Electrochemical Assessment of Mitigation of Desulfovibrio ferrophilus IS5 Corrosion against N80 Carbon Steel and 26Cr3Mo Steel Using a Green Biocide Enhanced by a Nature-Mimicking Biofilm-Dispersing Peptide

MIC (microbiologically influenced corrosion) is problematic in many industries, especially in the oil and gas industry. In this work, N80 carbon steel for pipelines was tested with 26Cr3Mo chromium pipeline steel for comparison in SRB (sulfate-reducing bacterium) MIC mitigation using a THPS (tetrakis hydroxymethyl phosphonium sulfate)-based commercial biocide (Biotreat 5475 with 75–80% THPS by mass). Peptide A, a nature-mimicking synthetic cyclic peptide (cys-ser-val-pro-tyr-asp-tyr-asn-trp-tyr-ser-asn-trp-cys) with biofilm dispersal ability was used as a biocide enhancer. Metal coupons covered with 3-d old Desulfovibrio ferrophilus IS5 biofilms were immersed in different biocide solutions. After 1-h treatment, 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM (360 ppb) Peptide A, and 400 ppm Biotreat 5475 achieved 0.5-log, 1.7-log and 1.9-log reductions in sessile cell count on N80, and 0.7-log, 1.7-log, and 1.8-log on 26Cr3Mo, respectively. The addition of 200 nM Peptide A cut the THPS biocide dosage by nearly half. Biocide injection tests in electrochemical glass cells after 1 h exhibited 15%, 70%, and 72% corrosion inhibition efficiency (based on corrosion current density) on N80, and 27%, 79%, 75% on 26Cr3Mo, respectively. Linear polarization resistance and electrochemical impedance spectrometry results also indicated antimicrobial efficacies.


Introduction
N80 carbon steel is a common pipeline steel in oilfield applications. Like other carbon steels, N80 is susceptible to different types of corrosion such as CO 2 (carbon dioxide) corrosion and microbiologically influenced corrosion (MIC) [1][2][3][4]. Low chromium steels have much higher corrosion resistance to CO 2 and they are more economical than stainless steels [5][6][7]. The corrosion product films on chromium steels serve as a passive layer to hinder the outward diffusion of Fe 2+ , and the enrichment of Cr prevents corrosive anions from attacking the steel surface [8][9][10]. The addition of Cr in steels also improves MIC through the passive film a certain degree of antimicrobial effect of Cr ions [11]. 26Cr3Mo is a new grade chromium steel that belongs to 3Cr-L80 series pipeline steels [12].
MIC was first identified a century ago [13]. It is a major issue in the oil and gas industry due to its contribution to equipment failures including pipelines, and pressure vessels [14][15][16]. MIC is reported to be responsible for more than 20% of the total corrosion costs [17,18]. Sulfate-reducing bacteria (SRB) are involved in the majority of severe MIC cases. SRBs are ubiquitous in oilfield operations, because sulfate is widely present in many oilfield operations owing to the use of seawater injection in enhanced oil recovery [19][20][21][22]. To mitigate SRB MIC, biocide treatment is a commonly used technique in oilfields [23,24] together with pigging.
Chromium steels provide good passivation against CO 2 corrosion. However, their passive films are (semi-)conductive, allowing an SRB biofilm to harvest electrons across the passive film. Once there is a defective or damaged passive film spot, the spot will be preferentially used for Fe 2+ outward diffusion in a large cathode-small anode scenario, leading to classical pitting or pit amplification [49]. It was found that after a 7-d incubation with Desulfovibrio ferrophilus, 13%Cr had a weight loss of 4.4 mg/cm 2 (0.30 mm/a uniform corrosion rate) while its maximum pit depth reached 288 µm (15 mm/a pitting rate), compared to a much larger weight loss of 15.2 mg/cm 2 (1.0 mm/a uniform corrosion rate) accompanied by a much lower maximum pit depth of 7.3 µm (0.38 mm/a pitting rate) for N80 which had much inferior FeS passivation to the passivation provided by Cr oxides/hydroxides on 13%Cr [49]. This means that to deploy a Cr steel for CO 2 corrosion resistance applications, an MIC assessment and mitigation plan should be in place because there is potential for severe MIC attack.
