Surface conditions for microcosm development and proliferation of SRB on steel with cathodic corrosion protection

Construction and Building Materials 243 (2020) 118209

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Surface conditions for microcosm development and proliferation of SRB on steel with cathodic corrosion protection Samanbar Permeh a,⇑, Kingsley Lau a, Berrin Tansel a, Matthew Duncan b a b

Department of Civil and Environmental Engineering, Florida International University, Miami, FL, United States State Materials Office, Florida Department of Transportation, Gainesville, FL, United States

h i g h l i g h t s
Corrosion conditions were evaluated with externally applied cathodic polarization.
Field and laboratory experiments were conducted.
Proliferation of bacteria was not inhibited by cathodic polarization at 1000 mVCSE.
Sulfate reduction can be a significant contributor in corrosion for steel with CP.

a r t i c l e

i n f o

Article history: Received 22 August 2019 Received in revised form 20 December 2019 Accepted 16 January 2020

Keywords: Microbiologically influenced corrosion Cathodic protection Sulfate-reducing bacteria Marine fouling Potentiostatic polarization

a b s t r a c t Steel components used in civil infrastructure are susceptible to microbiologically influenced corrosion (MIC) in marine environments due to the interactions between the metal and its environment with varying water quality parameters (chemical and biological). Sulfate-reducing bacteria (SRB) has been associated with MIC by a cathodic depolarization mechanism due to the reaction of surface hydrogen. Recent findings at a Florida bridge showed severe localized corrosion associated with microbial activity and heavy marine fouling of the submerged steel piles. Cathodic protection can be applied to control the corrosion of steel; however, the mitigation may be impaired by the microbial activity and marine fouling. The objective of this work was to explore reduction reactions on cathodically polarized steel surfaces and evaluate the influence of cathodic polarization on sulfate reduction reaction. It was of interest to characterize the reduction reaction in crevices representative of different physical properties of the marine foulers where SRB can be supported. Field testing was conducted to differentiate the cathodic protection currents that develop in the presence of microorganisms under the layer of marine fouling in natural environments. Laboratory tests were conducted to elucidate the cathodic reactions by potentiostatic cathodic polarization of steel specimens in solutions inoculated with SRB. Field testing showed that proliferation of the bacteria was not inhibited in the presence of cathodic polarization at about 1000 mVCSE and corrosion continued in the localized regions under fouling encrustation. The larger cumulative charge relating to the apparent sulfate reduction by SRB corresponded to higher levels of cathodic polarization. Sulfate reduction by SRB can be a significant contributor to the electrochemical process for steel corrosion with cathodic polarization. For the porous crevice environments, large cathodic currents and high sulfate reduction coincided with the development of sustained SRB growth. Ó 2020 Elsevier Ltd. All rights reserved.

1. Introduction Microbiologically-influenced corrosion (MIC) of steel components used in civil infrastructure occurs in soil and water environments relating to the microbial activities and the interfaces that develop on the metal surface [1]. The interface typically contains

⇑ Corresponding author. E-mail address: [email protected] (S. Permeh). https://doi.org/10.1016/j.conbuildmat.2020.118209 0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.

a biofilm constituted of extracellular polymeric substances (EPS) produced by the bacteria [2]. Sulfate-reducing bacteria (SRB) have been associated with MIC [3,4]. Early explanations of the corrosion mechanism proposed that cathodic depolarization occurs by the metabolic activities of the microorganisms where atomic hydrogen on the iron surface is depleted by the hydrogenase enzyme produced by the bacteria and the cathodic reactions support the corrosion of iron [5]. The reactions associated with the corrosion process are represented by the following equations [6]:

2

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

4 Fe ! 4 Fe2þ þ 8 e ðanodic reactionÞ

ð1Þ

8 Hþ þ 8 e ! 8 H ðcathodic reactionÞ

ð2Þ

SO4 2 þ 8 H ! S2 þ 4H2 O ðcathodic depolarization by SRBÞ ð3Þ 8 H2 O ! 8 OH þ 8 Hþ ðdissociation of waterÞ

ð4Þ

2 Hþ þ S2 ! H2 S ðreversible reactionÞ

ð5Þ

Fe2þ þ S2 ! FeS ðanode corrosion productÞ

ð6Þ

3 Fe2þ þ 6 ðOHÞ ! 3 FeðOHÞ2 ðanode corrosion productÞ

ð7Þ

Later refinements to the corrosion mechanism were developed in part due to observations of MIC in the presence of hydrogenase negative strain of SRB and in consideration of the limitations on hydrogen availability to support the occurrence of accelerated corrosion [7–9]. Research efforts have shown that the metabolic reaction products (i.e., hydrogen sulfide and ferrous sulfide) that are formed by SRB can also act as depolarizing agents and contribute to the corrosion process. Some researchers [10–12] suggested a more complex mechanism involving both sulfide and phosphide ions. Starkey [13] pointed out that several processes concerning the effect of ferrous sulfide, sulfur, ferrous hydrate, phosphide, and other products are involved in anaerobic corrosion [13]. Nevertheless, hydrogenase is considered to have some role in removing the atomic hydrogen and providing a supply of hydrogen sulfide for the cathodic reaction. The corrosion in SRB-induced MIC can be accounted for by the cathodic corrosion reactions either by the removal of atomic hydrogen by reduction to hydrogen sulfide or oxygen reduction [14]. Descriptions of SRB-induced MIC of steel by Gu et. al., 2009 [15] and in later works provide a framework for bio-electrochemical reactions in regions with low carbon availability, where local anodes that develop on the steel surface are supported by sulfate-reduction reactions by SRB utilizing extracellular electrons from the steel transferred to the bacteria cell by an extracellular electron transfer, EET, mechanism [15– 17]. The corrosion process generally conforms to the cathodic depolarization theory that describes the metabolic activities of the microorganisms to support iron oxidation, where atomic hydrogen on the iron surface is depleted and the electrons from the iron oxidation is used in intracellular biocatalysis required for the cathodic reactions within the biofilm [18,19]. Despite the body of work available in the literature regarding electron transfer reactions associated with the microorganisms involved in MIC, there is very little information on the phenomena occurring on polarized surfaces (e.g., systems with cathodic protection to mitigate corrosion of steel). A review by Rosenbaum [20] highlighted the mechanisms for extracellular electron transfer involving ctype cytochromes as well as the hydrogenase enzyme and discussed the possibility for reduction reactions to be catalyzed on polarized electrodes during physiological activities. Severe localized corrosion of submerged steel bridge piles was observed in an environment with high sulfate concentrations that facilitated the metabolic activities of SRB which also coincided with heavy marine fouling [21,22]. Conditions under the fouling organisms could allow SRB growth where the metabolic activities mediate cathodic reactions to support the corrosion [22,23]. Application of cathodic protection (CP) was assessed to identify cathodic polarization behavior of steel subjected to marine fouling and MIC [22,24]. Testing showed that the proliferation of the bacteria was not inhibited in the presence of cathodic polarization at about 1000 mV vs. the copper/copper-sulfate reference electrode, CSE.

