Desulfovibrio vulgaris ATCC 7757, a strain of SRB utilized in this study, was obtained from the General Microbiological Culture Collection Center in China. The ATCC 1249 medium (Supplementary Text S1) was used for bacterial cultivation (Jia et al., 2018a). The medium underwent a 40-min deoxygenation process using N₂ and was subsequently autoclaved at 121°C for 25 min. Additionally, L-cysteine (100 mg/L, Macklin, Shanghai, AR>98%) was introduced as an oxygen scavenger (Jia et al., 2018b). D-Phe (Macklin, Shanghai, AR>98%) and PASP (Macklin, Shanghai, AR > 95%) were dissolved in deionized water and used after filtration with 0.22 μm filters. The information on growth curve was provided in Supplementary Text S2.

The microbial corrosion was evaluated in four reactors. The dimensions of each reactor were 92 mm in height, 60 mm in inner diameter, and a volume of 200 mL (Supplementary Figure S1). Eight 20# carbon steel coupons (10 × 10 × 3 mm), a commonly used material in industrial pipelines, were placed in each reactor. The elemental composition of the carbon steel is presented in Supplementary Table S1. The coupons first underwent a sequential grinding process using silicon carbide paper with varying grit sizes (220, 400, 800, 2000) and then received ultraviolet light exposure for 30 min to ensure sterilization (Jia et al., 2018b). Subsequently, all coupons were firmly secured in rubber sleeves within the reactors. The SRB was cultured in reactors with a medium that included PASP, D-Phe, and PASP + D-Phe, denoted as PASP, D-Phe, and P + D group, respectively. The concentrations of PASP and D-Phe were 40 mg/L and 10 mg/L according to our previous study, respectively (Gong et al., 2023). A blank group was prepared without the addition of PASP and D-Phe. Additionally, an abiotic group was designed without SRB. All reactors were incubated at 30°C for 14 days.

After 14-day incubation, the coupons underwent a treatment to remove biofilm and corrosion products according to the ASTM G1-03 standard (Zhang B. et al., 2015). Subsequently, the coupons underwent three rounds of cleaning with anhydrous alcohol, followed by drying in a vacuum oven at 60°C. Then, the weight loss was calculated via measuring the difference in mass of the coupons before and after the incubation. In each reactor, four coupons were prepared for weight measurement, the average weight loss was calculated for each data point, and the average corrosion rate (V) and inhibition rate (η) were determined using the following formulas: V = 8760 × 10 × m − m 0 A ∙ T ∙ ρ (1) η = v 0 − v 1 v 0 × 100 % (2) where V is the corrosion rate (mm/a). m is the mass reduction of the coupons. m₀ is the average weight loss of the blank test for the acid etching of the test coupons (g). A is the coupon surface area (cm²). T is the incubation time (h). ρ is the density of the coupon (g/cm³). v ₀ is the corrosion rate in blank group and v ₁ is the corrosion rate in other groups.

Additionally, a synergy parameter (S) was determined in terms of corrosion activities according to Aramaki–Hackerman model: S = 1 − η A t a m a k i t h r e s h o l d 1 − η m e a s u r e d (3) where η t h r e s h o l d is the threshold inhibition efficiency. η m e a s u r e d is the evaluated mixture’s inhibition efficiency. S > 1 indicates synergism, and S < 1 indicates antagonism (Aramaki and Hackerman, 1969; Kozlica et al., 2021; Kokalj, 2023).

After the incubation, the coupons with biofilm were collected and rinsed with phosphate-buffered saline (PBS, pH = 7.4) three times to clear the residual medium. Then, the coupons were fixed with 2.5% glutaraldehyde in PBS solution for 2 h and washed with PBS solution three times, followed by dehydration. Ultimately, the coupons were dried in a CO₂ dryer, and subsequent to this, palladium coating was applied to enhance conductivity before conducting SEM (Quanta 250 FEG, FEI, United States of America) analysis. The coupons were subjected to pretreatment to remove biofilm prior to SEM analysis to observe the pitting corrosion morphology. Subsequently, corrosion pits of coupons were observed and analyzed with a confocal laser scanning microscope (CLSM, LSM 800, Carl Zeiss, Germany) and ZEISS ConfoMap software, respectively (Li et al., 2021). Each sample was scanned in 10 regions to find the maximum pit depth. In addition, the corrosion products were collected, and the composition of sulfur-related components was recorded with X-ray photoelectron spectroscopy (XPS, Thermo Fisher Nexsa, United States of America). The XPS spectra was analyzed with Avantage software and the XPS reference binding energies (B.E.) of sulfur species were listed in Supplementary Table S2.

CLSM was also applied to observe cells in biofilm. Briefly, the collected coupon containing biofilm was gently washed using PBS solution and stained using SYTO 9 (live cells) and PI (dead cells) stain. Stained coupons were incubated in dark for 20 min to ensure optimal staining, after which the coupons were carefully washed using a PBS solution again to remove excess stain and to enhance the clarity of observations. Finally, the coupons were observed with CLSM (LSM 900, Carl Zeiss, Germany), and the images were obtained with ZEISS ZEN 3.7 software. ImageJ software was employed to measure the total fluorescence intensity of each image, and the relative amounts of cells were calculated in comparison with in blank group.