Before a biocide is field tested, lab testing is desired to probe dosage and efficacy. There are two kinds of biocide tests. One is the biofilm prevention test, in which treatment chemicals are added to the culture medium upon inoculation [48]. The other is the biofilm kill (or biofilm eradication) test that uses biocides to treat pre-established biofilms [46,50,51]. In pipelines, the kill test is relevant, in which a biocide liquid "plug" moves downstream between two pigs driven by pressure. The residence time is rather short. In lab tests, such a short-term (e.g., 0.5 h or 1 h) will not generate a measurable weight loss or pit depth difference, unlike in biofilm prevention tests that last for days. Thus, real-time or near-real-time corrosion rate measurements are needed to calculate corrosion rate changes for the calculation of biocide corrosion inhibition efficiency that can be used to support biofilm sessile cell count reduction. Only electrochemical tests can provide near-real-time transient corrosion rate measurements. OCP (open circuit potential), LPR (linear polarization resistance), EIS (electrochemical impedance spectroscopy), and PDP (potentiodynamic polarization) scans are commonly used electrochemical techniques to assess corrosion [52][53][54].
This work investigated the mitigation efficacy of a THPS-based biocide, namely Biotreat 5475, enhanced by Peptide A against SRB MIC on N80 and 26Cr3Mo. The sessile cell reductions after a 1-h biocide treatment, and electrochemical responses to biocide injection were examined. Because sessile cells in a mature biofilm are more difficult to treat than preventing a biofilm from establishing on a metal surface, relatively high dosages (200 ppm and 400 ppm) of Biotreat 5475 were employed.

Electrochemical Test Results
OCP responses of N80 and 26Cr3Mo during the 1 h after the biocide injection treatment are shown in Figures 3 and 4. Without biocide injection, there were very few variations in OCP for N80 and 26Cr3Mo (controls). After biocide injection, OCP values of both N80 and 26Cr3Mo exhibited a decreasing trend over the 1-h period. This is obviously misleading. Theoretically, a lower OCP value indicates a greater corrosion tendency (working electrode losing electrons). However, in most SRB MIC cases, OCP trends are wrong because OCP only indicates corrosion tendency, not the actual corrosion kinetic process or corrosion outcome [47]. Kinetic electrochemical measurements which have been repeatedly proven reliable should be depended upon instead [25,48,52].

Electrochemical Test Results
OCP responses of N80 and 26Cr3Mo during the 1 h after the biocide injection treatment are shown in Figures 3 and 4. Without biocide injection, there were very few variations in OCP for N80 and 26Cr3Mo (controls). After biocide injection, OCP values of both N80 and 26Cr3Mo exhibited a decreasing trend over the 1-h period. This is obviously misleading. Theoretically, a lower OCP value indicates a greater corrosion tendency (working electrode losing electrons). However, in most SRB MIC cases, OCP trends are wrong because OCP only indicates corrosion tendency, not the actual corrosion kinetic process or corrosion outcome [47]. Kinetic electrochemical measurements which have been repeatedly proven reliable should be depended upon instead [25,48,52].     show the polarization resistance (R p ) variation during the 1-h biocide treatment. Because the inverse of R p from LPR can be used to represent corrosion rate, the decrease of 1/R p can estimate corrosion inhibition efficacy for a biocide treatment,    In the no-biocide injection control glass cell, there were only small variations in R p . After biocide injection, R p values of N80 and 26Cr3Mo gradually increased, suggesting corrosion rate reduction due to biocide treatment. After the 1-h treatment, 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM Peptide A, and 400 ppm Biotreat 5475 achieved 24%, 31%, and 43% 1/R p reductions, respectively for N80. For 26Cr3Mo, the reductions were 34%, 41%, and 40%, respectively. The cocktail of 200 ppm Biotreat 5475 + 200 nM Peptide A led to an extra 1/R p reduction for both N80 and 26Cr3Mo compared to 200 ppm Biotreat 5475 alone.