Although the overall corrosion rates were reduced, localized regions of corrosion continued under fouling encrustation. Olivares [25] reported systems that had lower corrosion rates and reduced mass loss with applied polarization levels as negative as about 925 mVCSE but where the SRB population continued to proliferate due to an electrostatic attraction between the bacteria and the electric charges created by cathodic protection [25]. There are different views on how these microorganisms affect the cathodic protection efficiency. Guezennec [26] described the relationship between cathodic polarization and development of biofilm on surfaces exposed to both synthetic and natural seawater and showed that cathodically-produced hydrogen can encourage the growth of hydrogenase-containing bacteria such as SRB and the biologicallyproduced iron sulfide can contribute to an increase in cathodic current demand [26]. Bacteria in the biofilm can act as a depolarizing agent and increase the required current for cathodic protection [27–29]. The objective of the work presented here was to explore the reduction reactions on cathodically-polarized steel surfaces and assess the influence of the cathodic polarization on the microbiologically-influenced sulfate-reduction reaction in crevice environments. The work was conducted as part of research to elucidate the causes and mitigation of severe localized corrosion of steel bridge piles associated with micro- and macro-fouling. It was of interest to characterize the reduction reaction in crevices representative of different physical properties of the marine foulers where SRB can be sheltered [22,30–33]. Tests were conducted both in field conditions and in the laboratory. Field testing was conducted to identify and differentiate the cathodic polarization behavior of submerged steel with cathodic protection, that develops in the presence of microorganisms under occluded spaces of naturally developed marine fouling. Laboratory tests, consisting of potentiostatic cathodic polarization of steel specimens in solutions inoculated with SRB, were conducted to elucidate and verify the cathodic reactions that occur under compact (hard) and porous (soft) crevices representative of the different marine fouling conditions observed in the field.

2. Materials and methods 2.1. Field site testing Steel coupons of 7.62 cm  12.70 cm  0.3175 cm (composition: 0.02 %C, 0.16 %Mn, 0.006 %S and 0.03 %Si, and balance Fe) were exposed at two natural brackish water river sites in Florida. Fourteen steel coupons were electrically coupled to a commercial bulk zinc anode (composition: 0.1–0.5 %Al, 0.02–0.07 %Cd, 0.005 %Fe, 0.006 %Pb, 0.005 %Cu, and balance Zn) for ~200 days at each test site. The duration of testing at test Site II was 25 days shorter than test Site I due to scheduling and weather logistic issues for the remote site testing. The time difference is not expected to create significant differentiation in the test results. Table 1 presents the field test conditions. An electrical switch was used for current measurements between the steel specimens and the zinc anode which allowed the measurement of the local CP current from the steelanode system to a single isolated test coupon. After decommissioning the testing at ~250 days of exposure, the surface area of the steel under marine fouling encrustation was swabbed and the population of SRB was analyzed using BART biological reaction tests by Droycon Bioconcepts, Inc. (Regina, Canada). The BART test was adopted for the field testing to provide a practical means to analyze and compare SRB levels on the specimen surfaces while the research team was at the remote test site. The surface film material within the ~6.45 cm2 sampling area under the fouling organisms was collected from the sterile swab by gently scraping and

3

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209 Table 1 Field site test conditions.

a

Test sites

Duration of exposure (days)

Duration of CP (days)

Sulfate concentration (mg/L)

SRB population (CFU/mL)

I II

279 245

191 169

2,800 1,900

27,000 (A)a 325 (M)a

Aggressivity: (A) Aggressive, (M) Moderately Aggressive. General guidelines for BART test for corrosion.

agitating the swab tip in the prescribed test volume of deionized water within the BART test vial. The vials were then stored in a sheltered environment and allowed to react. Aggressivity was defined based on the general guidelines for the BART test provided by the manufacturer. 2.2. Laboratory testing Steel bars with similar composition as the field-exposed test specimens were used for all laboratory testing. The experimental parameters for the laboratory tests are shown in Table 2. Cathodic potentiostatic polarization levels in the range associated with cathodic protection of steel (850 and 950 mV vs. the saturated-calomel reference electrode, SCE), were conducted for up to 7 days. Supplemental testing at 500 mVSCE was also conducted for comparative testing with anodic polarization. The net cathodic or net anodic currents were periodically measured (more frequently for the first 2 days and then daily until day 7). The cumulative charge, Q, at each measurement period, n, was calculated by Equation (8) for the duration of the tests, where I (A) is the measured current and t (s) is the measurement period.

Q ðtnþ1 Þ ¼

X n¼0

Inþ1  ðtnþ1  tn Þ

hard-shell barnacles and soft marine flora and fauna deposits as shown in Fig. 1. An SCE was used as a reference electrode. An activated-titanium mesh was used as the counter electrode. The test cells were filled with 300 mL deionized water and 20 mL modified-Postgate B medium solution [34]. The composition of the medium is presented in Table 3. The pH of all test solutions was periodically measured with a glass pH-electrode and was between 6.5 and 8.0 throughout the testing. A schematic of the test cell is shown in Fig. 2. All test cells were assembled with sterile components and the test specimens were rinsed in deionized water and sterilized with ethanol solution prior to testing. Chemical oxygen demand (COD), sulfide production, and sessile SRB population levels were monitored to determine the extent of microbiological activity. The COD values provide an indication of environments that can support microbial development and thus is a marker for microbial activity in the test solution. COD of each sample was measured by a colorimetric COD method on the first and final days of testing [35]. The sulfide concentrations relate to the level of sulfate reduction that is in part due to the metabolic activity of the SRB in the

ð8Þ

Plastic Cap Plastic Sheet with 0.16 cm Hole

Sponge

Shim (thickness 0.0076 cm) Mounted Steel sample

1.27cm

Supplemental potentiodynamic polarization tests were conducted for representative non-inoculated control specimens in Postgate B medium from the open-circuit potential (OCP) to 1.1 VSCE at a scan rate of 0.05 mV/s to verify the reduction reactions. The working electrode consisted of the transverse cross-section of a steel bar with a 1.27-cm diameter and a 10-mm grit surface roughness that either had an open- or crevice-surface condition. Crevice environments were considered as representations of the physical compact- and porous-crevice conditions characteristic of

Mounted Steel sample

Control

Mounted Steel sample

Porous (Soft) Crevice (Rep. of soft marine flora)

Compact (hard) Crevice (Rep. of barnacle shell)

Fig. 1. Schematic of working electrode in laboratory testing.

Table 2 Experimental test condition. Polarization levela

Cathodic

Inoculation

850 mVSCE

Inoculated

c

Non-inoculated

950 mVSCE

Inoculatedb

Non-inoculated

Anodic

500 mVSCE

Inoculatedb

Non-inoculated

a b c

Working electrode condition

No. of samples Naturally aerated

De-aeratedb

Control Compact crevice Porous crevice Control Compact crevice Porous crevice Control Compact crevice Porous crevice Control Compact crevice Porous crevice

2 2 2 2 2 2 2 2 2 2 2 2

2 2 2 2 2 2 2 2 2 2 2 2

Control Compact crevice Porous crevice Control Compact crevice Porous crevice

2 2 2 2 2 2

2 2 2 2 2 2

Open circuit potential (OCP) 650 to 750 mVSCE. On days 1 and 3, high-purity nitrogen gas was introduced into the test cells for 10 min. Inoculated with 10 mL SRB in Postgate medium.

