemical chemical components components on on the the pitting pitting corrosion corrosion development development process in CR bars, the chemical compositions inside and outside of the corrosion pits versus the immersion in CR bars, the chemical compositions inside and outside of the corrosion pits versus the immersion duration are dispersive spectroscopy (EDS) analysis. It should be noted that duration are analysed analysedusing usingananenergy energy dispersive spectroscopy (EDS) analysis. It should be noted the correction method is ZAF in EDS analysis. The results are shown in Figure 9. that the correction method is ZAF in EDS analysis. The results are shown in Figure 9.

(a)

(b)

(c) Figure 9. Cont.

Materials 2017, 10, 903

16 of 21

Materials 2017, 10, 903

16 of 21

(d)

(e)

(f) Figure 9. Energy dispersive spectroscopy (EDS) photos of corrosion-resistant alloy reinforcement bars Figure 9. Energy dispersive spectroscopy (EDS) photos of corrosion-resistant alloy reinforcement bars at different corrosion exposure times. (a) EDS result of CR at 0 h; (b) EDS result of CR at 24 h; (c) EDS at different corrosion exposure times. (a) EDS result of CR at 0 h; (b) EDS result of CR at 24 h; (c) EDS result of CR at 60 h; (d) EDS result of CR at 100 h; (e) EDS result of CR at 120 h; and (f) EDS result of result of CR CR at 200 h. at 60 h; (d) EDS result of CR at 100 h; (e) EDS result of CR at 120 h; and (f) EDS result of CR at 200 h.

A determination of the amount of the elements O, Fe, Cr, and Mn, both inside and outside of the corrosion pits in CR bars at amount differentofcorrosion exposure is Mn, obtained by analysing the results A determination of the the elements O, Fe,times, Cr, and both inside and outside of the shown in Figure The details are shown in Figures 10 and 11. is obtained by analysing the results corrosion pits in 9. CR bars at different corrosion exposure times, shown in Figure 9. The details are shown in Figures 10 and 11.

O O-pit

10

Fe Fe-pit 80

8

Wt/%

Wt/%

60

6 Materials Materials2017, 2017,10, 10,903 903

17 17 of of 21 21 40

4 12

O O-pit

2 10

Wt/%

8

0

24

60

100 120

80 0

200

0

24

60

100 120

200

Time/h

Time/h

60

(a)

6

Fe Fe-pit

(b)

Wt/%

0

100 20

40 Figure 10. Analysis of the chemical elements inside and outside of the corrosion pits at different 4 corrosion exposure times in corrosion-resistant alloy steel bars. (O-pit represents the O content inside 20 the pits, 2 Fe-pit represents the Fe content inside the pits). (a) O element contained in inside and outside of the corrosion pits; and (b) Fe element contained in inside and outside of the corrosion pits.

0

0

24

60

100 120

0

200

0

24

60

100 120

200

The data in Figure 10 that there is more O inside the corrosionTime/h pits than outside of the pits, (a) the pits than outside of the pits. This is in(b) whereas there is less Fe inside agreement with the results of Figueira et al. [26]. The reason for the oxygen distribution is that the corrosion products generated Figure 10. 10. Analysis Analysis of of the the chemical chemical elements elements inside inside and and outside outside of of the the corrosion corrosion pits pits at at different different Figure inside the corrosion pits contain oxygen, whilstalloy thesteel areabars. outside ofrepresents the pits the is in the passive state. corrosion exposure times in corrosion-resistant (O-pit O content inside corrosion exposure times in corrosion-resistant alloy steel bars. (O-pit represents the O content inside Hence, the oxygen content inside the pits is high. In addition, the corrosion reaction occurs inside the pits, pits, Fe-pit Fe-pit represents representsthe theFe Fecontent contentinside insidethe thepits). pits). (a) (a) O O element element contained contained in in inside inside and and outside outside the the pits. Therefore, the iron is oxidised into corrosive which are of found inside the pits. of the corrosion pits; and (b) Fe element containedproducts, in inside and outside the corrosion pits. Time/h show

of the corrosion pits; and (b) Fe element contained in inside and outside of the corrosion pits.