The equivalent electrical circuit model shown in Figure 7 was used for EIS spectra modeling. R s and R f are solution resistance and resistance of the film consisting of the biofilm and the corrosion products, respectively. R ct denotes charge transfer resistance. Q f and Q dl stand for the capacitance of the biofilm/corrosion product film and the doublelayer capacitance, respectively. Nyquist and Bode plots of N80 and 26Cr3Mo are displayed in  In the no-biocide injection control glass cell, there were only small variations in Rp. After biocide injection, Rp values of N80 and 26Cr3Mo gradually increased, suggesting corrosion rate reduction due to biocide treatment. After the 1-h treatment, 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM Peptide A, and 400 ppm Biotreat 5475 achieved 24%, 31%, and 43% 1/Rp reductions, respectively for N80. For 26Cr3Mo, the reductions were 34%, 41%, and 40%, respectively. The cocktail of 200 ppm Biotreat 5475 + 200 nM Peptide A led to an extra 1/Rp reduction for both N80 and 26Cr3Mo compared to 200 ppm Biotreat 5475 alone.
The equivalent electrical circuit model shown in Figure 7 was used for EIS spectra modeling. Rs and Rf are solution resistance and resistance of the film consisting of the biofilm and the corrosion products, respectively. Rct denotes charge transfer resistance. Qf and Qdl stand for the capacitance of the biofilm/corrosion product film and the doublelayer capacitance, respectively. Nyquist and Bode plots of N80 and 26Cr3Mo are displayed in Figures 8 and 9. The Nyquist plots in Figure 8A-D and Figure 9A-D indicate that biocide treatment increased the semi-circle diameters, indicating increased corrosion resistance. The EIS fitting results are listed in Tables 1 and 2.    Figure 8.   Table 2. Fitted electrochemical parameters of 26Cr3Mo from EIS data in Figure 9.  [56,57]. Because its inverse can represent corrosion rate, 1/(Rct + Rf) decrease is used to estimate corrosion inhibition efficacy for a biocide treatment,

Rs$$(Ω cm 2 )
In Figures 10 and 11, (Rct + Rf) was stable for N80 and 26Cr3Mo during the 1-h period without biocide injection. One h after the biocide injection, (Rct + Rf) of both N80 and 26Cr3Mo increased significantly, suggesting decreased corrosion rates due to biocide injection. Based on fitted EIS parameters in Tables 1 and 2, Figure 8.  [56,57]. Because its inverse can represent corrosion rate, 1/(R ct + R f ) decrease is used to estimate corrosion inhibition efficacy for a biocide treatment, In Figures 10 and 11, (R ct + R f ) was stable for N80 and 26Cr3Mo during the 1-h period without biocide injection. One h after the biocide injection, (R ct + R f ) of both N80 and 26Cr3Mo increased significantly, suggesting decreased corrosion rates due to biocide injection. Based on fitted EIS parameters in Tables 1 and 2, for N80, the 1-h biocide treatment  with 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM Peptide A, and 400 ppm Biotreat 5475 reduced 1/(R ct + R f ) by 41%, 58%, and 61%, respectively. For 26Cr3Mo, the reductions were 25%, 28%, and 60%, respectively. The EIS results indicated that the efficiency sequence was 200 ppm Biotreat 5475 < 200 ppm Biotreat 5475 + 200 nM Peptide A < 400 ppm Biotreat 5475, which is consistent with LPR results.   Tables 3 and 4. A higher corrosion current density (icorr) correlates to a higher uniform corrosion rate. Without biocide injection, the icorr values were 0.54 mA/cm 2 and 0.26 mA/cm 2 for N80 and 26Cr3Mo. A lower icorr value of 26Cr3Mo also suggests that   Tables 3 and 4. A higher corrosion current density (icorr) correlates to a higher uniform corrosion rate. Without biocide injection, the icorr values were 0.54 mA/cm 2 and 0.26 mA/cm 2 for N80 and 26Cr3Mo. A lower icorr value of 26Cr3Mo also suggests that Figure 11. 