4

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209 Table 3 Composition of the modified Postgate B medium. Constituents

Composition (%)

Potassium Phosphate (KH2PO4) Ammonium Chloride (NH4Cl) Sodium Sulfate (Na2SO4) Sodium Chloride (NaCl) Iron Sulfate (FeSO4.7H2O) Sodium Lactate Yeast extract

0.05 0.1 0.1 2.5 0.05 0.5 0.1

Sampling Air Port Working Outlet Electrode Nitrogen Gas Inlet

Site I

Site II

Reference Electrode (SCE) Counter Electrode (Titanium Mesh) Thin Layer of Mineral Oil

LugginCapillary Probe

Fig. 2. Schematic of laboratory corrosion cell.

bulk solution and on the metal surface. The changes in the concentrations of sulfide products and sulfate reactant during exposure can be used as a measure of SRB activity in MIC [36–38]. A hydrogen sulfide color disc test kit was used for monitoring sulfide concentration (in the form of hydrogen sulfide and metal sulfide) from extracted aliquots of the test solution. As a first approach to identify the role of SRB in corrosion development, the total sulfide production levels during the course of the testing were considered to be proportionate to the sulfate-reduction reactions on the metal surface. Thus, the changes in sulfide levels are considered here qualitatively. After ~7 days, the steel working electrodes were removed from the test solution, and coverings were removed from crevice specimens. Sessile test kits by Biotechnology Solutions (Houston, Texas, US) were used for detection of sulfate-reducing bacteria by serial dilutions in Modified Postgate B (MPB) following the NACE standard TM0194-2014 [39]. Sterile cotton swabs were used to gently scrape the surface of the sample (~1 cm2 area) and the soft deposits (formed by microbial activity) collected on the swab was placed into a sterile phosphate buffer solution (PBS). Serial dilutions of the 1-mL PBS ranged from 4 to 8 times. Images of the corrosion development and remnant physical effects of microbial activity were captured by a digital camera.

3. Results and discussion 3.1. Field testing Fig. 3 shows the representative marine growth observed on the steel coupons at the test sites. At Site I, hydroids and marine flora amassed with sporadic growth of barnacles at test depths 1.6– 2.6 m below marine growth (BMG). At Site II, barnacles were the

Fig. 3. Examples of marine fouling observed on coupons at Site I and Site II.

predominant macrofoulers at test depths down to 1.6 m BMG. The initial free-corrosion potential of the uncoupled steel coupons and zinc anodes as well as the mixed-potential after coupling was measured. The open-circuit potential (OCP) of the uncoupled bulk zinc anode was more negative than 1,000 mVCSE throughout the testing (as expected). The uncoupled steel specimens had OCP of about 700 mVCSE. The mixed-potential of the coupled system had on-potentials of about 1,000 mVCSE indicating that the anode remained in active condition and was able to provide cathodic polarization to the steel. The local electrical current between an isolated steel coupon and the remaining coupled steel-anode array was measured in part to differentiate the current for individual coupons by submersion depth and with varying levels of surface fouling. Detailed examination of the currents developed from the coupling of the zinc anode and the steel array can be found elsewhere [22,24]. Results for steel specimens at various submersion depths are shown in Fig. 4. CP current density on day ~200 was between 1 and 10 lA/cm2 exceeding the general protection levels suggested in the literature [40]; however, there was variability in the magnitude of the current between the specimens. At both test sites, lower values were unexpectedly observed for specimens at locations in closer proximity to the zinc anode. Higher currents would be expected for specimens placed closer to the zinc anode, and the lower currents were thought to be due to variations in the type of marine fouling attachment and coverage as well as the extent of calcareous deposits. Indeed, unlike Site II, cathodically-polarized specimens from Site I had positive identification of calcareous carbonates by X-ray diffractometry technique [24]. Fig. 5 compares the measured CP currents with the apparent corrosion rates estimated from the mass loss measurements. The lower CP currents measured at Site I generally corresponded to higher apparent corrosion rates. Coincident with the heavy marine fouling (compacted marine flora and varied sedentary fauna), the results indicated that there were portions of the steel surface that did not receive sufficient cathodic polarization thus allowing for differential corrosion cells to develop. Reciprocally, Site II had higher CP currents and lower overall apparent corrosion currents, even though the specimen surface was well covered primarily with bay barnacles. It was noted that the level of adhesion of the fouling layer (organisms) to the steel surface was relatively weaker at Site II than at Site I as much of the fouling bulk in the former was in the form of mounded and poorly-adhered fused layers of barnacle shells (Fig. 3). This type of marine fouling coverage allowed

5

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

Measured Current (mA)

10

1

0.1

Site I -191 Days Site II -245 days

Anode Location at Site I

Anode Location at Site II

0.01 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

Depth Below Marine Growth (m)

1

0.1 1

10 Apparent Corrosion Rate (MDD)

100

Fig. 5. Apparent corrosion rates corresponding to CP current measurements for the field exposed specimens.

relatively better interconnectivity for ion mobility and surface polarization of steel than those occurring at Site I. Both Sites I and II had high populations of SRB regardless of the effect of the cathodic polarization as presented in Table 4. The measured bacteria count indicated that aggressive conditions maintained high populations of SRB which could support MIC of steel. Any increase in solution pH within occluded regions did not have a strong effect to diminish the proclivity for the SRB to develop. This in part related to the non-uniform cathodic polarization under the fouling as well as dilution due to some level of ionic interaction with the bulk solution depending on the geometry and characteristics of the macrofouling. Field tests indicated differentiation in the magnitude and efficacy of the CP currents in relation to the surface coverage by calcareous deposits or shielding by tight deposits by marine fouling (marine flora and fauna) [22,24].

tions with apparent cathodic ennoblement. Polarization at 850 and 950 mVSCE can provide net cathodic currents. The total net cathodic reactions in the presence of the cathodic polarization include oxygen reduction (diffusion-controlled), hydrogen formation by activation polarization, and the sulfate reduction associated with SRB. In the non-inoculated solutions, the reduction reactions include oxygen reduction and hydrogen reduction. The limiting current density for the oxygen-reduction reaction should be lower due to concentration polarization in the de-aerated condition. Potentiodynamic polarization testing (Fig. 6) of non-inoculated control specimens confirmed (by Tafel characteristics) that concentration polarization of the oxygen-reduction reaction was dominant at 850 mVSCE and the hydrogen-reduction reaction become significant at 950 mVSCE for the tested aeration conditions. Fig. 7 presents the cumulative charge associated with the measured cathodic reactions for all test conditions for the specimens polarized to 850 and 950 mVSCE. As expected, the results showed an increase in the cumulative cathodic charge with time; and in general, specimens polarized to 950 mVSCE showed greater cumulative cathodic charge than specimens polarized to 850

-0.6 De-aerated conditon Naturally Aerated Condition -0.7

Ptential/VSCE

Site I Site II

Average corrosion rate

Measured Current (mA)

10

Maximum corrosion rate

Fig. 4. Current measurements for the field-exposed CP specimens.

-0.8

-0.9

-1

3.2. Laboratory testing

-1.1 1.E-06

3.2.1. Electrochemical cathodic and anodic charge The open-circuit potential for the specimens in the laboratory test solution was between 650 and 750 mVSCE, including condi-

1.E-05

1.E-04

Current Density/A cm-2 Fig 6. Potentiodynamic polarization curve for control laboratory specimens in Postgate B medium.

Table 4 Number of viable SRB (CFU/mL) for the field exposed samples. Site I

a

Site II

Control

CP

Control

CP

Zinc anode

6000 (A)a

1400 (M)

27,000 (A)

27,000 (A)

325 (M)

Aggressivity: (A) Aggressive, (M) Moderately Aggressive. General guidelines for BART test for corrosion.