Wt/%

Wt/%

12 The 45 data in Figure 10 show that there is more O inside the corrosion pits than outside of the pits, Cr Mn whereas 40 there is less Fe inside the pits than outside of the pits. This is in agreementCr-pit with the results Mn-pit 10 of Figueira 35 et al. [26]. The reason for the oxygen distribution is that the corrosion products generated inside the30 corrosion pits contain oxygen, whilst the area outside of the pits is in the passive state. 8 Hence, the oxygen content inside the pits is high. In addition, the corrosion reaction occurs inside the 25 6 pits. Therefore, the iron is oxidised into corrosive products, which are found inside the pits. 20

15

4

45 10

12

Mn Mn-pit

40 5

Wt/%

30

0

0

24

60

25

100 120

Time/h

200

(a)

20

0

24

60

8

Wt/%

35 0

Cr Cr-pit

2 10 100 120

200

Time/h

(b)

6

Figure 11. 11. Contents of chemical inside and and outside outside of the pits at at different corrosion 4 Figure Contents of chemical elements elements inside of the corrosion corrosion pits different corrosion 15 exposure times times in in corrosion-resistant corrosion-resistant alloy alloy steel steel bars. bars. (Mn-pit (Mn-pit represents represents Mn Mn content content inside inside the the pits, pits, exposure 10 2 Cr-pit represents representsthe theCr Cr content contentinside insidethe thepits). pits). (a) (a) Mn Mn element element contained in in inside inside and and outside outside of of the the Cr-pit contained 5 corrosion pits; and (b) Cr element contained in inside and outside of the corrosion pits. corrosion pits; and (b) Cr element contained in inside and outside of the corrosion pits. 0

0

0

24

60

100 120

200

0

24

60

100 120

200

Time/h Time/h The data in Figure 11 show that the quantity of Mn inside the corrosion pits is much higher than The data in Figure 10 show that there is more O inside the corrosion pits than outside of the pits, (a) is because the pitting corrosion occurs at the (b)Mn inclusions during the that outside of the pits. This whereas there is less Fe inside the pits than outside of the pits. This is in agreement with the results nucleation period [27], resulting in pitting corrosion. Generally, the pitting corrosion formed in these of Figueira [26]. The reason for the oxygen distribution that the corrosion products generated Figure et 11.al. Contents of chemical elements inside and outside ofis the corrosion pits at different corrosion locations will continue to develop into steady-state pitting. The quantity of Cr inside the pits is low times inpits corrosion-resistant alloy steel bars. (Mn-pit represents Mn content the pits,state. insideexposure the corrosion contain oxygen, whilst the area outside of the pits is in inside the passive for the following reasons. On one hand, pits). according to the contained literaturein[28], Crand is outside involved in the Cr-pit the Cr content inside element inside of the Hence, the represents oxygen content inside the pitsthe is high.(a) InMn addition, the corrosion reaction occurs inside the formation of passive films, thereby resulting in a film that consists of an inner layer that contains Cr– corrosion pits; elementinto contained in inside and outside thefound corrosion pits.the pits. pits. Therefore, theand iron(b) is Cr oxidised corrosive products, whichofare inside Fe oxides and an outer layer that contains Fe oxides, whose thickness presents a slight increase as the The data in Figure 11 show that the quantity of Mn inside the corrosion pits is much higher than content Cr increases. Crshow can effectively enhance the steel’s to corrosion. Therefore, Theofdata Figure that thethe quantity MnCR inside the resistance corrosion pits is much higher that outside ofin the pits. 11 This is because pittingofcorrosion occurs at the Mn inclusions duringthan the pitting occurs in locations where the Cr content is low. On the other hand, in the electrochemical that outside of the pits. This is because the pitting corrosion occurs at the Mn inclusions during the nucleation period [27], resulting in pitting corrosion. Generally, the pitting corrosion formed in these nucleation period [27], resulting in pitting corrosion. Generally, the pitting corrosion formed in these locations will continue to develop into steady-state pitting. The quantity of Cr inside the pits is low for locations will reasons. continueOn to develop into steady-state pitting. The[28], quantity Cr inside the formation pits is low the following one hand, according to the literature Cr is of involved in the for the following reasons. On one hand, according to the literature [28], Cr is involved in the of passive films, thereby resulting in a film that consists of an inner layer that contains Cr–Fe oxides formation of passive films, thereby resulting in a film that consists of an inner layer that contains Cr– and an outer layer that contains Fe oxides, whose thickness presents a slight increase as the content FeCr oxides and anCr outer that contains oxides, whoseresistance thicknessto presents a slight increasepitting as the of increases. can layer effectively enhanceFethe CR steel’s corrosion. Therefore, content of Cr increases. Cr can effectively enhance the CR steel’s resistance to corrosion. Therefore, occurs in locations where the Cr content is low. On the other hand, in the electrochemical corrosion pitting occurs in locations where the Cr content is low. On the other hand, in the electrochemical