26Cr3Mo R f + R ct curves from EIS modeling during 1-h time after biocide injection. Figures 12 and 13 present Tafel curves of N80 and 26Cr3Mo after biocide treatment with Tafel scans starting 1.5 h after the biocide injection. The fitted Tafel parameters are summarized in Tables 3 and 4. A higher corrosion current density (i corr ) correlates to a higher uniform corrosion rate. Without biocide injection, the i corr values were 0.54 mA/cm 2 and 0.26 mA/cm 2 for N80 and 26Cr3Mo. A lower i corr value of 26Cr3Mo also suggests that 26Cr3Mo had a greater (uniform) corrosion resistance than N80.  Table 3. Tafel parameters of N80 fitted from potentiodynamic polarization curves in Figure 12.  Table 4. Tafel parameters of 26Cr3Mo fitted from potentiodynamic polarization curves in Figure 13.   Table 3. Tafel parameters of N80 fitted from potentiodynamic polarization curves in Figure 12.  Table 4. Tafel parameters of 26Cr3Mo fitted from potentiodynamic polarization curves in Figure 13. Corrosion inhibition efficiency (ηi) calculated from icorr is shown in Table 5. After the 1-h biocide treatment, ηi values for N80 were 15%, 70%, and 72% for 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM Peptide A, and 400 ppm Biotreat 5475, respectively. For 26Cr3Mo, the values were 27%, 79%, and 75%, respectively. The 200 ppm Biotreat 5475 didn't achieve a high corrosion inhibition, but 200 nM peptide turned out to enhance the biocide efficacy considerably, reaching a similar corrosion inhibition outcome as 400 ppm Biotreat 5475 did, which means cutting THPS dosage by half, consistent with the trend in sessile cell counts. Table 5 summarizes the biocide efficacy data from sessile cell reduction, LPR 1/Rp reduction, EIS 1/(Rct + Rf) reduction, and PDP icorr reduction. The data in this work indicated that all three electrochemical methods correctly supported sessile cell reduction outcomes. Table 5 data indicate that biocide treatment efficacies were similar for the two metals. Although different electrochemical methods provided different corrosion inhibition efficacy values, their trends were consistent. Thus, LPR, EIS, and PDP were all able to support the sessile cell count trend.

Ecorr$$(V) vs. SCE
The findings in this work were comparable to other Peptide A studies in that Peptide A turned out to be an effective biocide enhancer that was able to cut the biocide dosage by nearly half [47,48,52]. Peptide A was found attractive as it functioned at a very low concentration (sub-ppm level). The biofilm dispersal effect appeared crucial in preventing biofilm formation and MIC mitigation. The mechanism of the biofilm dispersal effect of Peptide A was speculated in a previous study. It will be worthwhile to investigate it in depth in the future. Corrosion inhibition efficiency (η i ) calculated from i corr is shown in Table 5. After the 1-h biocide treatment, η i values for N80 were 15%, 70%, and 72% for 200 ppm Biotreat 5475, 200 ppm Biotreat 5475 + 200 nM Peptide A, and 400 ppm Biotreat 5475, respectively. For 26Cr3Mo, the values were 27%, 79%, and 75%, respectively. The 200 ppm Biotreat 5475 didn't achieve a high corrosion inhibition, but 200 nM peptide turned out to enhance the biocide efficacy considerably, reaching a similar corrosion inhibition outcome as 400 ppm Biotreat 5475 did, which means cutting THPS dosage by half, consistent with the trend in sessile cell counts.  Table 5 summarizes the biocide efficacy data from sessile cell reduction, LPR 1/R p reduction, EIS 1/(R ct + R f ) reduction, and PDP i corr reduction. The data in this work indicated that all three electrochemical methods correctly supported sessile cell reduction outcomes. Table 5 data indicate that biocide treatment efficacies were similar for the two metals. Although different electrochemical methods provided different corrosion inhibition efficacy values, their trends were consistent. Thus, LPR, EIS, and PDP were all able to support the sessile cell count trend.