6

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

Open Naturally Aerated Condition

Inoculated at -850mV

Non-Inoculated -850mV

Inoculated at -950mV

10

Non-Inoculated at -950mV

1 4

1. C=1•t1

8

2. C=1.14•t0.97

0.1 5

3. C=1.40•t0.90

6

7 3

4. C=1.78•t0.91 5. C=2.75•t0.85

0.01

De-aerated Condition

100

Inoculated at -850mV

Cumulative Cathodic Charge (C)

Cumulative Cathodic Charge (C)

100

1

6. C=2.99•t0.93

Non-Inoculated at -850mV

10

7 6 5 8 3

Inoculated at -950mV

Non-Inoculated at -950mV

1 1. C=0.62•t0.98

4

2. C=0.95•t0.86

0.1

3. C=1.47•t0.82

4. C=0.75•t0.89 2

0.01

5. C=2.21•t0.88 6. C=2.66•t0.96

1

7. C=3.38•t0.98

7. C=2.83•t0.97

8. C=0.75•t0.90

0.001 0.001

0.01

0.1

1

10

8. C=1.781•t0.91

0.001 0.001

0.01

0.1

1

10

Time (Day)

Time (Day)

Porous (Soft) Crevice

10

100

Inoculated at -850mV Non-Inoculated at -850mV Inoculated at -950mV Non-Inoculated at -950mV

5

6

8

1 1. C=1.93•t1.09 2. C=0.95•t0.86

0.1

3. 1

4

0.01

4. C=0.75•t0.89

3

5. C=2.21•t0.88

2

7

C=1.47•t0.82

6. C=2.66•t0.96

0.001 0.001

0.01

0.1

7.

C=3.38•t0.97

8.

C=1.78•t0.91

1

10

Cumulative Cathodic Charge (C)

Cumulative Cathodic Charge (C)

100

Naturally Aerated Condition

10

De-aerated Condition Inoculated at -850mV Non-Inoculated at -850mV Inoculated at -950mV Non-Inoculated at -950mV

5

1 1. C=1.56•t1.14 2. C=0.40•t0.82

0.1

3. C=1.45•t0.71

6

7

4. C=0.55•t0.69 8

0.01

5. C=3.75•t1.08

3 2

4

6. C=2.26•t0.94

1

7. C=1.59•t0.68

8. C=2.38•t0.75

0.001 0.001

0.01

0.1

1

10

Time (Day)

Time (Day)

Compact (Hard) Crevice

10

100

Inoculated at -850mV Non-Inoculated at -850mV Inoculated at -950mV Non-Inoculated at -950mV 5 8 6

1 1. C=0.46•t1.01 2. C=0.37•t0.90

0.1

3. C=0.45•t0.83

0.01

0.001 0.001

5.

C=0.80•t0.98

6. C=0.49•t0.86

7 3 4

4.

C=0.61•t0.94

1

10

1

1. C=0.33•t1.07

0.1

2. C=0.24•t0.87 3. C=0.21•t0.92 4. C=0.39•t0.94

0.01

5

7

10

Time (Day)

5. C=0.62•t0.91

6 4

8. C=0.63•t0.90

0.1

De-aerated Condition Inoculated at -850mV Non-Inoculated at -850mV Inoculated at -950mV Non-Inoculated at -950mV

1

7. C=0.89•t0.89

2

0.01

Cumulative Cathodic Charge (C)

Cumulative Cathodic Charge (C)

100

Naturally Aerated Condition

0.001 0.001

2 3

6. C=0.64•t0.92 1

0.01

7. C=0.80•t0.85

0.1

1

10

Time (Day)

Fig. 7. Cumulative cathodic charge measurements for the laboratory specimen at 850 mVSCE and 950 mVSCE polarization levels. The lines show the fitted power function.

mVSCE due to the higher reaction rates for the hydrogen-reduction reaction in the former. The trend of the potentiostatic data with time was evaluated by regression analysis using a power model as follows:

C ¼ a  tb

ð9Þ

where a is the scaling factor, and b is the power factor. The regression analyses showed that some decay in cathodic rates was apparent but was not significant in any of the test cases. Table 5 presents the values of the estimated model parameters (a and b) for all test

cases. The power coefficient estimated by the regression was typically about 1 in both aeration conditions at 850 and 950 mVSCE. At 850 mVSCE, the power factor was typically greater in the naturally-aerated conditions than the de-aerated conditions as expected due to the larger oxygen concentration in the bulk solution. At 950 mVSCE, the scaling factor was similar in both aeration conditions as hydrogen reduction would be dominant. The transient rate limitations in oxygen and hydrogen reduction, as well as possible limitations in available surface atomic hydrogen, were not significant with time during the testing [41,42].

7

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209 Table 5 Estimated values of parameter for the power function C = a∙tb fitted to the cumulative cathodic charge data (a: scaling factor, b: power factor). Test condition

At 850mVSCE

Control

De-aerated

Inoculated Non-inoculated

Porous crevice

Inoculated Non-inoculated

Compact crevice

Inoculated Non-inoculated

At 950mVSCE

Control

Inoculated Non-inoculated

Porous crevice

Inoculated Non-inoculated

Compact crevice

Inoculated Non-inoculated

a

b

a

b

0.62 0.95 0.75 1.47 0.45 0.56 0.40 0.55 0.33 0.24 0.21 0.39

0.98 0.86 0.89 0.82 0.72 1.14 0.82 0.69 1.07 0.87 0.92 0.94

1.00 1.14 1.40 1.78 0.64 1.93 0.92 0.45 0.46 0.37 0.61 –

1.00 0.96 0.90 0.91 0.81 1.09 0.70 0.82 1.01 0.90 0.93 –

2.21 2.66 1.78 3.37 2.26 3.75 1.59 2.38 0.62 0.64 0.80 –

0.88 0.96 0.91 0.97 0.94 1.08 0.68 0.75 0.91 0.92 0.85 –

2.75 2.99 0.75 2.83 2.75 3.08 1.07 1.14 0.80 0.49 0.89 0.63

0.85 0.93 0.90 0.97 0.93 1.04 0.65 0.72 0.98 1.01 0.89 0.90

-850 -950 -850 -950 DE NA non Inoculated

10 1 0.1

-850 -950 -850 -950 DE NA Inoculated

-850 -950 -850 -950 DE NA non Inoculated

10 1 0.1

-850 -950 -850 -950 DE NA Inoculated

-850 -950 -850 -950 DE NA non Inoculated

Cum. Cathodic Charge (C)

-850 -950 -850 -950 DE NA Inoculated

Cum. Cathodic Charge (C)

0.1

100

100

Cum. Cathodic Charge (C)

Open

1

Porous (Soft) Crevice

100

10

Day 7

Compact (Hard) Crevice

Cum. Cathodic Charge (C) Cum. Cathodic Charge (C)

100

Cum. Cathodic Charge (C)

Compact (Hard) Crevice

Porous (Soft) Crevice

Open

Day 1 100

Naturally aerated

100

10

1 0.1

-850 -950 -850 -950 DE NA Inoculated

-850 -950 -850 -950 DE NA non Inoculated

-850 -950 -850 -950 DE NA Inoculated

-850 -950 -850 -950 DE NA non Inoculated

-850 -950 -850 -950 DE NA Inoculated

-850 -950 -850 -950 DE NA non Inoculated

10 1 0.1

10 1 0.1

Fig. 8. Cumulative cathodic charge at days 1 and 7 for different geometric conditons. DE: de-aerated, and NA: naturally aerated.