Materials 2017, 10, 903

18 of 21

process involved during the development of pitting corrosion, the chromium (Cr) ions inside the pits are not dissolved as the iron ion is, but rather, they remain along the internal walls of the pits. Hence, the Cr content inside the pits is low. As demonstrated by the pitting corrosion model, the stable development of the corrosion pits is closely related to the chemical elements at the bottom of the pits. Iron, chromium, and molybdenum are dissolved by oxidation, and the metal ions are created inside the corrosion pits. These ions are hydrolysed, and then form hydroxide and hydroxyl chloride complexes [29]. A microscopic analysis shows that the steady-state corrosion pitting in CR bars has a generally circular shape, and the pitting develops longitudinally in the direction of gravity. The present study shows that pitting is caused by local damage of the passivation film. The damaged passivation film is the anode, while the intact passivation film is the cathode. Once the pits are formed, the surface of the CR bar in the area of the pitting is in a localised active state (the potential is low). The anode dissolves in the pits, generating the elements Fe2+ , Cr3+ , etc. Most of the steel surface outside the pits is still in the passivated state (with high potential). The cathode reduction reaction takes place on the outer surfaces of the adjacent pit via O2 + 2H2 O + 4e− → 4OH− . Therefore, the activation/passivation battery cell composed of a small anode and large cathode is formed inside and outside the pits, which accelerates the pitting corrosion. Simultaneously, in strongly alkaline conditions, there is a large quantity of OH− ions in the solution. Once Fe2+ ions precipitate from the pits, they will react with OH− to produce Fe(OH)2 . As the solution is not deoxygenated, the generated Fe(OH)2 rapidly reacts with the dissolved O2 in the solution to form Fe(OH)3 . The compound Fe(OH)3 decomposes rapidly to form Fe2 O3 . In addition, because of the high Cr content in the CR bars, Cr ions will participate in the reaction, generating Cr2 O3 . In this process, the oxygen diffusion is the controlling factor. The corrosion products generated are quickly adsorbed around the pits, creating an occlusion zone. The oxygen is depleted inside the pits, while outside the pits, the oxygen is rich. This creates an oxygen concentration battery cell. The main reason for the slow pitting of the CR bar during the first 60 h is possibly the inhabitation of Cr around the MnS inclusion. The redox corrosion reaction is difficult to produce in these areas, and leads to the pitting of the lower width, which inhibits the transport of ions into pits. Thus, the corrosion reaction in the pit is reduced. Some of the pits are repassive when the passive reaction rate is higher than the corrosion reaction rate, while other pits slowly continue to develop. The above analysis shows that, during pitting corrosion, the activation/passivation battery, and the oxygen concentration by the battery, occur inside and outside of the corrosion pits in the CR bars. The development of pitting corrosion in the CR bars is presented in Figure 12. As shown in Figure 12, coupled with the gravity, the pit has a deep excavating capacity. During the pitting corrosion development, the dissolved chromium ions are not all diffused and hydrolysed as the Fe2+ ions are. On the contrary, some of the chromium ions are re-deposited onto the internal walls of the pits. After discharging, the Cr ions deposit inside the pits, creating a surface layer of high chromium content. Therefore, a dissolution reaction occurs primarily at the bottom of the pit, and pitting corrosion develops a more regular shape as it develops vertically. In addition, a large area outside of the corrosion pit is in the passivated cathode state, further accelerating the metal ionisation process inside the pits. However, because the corrosion products are deposited in the openings of the pits, the passivation areas around the pit openings are cathodically protected by electrons released from the anodic process. This greatly suppresses the corrosion around the pits. For this reason, the pitting corrosion in the CR bars appears as the distribution of a single pit.