The findings in this work were comparable to other Peptide A studies in that Peptide A turned out to be an effective biocide enhancer that was able to cut the biocide dosage by nearly half [47,48,52]. Peptide A was found attractive as it functioned at a very low concentration (sub-ppm level). The biofilm dispersal effect appeared crucial in preventing biofilm formation and MIC mitigation. The mechanism of the biofilm dispersal effect of Peptide A was speculated in a previous study. It will be worthwhile to investigate it in depth in the future. Table 6 shows the elemental compositions of 26Cr3Mo and N80 steels. Biotreat 5475 was an industrial biocide from Clariant (Muttenz, Switzerland) that contained 75-80% of THPS by mass. Peptide A with 97.2% purity (based on peak area analysis in reverse-phase high-performance liquid chromatography) according to the supplier (Bachem Holding AG, Bubendorf, Switzerland) was used in this work. Tables 7 and 8 display the test matrices for this work. D. ferrophilus (strain IS5), a highly-corrosive pure strain SRB [49,58], was immersed in enriched artificial seawater (EASW) culture medium inoculated with D. ferrophilus at 28 • C for 3 days to yield mature biofilms on the metal coupon surfaces. The composition of EASW is listed in Table 9. The pre-cut 26Cr3Mo and N80 coupons were coated with an inert liquid Epoxy coating (3M Product 323) except for a 10 mm × 10 mm exposed top work surface. The coupons were left at room temperature for drying overnight. The top surface was polished to 600 grit and the coupon was sterilized with anhydrous isopropanol before testing. The initial pH of EASW was adjusted to 7.0 using 5% (w/w) NaOH. The medium was sterilized in an autoclave at 121 • C for 20 min. Then, it was sparged with filter-sterilized N 2 for 1 h to get rid of dissolved oxygen. Finally, an oxygen scavenger L-cysteine was added into the medium to reach a final concentration of 100 ppm. The biocide stock solution was filter-sterilized using a 0.22 µm Stericup single-use sterile filter (Millipore, Bedford, MA, USA). The inoculation (with a 1:100 volume ratio for inoculum vs. culture medium) was carried out in an anaerobic chamber filled with N 2 .

Sessile Cell Counts
For counting sessile cells, coupons of each metal type were put into separate 125 mL anaerobic vials with 50 mL EASW. After the 3-d pre-growth of D. ferrophilus, all coupons were taken out in the anaerobic chamber and rinsed with pH 7.4 phosphate-buffered saline (PBS) to remove planktonic cells. The coupons were then transferred into Petri dishes containing the pH 7.4 PBS solutions with and without biocide chemicals (1 cm 2 coupon surface per 25 mL PBS) in the anaerobic chamber. After the 1-h biocide soaking treatment, sessile cells, which were motile, were counted using a hemocytometer under a 400× microscope [25].

Electrochemical Measurements
Electrochemical tests were performed in 450 mL glass cells, each filled with 250 mL EASW. For each metal type (N80 or 26Cr3Mo), 4 glass cells were used (1 for no-biocide control, 3 for 3 different biocide treatments). A three-electrode setup was adopted for measurements ( Figure 14A). The working electrode (WE) Epoxy cake contained two replicate coupons as replicates ( Figure 14B). The counter electrode (CE) was a thin platinum sheet, and a saturated calomel electrode (SCE) was used as the reference electrode (RE). Electrochemical measurements were carried out using a PCI4/750 potentiostat (Gamry Instruments, Inc., Warminster, PA, USA). OCP, LPR, EIS and PDP were measured in this work. LPR was measured at a scanning rate of 0.167 mV/s from −10 mV to 10 mV (vs. OCP) after OCP became stable. EIS was scanned from 10 5 Hz to 0.01 Hz with a 10 mV amplitude sinusoidal signal. PDP was measured at the end of incubation. Tafel curves were obtained from two half-scans on the same working electrode starting from 0 mV to −200 mV (vs. OCP) and 0 mV to +200 mV (vs. OCP) with a scanning rate of 0.167 mV/s.