The incubation, while maintaining stringent control over experimental conditions, was reproduced in three-electrode cells. The platinum sheet functioned as the counter electrode. Carbon steel, with a 1 cm² exposed surface, served as the working electrode, and Ag/AgCl (KCl-saturated) was used as the reference electrode. The open circuit potential (OCP) and Electrochemical Impedance Spectroscopy (EIS) were measured at 1, 3, 6, 10, and 14 days, and potentiodynamic polarization tests were conducted at 14 days using an electrochemical workstation (CHI 660e, CH Instruments Inc., United States of America). The EIS measurement was conducted at the OCP using a sinusoidal signal with an amplitude of 10 mV, covering a frequency range from 100,000 Hz–0.01 Hz. The EIS data were fitted with ZsimpWin software. Additionally, the Tafel curve was automatically recorded at a scanning rate of 0.167 mV/s, ranging from −0.25 V to +0.25 V vs. the OCP.

The difference among samples were assessed employing the SPSS software through one-way ANOVA, and the levels of significance were denoted by p-values less than 0.05.

Weight loss was first measured and corrosion rates were calculated to assess corrosion. As shown in Figure 1, the corrosion rate increased from 0.018 mm/a in abiotic group to 0.034 mm/a in blank group, confirming that SRB can cause serious metal corrosion. Less weight loss was observed in both PASP group and D-Phe group with corrosion rates at 0.017 mm/a and 0.024 mm/a, respectively. The corrosion inhibition by D-Phe might be attributed to that D-Phe decreased the initial attachment of SRB on coupons, consequently limiting the consumption of iron by SRB. Besides, the corrosion greatly reduced in P + D group, with the corrosion rate decreasing by 76.54% than that in blank group. In previous studies, a corrosion inhibition of 80% was attained with the application of 4000 mg/L PASP, which is 100 times than our used dosage (Cui et al., 2011). Besides, Zeino et al. achieved high corrosion inhibition at 97% through the joint application of 500 mg/L PASP and 10 mg/L zinc ions (Zeino et al., 2018). In this study, a remarkable corrosion inhibition, up to 76.54%, was achieved with low doses of PASP (40 mg/L) and D-Phe (10 mg/L), both of which are non-toxic and biodegradable. Furthermore, according to the Aramaki-Hackerman model, the synergy parameter S was 1.48, indicating the synergism between D-Phe and PASP on corrosion inhibition.

After incubation, the coupons were cleaned to remove attached cells and observed with SEM and CLSM. Serious corrosion was detected on coupons surface in blank group with a huge area and depth at 7.65 μm (Figure 2A, A′). In the D-Phe group, some small and shallow-depth pits were observed, and the maximum depth was 3.26 μm (Figure 2C, C′), suggesting that D-Phe could alleviate pitting corrosion. Moreover, few deep pits were detected in PASP group and P + D group (Figures 2B,D) with maximum depths at 1.44 and 0.95 μm, respectively (Figures 2B’, D’). Typically, ferrite on the iron coupons is more susceptible to corrosion than cementite (Li et al., 2021; Li et al., 2023). It can be explained that cementite contains covalent bonds, which reduce electron activity and make it difficult for bacteria to utilize electrons (Hao et al., 2016; Liu et al., 2021). In this study, the joint application of PASP and D-Phe led to a decrease in the depth of pitting corrosion, consequently preventing the exposure of ferrite, ultimately resulting in fewer electrons lost from the coupons and utilized by SRB. These results are consistent with corrosion rates and confirm the synergetic role of PASP and D-Phe in corrosion inhibition.

Hydrogen sulfide and other sulfides, the corrosive products formed during the reduction of sulfate by SRB, play an important role in metal corrosion by Desulfovibrio vulgaris in anaerobic environments (Yuan et al., 2013). In this study, XPS analysis was performed on the sulfur components in corrosion products (Supplementary Figure S2). After a 14-day incubation, the curve-fitted S 2p spectra indicated a complex mixture of sulfur species. The peak of SO₄ ²⁻ is likely attributed to the sulfate deposition from the culture medium (Pu et al., 2023). In P + D group, the relative amount of SO₄ ^2-, serving as the ultimate electron acceptor for SRB, is the highest. It may be attributed to the lower biofilm formation observed on coupon in the P + D group, as discussed in subsequent sections. Furthermore, the presence of organic sulfides implies that biologically derived organic proteins are likely contributing to the generation of sulfide species within the passive films (Yuan et al., 2013).