Materials 2017, 10, 903 Materials 2017, 10, 903

19 of 21 19 of 21

Figure 12. Development of pitting corrosion pits exposed to corrosive media in corrosion-resistant alloy steel Figure 12. bars. Development of pitting corrosion pits exposed to corrosive media in corrosion-resistant alloy steel bars.

4. Conclusions 4. Conclusions In this study, the pitting corrosion behaviours of LC bars and CR bars exposed to a simulated In this study, the pitting corrosion behaviours electrochemical of LC bars and CR exposed analyses to a simulated concrete pore solution are compared. Macroscopic andbars microscopic show concrete porehas solution are compared. Macroscopic andprimary microscopic analyses show that CR steel better resistance to pitting corrosionelectrochemical than LC steel. The reasons for the better that CR steel has better resistance to pitting corrosion than LC steel. The primary reasons for the performance of the CR steel are as follows: better performance of the CR steel are as follows: (1) CR steel has a high passivation capability. Compared with the polished LC samples that are (1) CR steel has a high passivation capability. Compared with the polished LC samples that are immersed into the corrosive solution, a passivation film is formed on the surfaces of the CR immersed into the corrosive solution, a passivation film is formed on the surfaces of the CR bars. bars. This film hinders the initiation of pitting. In LC steel, however, pitting corrosion occurs This film hinders the initiation of pitting. In LC steel, however, pitting corrosion occurs immediately, resulting in steady-state pitting corrosion. immediately, resulting in steady-state pitting corrosion. (2) The corrosion pits in CR bars are mostly individually developed shallow pits of circular shape. (2) The corrosion pits in CR bars are mostly individually developed shallow pits of circular shape. The corrosion pits in the LC samples, however, are large, irregular pits, created by the connection The corrosion pits in the LC samples, however, are large, irregular pits, created by the connection of adjacent pits. The corrosion pits in CR bars are regularly shaped micro-nanometre-scale of adjacent pits. The corrosion pits in CR bars are regularly shaped micro-nanometre-scale pits. pits. The pits are primarily caused by −Cl− absorption that compromises the passivation film. The pits are primarily caused by Cl absorption that compromises the passivation film. Compared to the rapid development of corrosion pits and the large pit size in LC steel samples, Compared to the rapid development of corrosion pits and the large pit size in LC steel samples, the corrosion pits in the CR bars develop slowly. Hence, the CR bars have great advantages over the corrosion pits in the CR bars develop slowly. Hence, the CR bars have great advantages over LC bars in controlling hazards caused by perforation corrosion, which is the result of uncontrolled LC bars in controlling hazards caused by perforation corrosion, which is the result of pitting corrosion development. uncontrolled pitting corrosion development. (3) The formation of metastable pitting corrosion greatly alleviates the pitting growth rate in CR bars. (3) The formation of metastable pitting corrosion greatly alleviates the pitting growth rate in CR Pitting corrosion develops immediately into steady-state pitting after pitting nucleation in LC bars. Pitting corrosion develops immediately into steady-state pitting after pitting nucleation in bars; however, the CR bars have a significant metastable pitting stage that occurs periodically. LC bars; however, the CR bars have a significant metastable pitting stage that occurs This is because the corrosion pits in the CR bars are shallow and the corrosion current density is periodically. This is because the corrosion pits in the CR bars are shallow and the corrosion small. Hence, closed pits are easily formed. For all of these reasons, the pitting corrosion at the current density is small. Hence, closed pits are easily formed. For all of these reasons, the pitting metastable pitting stage is passivated and then disappears. corrosion at the metastable pitting stage is passivated and then disappears. Author Contributions: Wei Sun introduced initial research ideas, andand the Author Contributions: Jin-yang Jin-yangJiang, Jiang,Hong-yan Hong-yanChu, Chu,and and Wei Sun introduced initial research ideas, research ideas were improved by by thethe co-authors; Danqian Wang the research ideas were improved co-authors; Danqian Wangand andYao YaoLiu Liu performed performed the the experiments; experiments; Danqian Wang Hong-yan Chu Chuanalysed analysedthe theexperimental experimental data; Han provided reagents, materials, Wang and Hong-yan data; Han MaMa provided reagents, materials, and and analysis tools; Jin-yang Jiang, Danqian Wang, and Hong-yan Chu wrotethe thepaper, paper,and andall all of of the the co-authors analysis tools; Jin-yang Jiang, Danqian Wang, and Hong-yan Chu wrote co-authors revised the manuscript. revised the manuscript. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.