Larger cumulative cathodic charge developed for the more electronegative test condition. Indeed, the scaling coefficient was higher for the 950 mVSCE polarization level than for the 850 mVSCE polarization level. The estimated scaling coefficient was generally smaller for the cases with crevice environments and had a lower cumulative cathodic charge in comparison to the specimens with the open no-crevice condition. This is consistent with the hypothesis that there is a smaller metal surface area available

to support the reduction reactions, especially for the compactcrevice conditions [43–45]. The compact-crevice environments can also reduce the interaction of the inoculated SRB (within occluded crevice regions) with the bulk solution; therefore, reducing the level of cathodic reactions associated with SRB. To better discriminate the trends, the cathodic cumulative charge on day 1 (allowing time for SRB growth) and at the end of the testing at day 7 for both polarization levels and for each test

8

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

cathodic reactions (e.g., reaction (2) shown in equation (2)) when those reactions are suppressed. The ubiquitous nature of SRB suggests that the microorganisms can continue to develop; however, limitations on available surface hydrogen required for sulfate reduction as described by the classical depolarization theory may regulate the SRB growth. Work by Moreno [46] and Kloeke [47] on the roles of cytochromes and hydrogenase in the transfer of electrons from charged surfaces elucidate the mechanisms of bioelectrochemical reductions occurring in engineering systems. The overall measured behavior would be accounted by a combination of the factors discussed.

condition were compared (Fig. 8). For the open-surface condition, the magnitude of the cumulative cathodic charge was not dissimilar for the polarized specimens in the inoculated and noninoculated solutions. The large steel surface area can accommodate the oxygen- and hydrogen-reduction reactions and any contribution of sulfate-reduction reactions by SRB was not readily discernable. The role of different crevice geometries on cathodic polarization behavior studied by other researchers [22,24] showed that large SRB populations can develop in the crevices regardless of cathodic polarization levels as electronegative as 1,000 mVCSE. These conditions allowed sustained growth of SRB within the crevices suggested that the cathodic reaction related to sulfate reduction can be an important mechanism. In this vein, the test results indicated that relatively high cathodic reaction rates developed in the porous-crevice environments with the presence of SRB. The porous-crevice environments provided shelter and localized low oxygen regions with a sufficient supply of nutrients to support SRB growth. On the other hand, the magnitude of cathodic reduction reactions (especially the oxygen and hydrogen surface reduction reactions) was evidently lower due to the decrease and nonuniform polarization of the steel surface under the compactcrevice spaces. Comparison of the cumulative cathodic charge for the inoculated and non-inoculated test solutions for the compact-crevice specimens did not strongly relate the contributions of charge-transfer reactions by the SRB. The specimens with the applied 500 mVSCE polarization consistently showed net anodic currents as shown in Fig. 9 and therefore does not directly relate to the cathodic polarization behavior. The anodic polarization behavior of the steel was not expected to be significantly different in the presence of SRB inoculum as the test solution environment itself allowed for corrosion initiation. The oxygen- and more so hydrogen-reduction reaction rates at this level of anodic polarization would be minimal as these electrochemical half-cell reactions would follow Tafel behavior. However, there were changes in the anodic currents during the testing period. This can in part be accounted for by the physical effects of rust accumulation or biofilm formation on the surface of the specimen. However, conceptually, any increases in reduction reaction rates that occur due to the growth of SRB would correspondingly have an inverse effect on the measured net anodic reaction rate. To the latter point, as shown in Figs. 9 and 10, there was a general indication of lower net anodic rates for test specimens in the inoculated solutions (i.e., with SRB activity) for the de-aerated open and porous-crevice conditions. Furthermore, there are complex mechanisms that sustain the SRB metabolic activities involving charge transfer associated with

Naturally Aerated Condition

1000

Inoculated Inoculated with Compact Crevice

Cumulative Anodic Charge (C)

Cumulative Anodic Charge (C)

1000

3.2.2. Microbiological activity Chemical and microbiological analysis was conducted for the solutions of the cathodic and anodic polarization test specimens to identify the levels of SRB activity. Although COD levels do not directly give an indication of the SRB populations. COD levels are considered as a metric of environmental conditions to support SRB activity [36]. As shown in Fig. 11, COD measurements of the test solutions at the onset of the test showed high COD levels indicating environments that can support SRB growth. The COD levels typically dropped overall by the end of the testing; however, the COD levels were generally higher in the inoculated solution than the control non-inoculated solutions throughout the test indicating that the environments in the former allowed more suitable conditions for the SRB growth as observed by increase in organic content that can develop with SRB proliferation. Correspondingly, the low final COD levels in the non-inoculated solutions are indicative of low SRB activity. Fig. 12 presents the apparent sulfide production rates estimated for the different test conditions including polarization levels, aeration levels, and crevice geometries. The test results showed that sulfide production occurred at different levels depending on the surface conditions throughout the duration of the test regardless of the level of cathodic polarization. The different forms of hydrogen sulfide detected by the test kit derived from sulfide, S2-, produced by the sulfate-reduction reaction as part of SRB metabolic activities were considered in part associated with charge-transfer reactions. The sulfide levels measured at discrete times during the exposure were used to calculate the apparent rate of sulfide production within the fixed solution volume. The apparent rate of sulfide production was assumed to be constant during the time intervals between sulfide measurements and on the first approach was assumed to be primarily related to SRB presence. It was apparent that overall, the level of sulfide production was greater with

Inoculated with Porous Crevice Non-Inoculated Non-Inoculated with Compact Crevice

10

Non-Inoculated with Porous Crevice

0.1

0.001 0.001

0.01

0.1

Time (Day)

1

10

De-aerated Condition Inoculated Inoculated with Compact Crevice Inoculated with Porous Crevice Non-Inoculated Non-Inoculated with Compact Crevice

10

Non-Inoculated with Porous Crevice

0.1

0.001 0.001

0.01

0.1

Time (Day)

Fig. 9. Cumulative anodic charge measurement for laboratory specimens at 500 mVSCE polarization level.

1

10

9

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

10 1 -500 -500 DE NA non Inoculated

100 10 1 0.1

-500 -500 DE NA Inoculated

-500 -500 DE NA non Inoculated

100 10 1 0.1

-500 -500 DE NA Inoculated

-500 -500 DE NA non Inoculated

Porous (Soft) Crevice

-500 -500 DE NA Inoculated

Compact (Hard) Crevice

0.1

Cum. Anodic Charge (C)

Open

100

100 10 1 0.1

-500 -500 DE NA Inoculated

Cum. Anodic Charge (C)

1000

Day 7 1000

1000

Cum. Anodic Charge (C)

Cum. Anodic Charge (C) Cum. Anodic Charge (C)

1000

Cum. Anodic Charge (C)

Compact (Hard) Crevice

Porous (Soft) Crevice

Open

Day 1 1000

1000

-500 -500 DE NA non Inoculated

100 10 1 0.1

-500 -500 DE NA Inoculated

-500 -500 DE NA non Inoculated

-500 -500 DE NA Inoculated

-500 -500 DE NA non Inoculated

100 10 1 0.1

Fig. 10. Cumulative anodic charge at days 1 and 7 for different geometric conditons. DE: de-aerated, and NA: naturally aerated.