Electrochemical Measurements
Electrochemical tests were performed in 450 mL glass cells, each filled with 250 mL EASW. For each metal type (N80 or 26Cr3Mo), 4 glass cells were used (1 for no-biocide control, 3 for 3 different biocide treatments). A three-electrode setup was adopted for measurements ( Figure 14A). The working electrode (WE) Epoxy cake contained two replicate coupons as replicates ( Figure 14B). The counter electrode (CE) was a thin platinum sheet, and a saturated calomel electrode (SCE) was used as the reference electrode (RE). Electrochemical measurements were carried out using a PCI4/750 potentiostat (Gamry Instruments, Inc., Warminster, PA, USA). OCP, LPR, EIS and PDP were measured in this work. LPR was measured at a scanning rate of 0.167 mV/s from −10 mV to 10 mV (vs. OCP) after OCP became stable. EIS was scanned from 10 5 Hz to 0.01 Hz with a 10 mV amplitude sinusoidal signal. PDP was measured at the end of incubation. Tafel curves were obtained from two half-scans on the same working electrode starting from 0 mV to −200 mV (vs. OCP) and 0 mV to +200 mV (vs. OCP) with a scanning rate of 0.167 mV/s. In the electrochemical tests, D. ferrophilus biofilms were pre-grown on the working electrode surfaces to reach maturity, which took 3 days of incubation. After that, a concentrated biocide stock solution was injected into the anaerobic glass cell. The glass cell was then gently shaken for 3 min to disperse the biocide. OCP and LPR were measured every 20 min during the 1 h following the biocide injection. EIS scan was conducted just before and after the 1-h biocide treatment to evaluate the biofilm's response to the biocide treatment. Tafel scans were performed after other electrochemical measurements were performed. To evaluate corrosion inhibition efficiency (ηi), reduction in icorr was calculated using the following equation [52]: where icorr and icorr,0 represent corrosion current densities with and without biocide treatment, respectively. icorr,0 was obtained in the biotic control glass cell. The biofilm kill test by soaking coupons in a biocide solution simulates a concentrated biocide plug between two pigs moving down a pipeline. Please note that corrosion weight loss change and corrosion pit depth change in this kind of 1-h (to simulate 1 h contact time or residence time) biofilm kill test are not measurable. Electrochemical tests In the electrochemical tests, D. ferrophilus biofilms were pre-grown on the working electrode surfaces to reach maturity, which took 3 days of incubation. After that, a concentrated biocide stock solution was injected into the anaerobic glass cell. The glass cell was then gently shaken for 3 min to disperse the biocide. OCP and LPR were measured every 20 min during the 1 h following the biocide injection. EIS scan was conducted just before and after the 1-h biocide treatment to evaluate the biofilm's response to the biocide treatment. Tafel scans were performed after other electrochemical measurements were performed. To evaluate corrosion inhibition efficiency (η i ), reduction in i corr was calculated using the following equation [52]: where i corr and i corr,0 represent corrosion current densities with and without biocide treatment, respectively. i corr,0 was obtained in the biotic control glass cell. The biofilm kill test by soaking coupons in a biocide solution simulates a concentrated biocide plug between two pigs moving down a pipeline. Please note that corrosion weight loss change and corrosion pit depth change in this kind of 1-h (to simulate 1 h contact time or residence time) biofilm kill test are not measurable. Electrochemical tests are the only methods that can provide near-real time corrosion rate changes for biocide efficacy assessment to support sessile cell count reduction data.

Conclusions
26Cr3Mo showed a higher uniform corrosion resistance compared to N80 under the same conditions based on R p and i corr values before biocide treatment. The SRB biofilms on both metals responded similarly to Biotreat 5475 treatment. The various electrochemical corrosion measurements indicated that 200 ppm Biotreat 5475 was able to mitigate MIC by D. ferrophilus, but it was not very effective. The addition of 200 nM Peptide A considerably enhanced the efficacy of 200 ppm Biotreat 5475, which achieved similar efficacy as 400 ppm