Materials 2017, 10, 903

20 of 21

References 1. 2. 3.

4.

5.

6.

7.

8. 9. 10. 11. 12. 13.

14. 15. 16. 17. 18. 19. 20.

21.

Burstein, G.T.; Soouto, R.M.; Liu, C.; Vines, S.P. New processes for passive metal surfaces. In Proceedings of the Joint Conference of the Australasian Corrosion Association, Melbourne, Australia, January 2003. Tan, Y.T.; Wijesinghe, S.L.; Blackwood, D.J. The inhibitive effect of bicarbonate and carbonate ions on carbon steel in simulated concrete pore solution. Corros. Sci. 2014, 88, 152–160. [CrossRef] Wang, Y.; Cheng, G.; Wu, W.; Qiao, Q.; Li, Y.; Li, X.F. Effect of pH and chloride on the micro-mechanism of pitting corrosion for high strength pipeline steel in aerated NaCl solutions. Appl. Surf. Sci. 2015, 349, 746–756. [CrossRef] Bautista, A.; Alvarez, S.M.; Paredes, E.C.; Velasco, F.; Guzman, S. Corrugated stainless steels embedded in carbonated mortars with and without chlorides: 9-Year corrosion results. Constr. Build. Mater. 2015, 95, 186–196. [CrossRef] Shi, J.; Sun, W.; Jiang, J.; Zhang, Y. Influence of chloride concentration and pre-passivation on the pitting corrosion resistance of low-alloy reinforcing steel in simulated concrete pore solution. Constr. Build. Mater. 2016, 111, 805–813. [CrossRef] Williams, D.E.; Mohiuddin, T.F.; Zhu, Y.Y. Elucidation of a trigger mechanism for pitting corrosion of stainless steels using submicron resolution scanning electrochemical and photoelectrochemical microscopy. Electrochem. Soc. 1998, 145, 2664–2672. [CrossRef] Vuillemin, B.; Philippe, X.; Oltra, R.; Vignal, V.; Coudreuse, L.; Dufour, L.C.; Finot, E. SVET, AFM and AES study of pitting corrosion initiated on MnS inclusions by microinjection. Corros. Sci. 2003, 45, 1143–1159. [CrossRef] Ameri, M.A.; Yi, Y.; Cho, P.; Al Saadi, S.; Jang, C.; Beeley, P. Critical conditions for pit initiation and growth of austenitic stainless steels. Corros. Sci. 2015, 92, 209–216. [CrossRef] ˇ Kocijan, A.; Crtomir, D.; Jenko, M. Electrochemical and XPS studies of the passive film formed on stainless steels in borate buffer and chloride solutions. Corros. Sci. 2007, 49, 2083–2098. [CrossRef] Natishan, P.M.; O’Grady, W.E.; Martin, F.J.; Rayne, R.J.; Kahn, H.; Heuer, A.H. The Effect of Chloride on Passive Oxide Film Breakdown on Stainless Steels and Aluminum. ECS Trans. 2012, 41, 49–57. Ernst, P.; Newman, R.C. Pit growth studies in stainless steel foils. I. Introduction and pit growth kinetics. Corros. Sci. 2002, 44, 927–941. Laycock, N.J.; Noh, J.S.; White, S.P.; Krouse, D.P. Computer simulation of pitting potential measurements. Corros. Sci. 2005, 47, 3140–3177. [CrossRef] Vignal, V.; Krawiec, H.; Heintz, O.; Oltra, R. The use of local electrochemical probes and surface analysis methods to study the electrochemical behaviour and pitting corrosion of stainless steels. Electrochim. Acta 2007, 52, 4994–5001. [CrossRef] Saremi, M.; Mahallati, E. A study on chloride-induced depassivation of mild steel in simulated concrete pore solution. Cem. Concr. Res. 2002, 32, 1915–1921. [CrossRef] Ghods, P.; Isgor, B.; Mcrae, A. Electrochemical investigation of chloride -induced depassivation of black steel rebar under simulated service condition. Corros. Sci. 2010, 52, 1649–1659. [CrossRef] Rocchini, G. Experimental verification of the validity of the linear polarization method. Corros. Sci. 1997, 39, 877–891. [CrossRef] Law, D.W.; Cairns, J.; Millard, S.G.; Bungey, J.H. Measurement of loss of steel from reinforcing bars in concrete using linear polarisation resistance measurements. NDT E Int. 2004, 5, 381–388. [CrossRef] Stern, M.; Geary, A.L. Electrochemical polarization: I. A theoretical analysis of the shape of polarization curves. J. Electrochem. Soc. 1957, 104, 56–63. [CrossRef] Li, L.; Sagüés, A.A. Chloride Corrosion Threshold of Reinforcing Steel in Alkaline Solutions Open-Circuit Immersion Tests. Corrosion 2001, 57, 19–28. [CrossRef] Gunay, H.B.; Isgor, O.B.; Ghods, P. Kinetics of Passivation and Chloride-Induced Depassivation of Iron in Simulated Concrete Pore Solutions Using Electrochemical Quartz Crystal Nanobalance. Corrosion 2015, 71, 615–627. [CrossRef] Angst, U.; Elsener, B.; Larsen, C.K.; Vennesland, O. Critical chloride content in reinforced concrete—A review. Cem. Concr. Res. 2009, 39, 1122–1138. [CrossRef]