1000

COD (mg/L)

800 600 400 200 0 -500 mV SCE

1

OCP

-850 mV SCE -950 mV SCE -500 mV SCE

Inoculated with SRB

OCP

1

-850 mV SCE -950 mV SCE

Non-Inoculated

Fig. 11. Range of chemical oxygen demand measured in the laboratory test specimens at 500 mVSCE, 850 mVSCE , 950 mVSCE and OCP condition. (Black lines: measurement on Day 1, Red lines: measurement on Day 7), 1: data from [22]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the larger cathodic polarization and smaller in the presence of the anodic polarization. Sulfide production relating to SRB activity would be expected to be larger in the de-aerated condition. This appeared to be so for the open-surface specimens at 850 mVSCE. However, there was an indication of greater sulfide production for similar specimens in the naturally-aerated solution at the 950 and 500 mVSCE polarization levels. This indicated that SRB was able to proliferate there. Yet, in congruity with the high total bacteria population (sampled from the surface of the specimens) in the crevices (Table 6), the apparent sulfide production rate for the specimens with crevices

was higher and prolonged relative to the open geometry. Indeed, sulfide production levels were high for the porous-crevice condition regardless of bulk solution aeration levels, consistent with the larger measured cathodic currents in the lab testing and relating to the benign environments there to support SRB as described earlier. Relatively high sulfide production in the compact-crevice environments was measured as well. In all cases, the level of sulfide production decreased over time indicating a decrease in SRB activity during testing. However, it was also apparent that SRB continued to grow to some extent as sulfide production continued in many cases throughout the test exposure. The effects of oxygen

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209 At -850 mVSCE/De-Aerated

7

Apparent Sulfide Production Rate (mg/day)

Apparent Sulfide Production Rate (mg/day)

10

Open Compact (Hard) Crevice

6

Porous (Soft) Crevice 5 4 3 2 1 0 4 Time (day)

6

8

10

Compact (Hard) Crevice

6

Porous (Soft) Crevice 5 4 3 2 1 0 0

Apparent Sulfide Production Rate (mg/day)

Open Compact (Hard) Crevice

6

Porous (Soft) Crevice 5 4 3 2 1 0 2

4 Time (day)

6

8

Compact (Hard) Crevice Porous (Soft) Crevice

5 4 3 2 1 0 0

2

4 Time (day)

6

Time (day)

6

8

10

Open Compact (Hard) Crevice

6

Porous (Soft) Crevice 5 4 3 2 1 0 0

2

4 Time (day)

6

8

10

At -500 mVSCE/Naturally Aerated

Open

6

4

7

10

At -500 mVSCE/De-Aerated

7

2

At -950 mVSCE/Naturally Aerated

At -950 mVSCE/De-Aerated

7

0

Apparent Sulfide Production Rate (mg/day)

2

Open

Apparent Sulfide Production Rate (mg/day)

Apparent Sulfide Production Rate (mg/day)

0

At -850 mVSCE/Naturally Aerated

7

8

7

Open Compact (Hard) Crevice

6

Porous (Soft) Crevice 5 4 3 2 1 0

10

0

2

4 Time (day)

6

8

10

Fig. 12. Apparent sulfide production rate in laboratory inoculated test specimens at 500 mVSCE, 850 mVSCE, 950 mVSCE polarization levels.

Table 6 Terminal bacteria concentrations (population per mL) for laboratory test samples. Aeration

Polarization mVSCE

SRB inoculation Open

Naturally aerated

De-aerated

a

950 850 OCPa 500 950 850 OCPa 500

OCP: from 650 to 750 mVSCE (data reported from [22]).

102–104 102–104 108 103–108 102–106 101–103 108 102–104

No SRB inoculation

Crevice

Open

Compact

Porous

103–106 103–108 107 102–104 103–108 101–108 108 101–106

0–106 0–106 107 0 0–106 0–108 107 0–103

0 0 103 10 104 101 0 0

Crevice Compact

Porous

0–102 102–103 0 0–104 0–102 0–102 103 0–102

0 0 0 0 0 0 0 0

11

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

3.2.3. Cathodic reactions The trends relating to the measured cathodic current and the apparent sulfide production rate shown in Fig. 14 were consistent with the observed surface corrosion characteristics for the specimens in solutions with SRB. The anodic polarization data for test specimens polarized to 500 mVSCE was not included in the figure. For the open-surface geometry, local cells under sulfide precipitates and biofilm created irregular and local tarnishing even though overall large cathodic-reduction reactions including oxygen and hydrogen reduction can develop with the polarization provided by the CP. For the porous-crevice environments, the large cathodic currents and the corresponding high level of sulfate reduction coincided with the SRB development. Localized surface heterogeneities developed under the porous crevices caused sinuous irregular surface corrosion as shown in Fig. 13B. Similar corrosion conditions can occur under the compact-crevice, especially as

A

SRB

NO SRB

B

NO SRB

SRB

NO SRB

-950 mVSCE NO SRB

SRB

Naturally De-aerated Naturally De-aerated Naturally De-aerated Aerated Aerated Aerated

Open Porous (Soft) Crevice

SRB

-850 mVSCE

-500 mVSCE SRB

Compact (Hard) Crevice

-950 mVSCE NO SRB

NO SRB

Naturally De-aerated Naturally De-aerated Naturally De-aerated Aerated Aerated Aerated

Open Porous (Soft) Crevice

Compact (Hard) Crevice

-850 mVSCE

-500 mVSCE SRB

Fig. 13. Laboratory specimens after testing and after sample cleaning.

300 -850 mVSCE

Charge by Sulfide Production (C)

and iron levels and ionic strength were assumed to be not significant in the oxidation of sulfide [48]. Table 6 presents the terminal surface SRB populations for all test cases on day 7. The results show that for the cathodicallypolarized specimens in the inoculated solutions, the crevice environments can facilitate the SRB growth in a similar manner to that occurring in anaerobic environments by providing shelter within the occluded space. In this case, the crevice spaces could adequately protect SRB even with strong cathodic polarization. As discussed earlier for the field conditions, any changes in solution pH due to cathodic oxygen reduction did not appear to have a significant effect on the SRB. Indeed, steel under porous crevices with SRB showed high cathodic currents reflecting favorable electrical properties (low resistance) through the sponge that allowed relatively high cathodic currents but did not produce a change in the environment that could reduce SRB growth. On the specimens subjected to anodic polarization, the SRB populations were relatively depressed. Fig. 13 shows the visual surface appearance of the test specimens for the de-aerated and naturally aerated conditions immediately upon removal from the test solution and after cleaning. The visual appearance of the specimens was similar regardless of aeration. The test specimens (with all tested surface conditions) placed in inoculated solutions showed thick accumulation of a black metal-sulfide precipitate consistent with the chemical and microbiological analyses discussed previously. The surface of the steel specimens with the open non-crevice geometry and inoculated with SRB had irregular and localized surface tarnishing for both cathodic polarization levels (-850 mVSCE and 950 mVSCE) whereas the surface was observed to be smooth and clean in the non-inoculated solutions. Surface pitting that formed due to the anodic polarization on all specimens polarized to 500 mVSCE appeared to be larger in the inoculated case. These observations indicate that the surface layer (due to microbial growth) created adverse underfilm conditions even with cathodic polarization. The surface of the steel specimens with compactand porous-crevices showed indication of corrosion regardless of the level of cathodic polarization as well as in the anodic polarization case. Corrosion was also apparent in the non-inoculated test cases; although, the surface oxidation was different if not more severe with the presence of SRB. The steel specimens with compact-crevices showed concentric surface tarnish from the center defect opening. The surface oxidation was black in color in the presence of SRB and was red–orange in the non-inoculated case. The steel surface of the specimens with the porous-crevice showed mottled surface oxidation that was more severe with the presence of SRB. The development of surface oxidation indicated that there was non-uniform cathodic polarization under the crevice environments.

-950 mVSCE

250

Open Porous Crevice

200

Compact Crevice

Porous Crevice

150 Compact Crevice

100

Filled Marker For De-aerated Condition

Open

50

0 0

10

20 30 Cumulative Charge (C)

40

50

Fig. 14. Cumulative charge associated with sulfide production and net cathodic reaction rates (lines correspond to the fitted curve).

12

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209

non-uniform protection of the steel is exacerbated by the higher electrical resistances under the crevice. This is consistent with the observed concentric geometry of the surface corrosion in the test specimens. The localized crevice environment created favorable conditions for SRB proliferation which allowed non-uniform cathodic polarization and development of localized corrosion. The cumulative molar sulfide content was estimated from the test data and assumptions based on the relevant charge-transfer theory and methodologies. As a first approach, based on the stoichiometry of the sulfate-reduction reaction (SO2 + 8H ? 4H2 4 O + S2) and associated reaction with surface adsorbed hydrogen (H+ + e ? H) by the hydrogenase enzyme in SRB, a charge associated to the sulfate-reduction reaction was derived from the sulfide levels ascribed by Faradaic conversion. A comparison of cumulative charge associated with the sulfide production and the measured net cathodic reaction rates is shown in Fig. 12. On first inspection, it was evident that the rate of sulfide production corresponded to an apparent cathodic charge larger than the total measured net cathodic reactions. As such, it was apparent that the measured sulfide levels developed in part by the reactions that do not exhibit charge-transfer characteristics corresponding to the idealized simple stoichiometry for the classical depolarization theory. Nevertheless, the results can be gleaned to provide qualitative generalizations on the cathodic reactions that can occur. For the open-surface and porous-crevice geometries, a positive trend relating the net cathodic charge to charge related to sulfate reduction was generally observed. The larger cumulative charge relating to the apparent sulfate reduction corresponded to the greater levels of cathodic polarization. This observation indicates that sulfate-reduction reactions due to SRB can be a significant part of the electrochemical process for steel with cathodic polarization (and reduce the effect of the CP). However, in the presence of SRB, the high cathodic rates for the steel in the cathodically-polarized condition would not necessarily mean enhanced steel corrosion if the electron donor is ascribed to the CP source. The large cathodic currents typically indicate cathodic polarization of the steel and corresponding reduced anodic corrosion currents. However, heterogeneities on the steel surface can occur due to the biofilm, microbial metabolites, and marine fouling. Local steel anodic sites, as exemplified by the irregular surface corrosion was observed in the laboratory specimens as well as field specimens (detailed in [22]). The results (especially for the crevice conditions) show that the surface heterogeneities including occluded spaces can have non-uniform polarization and linear resistances along the length of the specimen from the polarization source that can reduce the overall rate of the cathodic reaction. The presence of biofilm can also contribute to this effect. Non-uniform cathodic polarization for CP systems could allow localized corrosion to occur. This behavior was well manifested for the cases with the compact crevice and less so for the porous crevice reflecting the better ionic connectivity through the pores in the latter. Nevertheless, some portions of the steel in contact with the sponge (in the porous crevice) can exhibit similar non-uniform cathodic polarization.

4. Conclusions Field tests showed differentiation in both the magnitude and efficacy of CP current associated with reduced availability of surfaces due to marine fouling. The measured SRB bacteria counts in the field specimens indicated that aggressive conditions with high populations of SRB can be maintained even with the CP. Laboratory testing indicated non-uniform cathodic polarization of steel developed in the specimens with crevice geometries. Cathodic reactions related to SRB activity (sulfate reduction) was significant in the

presence of cathodic polarization. Consistent with the field results, SRB was sustained in the presence of the applied cathodic polarization during the lab testing period. Surface corrosion developed under crevice environments with externally applied cathodic polarization and was enhanced with the presence of SRB. CRediT authorship contribution statement Samanbar Permeh: Conceptualization, Investigation, Writing original draft. Kingsley Lau: Writing - review & editing, Supervision. Berrin Tansel: Writing - review & editing. Matthew Duncan: Project administration, Writing - review & editing. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgment This investigation was supported by the Florida Department of Transportation (FDOT). The opinions, findings, and conclusions expressed here are those of the authors and not necessarily those of the FDOT or the US Department of Transportation. Support from the FDOT State Materials Office is acknowledged here. The assistance by Dennis Baldi and the contributions by Mayren Echeverría Boan and Bin Li are acknowledged. References [1] B. Little, J.S. Lee, Microbially Influenced Corrosion, Wiley Series in Corrosion, Wiley, Hoboken, New Jersey, 2007. [2] S.C. Dexter, J.P. LaFontaine, Effect of natural marine biofilms on galvanic corrosion, Corrosion 54 (1998) 851–861. [3] R. Ray, J.S. Lee, B. Little, Factors contributing to corrosion of steel pilings in Duluth-Superior Harbor, Corrosion 65 (2009) 707–717. [4] A. Hu, Investigation of sulfate-reducing bacteria growth behavior for the mitigation of microbiologically influenced corrosion (MIC). Diss. Ohio University, 2004. [5] C.V.W. Kuhr, L.S. Van der Vlugt, The graphitization of cast iron as an electrobiochemical process in anaerobic soils, Water (den Haad) 18 (1934) 147–165. [6] S.W. Borenstein, Microbiologically Influenced Corrosion Handbook, Woodhead Publishing Ltd, Abington Hall, England, 1994. [7] R. Cord-Ruwisch, F. Widdel, Corroding iron as a hydrogen source for sulphate reduction in growing cultures of sulphate-reducing bacteria, Appl. Microbiol. Biotechnol. 25 (1986) 169–174. [8] I.P. Pankhania, A.N. Moosavi, W.A. Hamilton, Utilization of cathodic hydrogen by Desulfovibrio vulgaris (Hildenborough), Microbiology 132 (1986) 3357– 3365. [9] J.A. Hardy, Utilisation of cathodic hydrogen by sulphate-reducing bacteria, Br. Corrosion J. 18 (1983) 190–193. [10] J.A. Costello, Cathodic depolarization by sulfate-reducing bacteria, South African J. Sci. 70 (1974) 202–204. [11] R.A. King, J.D.A. Miller, Corrosion by the sulphate-reducing bacteria, Nature 233 (1971) 491. [12] W.P. Iverson, Mechanism of anaerobic corrosion of steel by sulfate reducing bacteria, Mater. Performance 23 (1984) 28–30. [13] R.L. Starkey, Anaerobic Corrosion–Perspectives About Causes, Biologically Induced Corrosion. (1985) 3–7. [14] B. Little, P. Wagner, F. Mansfeld, An overview of microbiologically influenced corrosion, Electrochim. Acta 37 (1992) 2185–2194. [15] T. Gu, K. Zhao, S. Nesic, A new mechanistic model for MIC based on a biocatalytic cathodic sulfate reduction theory. In: CORROSION 2009. NACE International. (2009), Paper No. 09390. [16] T. Gu, R. Jia, T. Unsal, D. Xu, Toward a better understanding of microbiologically influenced corrosion caused by sulfate reducing bacteria, J. Mater. Sci. Technol. 35 (2019) 631–636. [17] D. Xu, T. Gu, Carbon source starvation triggered more aggressive corrosion against carbon steel by the Desulfovibrio vulgaris biofilm, Int. Biodeterioration Biodegrad. 91 (2014) 74–81. [18] R. Jia, D. Wang, P. Jin, T. Unsal, D. Yang, J. Yang, D. Xu, T. Gu, Effects of ferrous ion concentration on microbiologically influenced corrosion of carbon steel by sulfate reducing bacterium Desulfovibrio vulgaris, Corros. Sci. 153 (2019) 127– 137.