Materials 2017, 10, 903

22.

23.

24. 25. 26.

27. 28. 29.

21 of 21

Wang, D.; Sun, W.; Shi, J. The Passive Film of A New Alloyed Corrosion-Resisitant Steel in Simulated Concerete Pore Solutions. In Proceedings of the 14th International Congress on the Chemistry of Cement, Beijing, China, October 2015. Kelly, R.G.; Kruger, J. Symposium on Critical Factors in Localized Corrosion. In Proceedings of the Symposium on Critical Factors in Localized Corrosion III: A Symposium in Honor of the 70th Birthday of Jerome Kruger, Boston, MA, USA, November 1999. Pistorius, P.C.; Burstein, G.T. Metastable Pitting Corrosion of Stainless Steel and the Transition to Stability. Philos. Trans. R. Soc. Math. Phys. Eng. Sci. 1992, 341, 531–559. [CrossRef] Feng, X.; Lu, X.; Zuo, Y.; Zhuang, N.; Chen, D. The effect of deformation on metastable pitting of 304 stainless steel in chloride contaminated concrete pore solution. Corros. Sci. 2016, 103, 223–229. [CrossRef] Figueira, R.B.; Sadovski, A.; Melo, A.P.; Pereira, E.V. Chloride threshold value to initiate reinforcement corrosion in simulated concrete pore solutions: The influence of surface finishing and pH. Constr. Build. Mater. 2017, 141, 183–200. [CrossRef] Stewart, J.; Williams, D.E. ChemInform Abstract: The Initiation of Pitting Corrosion on Austenitic Stainless Steel: On the Role and Importance of Sulfide Inclusions. Corros. Sci. 1992, 23, 457–463. [CrossRef] Liu, M.; Cheng, X.; Li, X.; Pan, Y.; Li, J. Effect of Cr on the passive film formation mechanism of steel rebar in saturated calcium hydroxide solution. Appl. Surf. Sci. 2016, 389, 1182–1191. [CrossRef] Ha, H.Y.; Park, C.J.; Kwon, H.S. Effects of non-metallic inclusions on the initiation of pitting corrosion in 11% Cr ferritic stainless steel examined by micro-droplet cell. Corros. Sci. 2007, 49, 1266–1275. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).