S. Permeh et al. / Construction and Building Materials 243 (2020) 118209 [19] R. Jia, T. Unsal, D. Xu, Y. Lekbach, T. Gu, Microbiologically influenced corrosion and current mitigation strategies: a state of the art, Int. Biodeterioration Biodegrad. 137 (2019) 42–58. [20] M. Rosenbaum, F. Aulenta, M. Villano, L.T. Angenent, Cathodes as electron donors for microbial metabolism: which extracellular electron transfer mechanisms are involved?, Bioresour Technol. 102 (2011) 324–333. [21] S. Permeh, C. Reid, M. Echeverría, K. Lau, B. Tansel, M. Duncan, K. Lau Microbiological Influenced Corrosion (MIC) In Florida Marine Environment: A Case Study. In CORROSION 2017. NACE International. (2017), Paper No. 9536. [22] S. Permeh,M. Echeverría, B. Tansel, K. Lau, Susceptibility of Bridge Steel and Concrete Components to Microbiological Influenced Corrosion (MIC) and Microbiological Influenced Deterioration (MID) in Florida. Florida Department of Transportation, Final Report, Contract No. BDV29-977-26. (2019). [23] F.L. LaQue, Corrosion: Causes and Prevention, John Wiley & Sons, New York, 1975. [24] S. Permeh, K. Lau, M. Echeverria, B. Tansel, M. Duncan, Cathodic polarization behavior of steel with different marine fouling morphologies on submerged bridge elements with cathodic protection, Forthcoming, J. Mater. Civ. Eng. (2020), https://doi.org/10.1061/(ASCE)MT.1943-5533.0003257. [25] G.Z. Olivares, G.M. Mejia, G.G. Caloca, R.G. Esquivel, L.G. Lopez, C.M. UlloaOchoa, F.R. Dabur, Sulfate reducing bacteria influence on the cathodic protection of pipelines that transport hydrocarbons, In CORROSION 2003, NACE International. (2003), Paper No 03087. [26] J.G. Guezennec, Cathodic protection and microbially induced corrosion, Int. Biodeterioration Biodegrad. 34 (1994) 275–288. [27] M.de Romero, Matilde, O. De Rincon, L. Ocando. Cathodic protection efficiency in the presence of SRB: State of the art, In CORROSION 2009. NACE International, (2009), Paper No 09407. [28] G.H. Booth, A.K. Tiller, Polarization studies of mild steel in cultures of sulphatereducing bacteria, Trans. Faraday Soc. 56 (1960) 1689–1696. [29] H. Venzlaff, D. Enning, J. Srinivasan, K.J. Mayrhofer, A.W. Hassel, F. Widdel, M. Stratmann, Accelerated cathodic reaction in microbial corrosion of iron due to direct electron uptake by sulfate-reducing bacteria, Corros. Sci. 66 (2013) 88– 96. [30] S. Permeh, M. Echeverría, B. Tansel, K. Lau, M. Duncan M, Exploration of the Influence of Microbe Availability on MIC of Steel Marine Fouling Environments. In CORROSION 2019. NACE International. (2019), Paper No. 13461. [31] S. Permeh, B. Li, M. Echeverría, B. Tansel, K. Lau, M. Duncan, Microbially influenced steel corrosion with crevice conditions in natural water. In CORROSION 2018. NACE International. (2018), Paper No. 11529. [32] S. Permeh, K. Lau, M. Duncan, Characterization of biofilm formation and coating degradation by electrochemical impedance spectroscopy, Coatings 9 (2019) 518.

13

[33] S. Permeh, K. Lau, Degradation of coatings for steel in environments susceptible to corrosion associated with fouling, Struct. Infrastruct. Eng. (2019), https://doi.org/10.1080/15732479.2019.1694543. [34] J.R. Postage, The Sulphate Reducing Bacteria, 2nd ed., Cambridge university Press, Cambridge, England, 1984. [35] J.W. O’Dell, The determination of chemical oxygen demand by semiautomated colorimetry-method 410.4. Cincinnati, Ohio: Environmental Monitoring Systems Laboratory, Office of Research and Development, US Environmental Protection Agency (1993). [36] P.J.B. Scott, Expert consensus on MIC: failure analysis and control, Mater. Performance 43 (2004) 46–50. [37] J.S. Lee, R. Ray, B. Little, Influence of experimental conditions on the outcome of laboratory investigations using natural coastal seawaters, Corrosion. 66 (2010) 015001-015001. [38] B.J. Webster, R.C. Newman, Producing Rapid Sulfate-Reducing Bacteria (SRB)lnfluenced Corrosion in the Laboratory, Microbiologically influenced corrosion testing. ASTM International, 1994. [39] TM0194, NACE Standard, Field monitoring of bacterial growth in oil and gas systems, Houston, TX, NACE, (2014), ISBN: 1-57590-192-7. [40] P.M. Chess, J.P. Broomfield, Cathodic protection of steel in concrete. E & FN Spon (an imprint of routledge), London , 2003. [41] G.H. Booth, A.K. Tiller, Polarization studies of mild steel in cultures of sulphatereducing bacteria. Part 3. — Halophilic organisms, Trans. Faraday Soc. 58 (1962) 2510–2516. [42] S.C. Dexter, Biologically induced corrosion: proceedings of the International Conference on Biologically Induced Corrosion, June 10-12, Gaithersburg, Maryland, 1986. [43] R.R. Fessler, A.J. Markworth, R.N. Parkins, Cathodic protection levels under disbonded coatings, Corrosion 39 (1983) 20–25. [44] S. Nunez, A.D. Eastman, R.J. Babnick, Cathodic protection shielding definition, AGA Corrosion Supervisory Committee (1989). [45] D. Kuang, Y.F. Cheng, Study of cathodic protection shielding under coating disbondment on pipelines, Corros. Sci. 99 (2015) 249–257. [46] C. Moreno, R. Franco, I. Moura, J. Gall, J. Moura, Voltammetric studies of the catalytic electron-transfer process between the Desulfovibrio gigas hydrogenase and small proteins isolated from the same genus, Eur. J. Biochem. 217 (1993) 981–989. [47] F.V.O. Kloeke, R.D. Bryant, E.J. Laishley, Localization of cytochromes in the outer membrane of Desulfovibrio vulgaris (Hildenborough) and their role in anaerobic biocorrosion, Anaerobe 1 (1995) 351–358. [48] F.J. Millero, The thermodynamics and kinetics of the hydrogen sulfide system in natural waters, Marine Chem. 18 (1986) 121–147.