SRB could attach to the coupons, utilize the electrons, and cause MIC (Jia et al., 2019). Therefore, biomass on the coupons is an important factor for MIC, which directly determines the utilization of electrons by SRB (Vahdati et al., 2022). Therefore, this study evaluated the growth of SRB in the presence of PASP and D-Phe. The growth curves exhibited a remarkably close overlap in all groups (Supplementary Figure S3), suggesting that both PASP and D-Phe showed negligible impact on the growth of SRB. Based on this, it was speculated that the joint application of PASP and D-Phe primarily reduced corrosion via inhibiting the attachment of SRB rather than inhibiting SRB growth. Therefore, SEM and CLSM were employed to observe the biofilm on the coupons. As shown in Figure 3A and Supplementary Figure S4A, the coupons in blank group exhibited complete coverage by biofilm and corrosion products after 14-day incubation. Biofilms with varying sizes and shapes were observed with large patchy and mushroom-like structures. The cells within the biofilm were tightly embedded, displaying a high degree of alignment and minimal open gaps. Moreover, the biofilm was looser, thinner, and contained significantly more gaps in D-Phe group (Figure 3C). Also, the cellular morphology with the addition of D-Phe was different from that in blank group, aligning with other studies (Li et al., 2018; Li et al., 2021). D-Phe can alter cell morphology via substituting the original D-amino acids in peptidoglycan, which serves as a cell wall framework (Dramsi et al., 2008; Lam et al., 2009; Kolodkin-Gal et al., 2010; Carniello et al., 2018). It also affects the synthesis of flagella and adhesion-like proteins, which contribute to initial bacterial attachment and irreversible adherence, respectively (Clark et al., 2007; Karatan and Watnick, 2009; Zhang et al., 2021). Furthermore, the presence of non-uniform bacterial shapes would diminish the efficacy of biofilm formation (Young, 2006). It might be a reason that D-Phe reduced the corrosion by SRB.

Moreover, cells were sparser on the coupon in PASP group (Figure 3B), and fewer corrosion products were observed than those in blank group, which might be attributed to the protective film formed by PASP (Supplementary Figure S4B). Generally, PASP can attach to the surface of coupon in a parallel orientation via carboxyl groups and amide groups in polymer chain to form a dense protective film (Zeino et al., 2018). The functional groups in PASP can complex with iron ions and prevent their further participation in corrosion reactions, and thus inhibit the dissolution of metal ions (anodic reaction) (Xu et al., 2016; Jia et al., 2019). Besides, the protective film effectively obstructs the reaction sites for H⁺ ions, leading to the inhibition of the hydrogen evolution reaction (cathodic reaction) (Qian et al., 2013). In this study, PASP might also impede direct contact between SRB and the carbon steel surface, decreasing the electron utilization of SRB from coupon. It is because D. vulgaris primarily accepts electrons from Fe⁰ through H₂, which serves as an intermediary electron carrier (Wang et al., 2020; Woodard et al., 2023). Furthermore, in P + D group, almost no biofilm and little corrosion products were found (Figure 3D), resulting from the combined effect of PASP and D-Phe. It is consistent with results in weight loss measurements.

CLSM was employed to evaluate the distribution of cells on the coupon (Figure 3A’–D’). Live cells (green) were obviously observed while there were almost no dead cells (red) on the coupon in all groups, confirming the minimal effect of PASP and D-Phe on SRB growth. Furthermore, fluorescence intensities were recorded to evaluate the relative amounts of adhered cells. Compared to blank group, the amounts of adhered cells in PASP group reduced by 64%, which might be due to the protective film formed by PASP. Besides, it reduced by 82% and 96% in D-Phe and P + D groups, respectively, confirming that D-Phe could inhibit bacterial adhesion.

The OCP tests were first performed (Figure 4A). On Day 1, the OCP values were at a low level in all groups, resulting from minimal bacterial adhesion on the coupon surface (Starosvetsky et al., 2000). Higher OCP values were detected in the presence of PASP, indicating high resistance to corrosion, which might be due to the protective film formed by PASP. Then, the values increased sharply on Day 3 due to the formation of FeS-biofilm (FeS film underneath the biofilm) on the coupon surface (Xu et al., 2019). Besides, there were higher OCP values in D-Phe group and PASP group at most time during incubation than those in blank group, implying a higher resistance to corrosion (Vasudevan et al., 1998; Li et al., 2021). These results indicated that D-Phe and PASP reduced the electron utilization of SRB from coupon, corresponding to weight loss measurement. For the P + D group, the values of OCP did not change dramatically and were always lower than other groups. It can be explained by the combined effect of PASP and D-Phe, resulting in minimal formation of the FeS-biofilm on the coupon surface throughout the entire 14-day incubation (Xu et al., 2019).

EIS was then measured with stable OCP on Day 1, 3, 6, 10, and 14, and the Nyquist plot was used to assess the corrosion (Figures 4B–F), where the diameter in the Nyquist plot is correlated with the corrosion resistance (Gamry Instruments, 2007). On day 1, the diameters were much larger in other groups compared to blank group, which is consistent with OCP values. Moreover, the diameters in P + D group were consistently larger than those in other groups, indicating that the coupon in P + D group exhibited the highest corrosion resistance during the incubation, confirming the synergy effect between PASP and D-Phe on corrosion inhibition. Furthermore, equivalent electrical circuits were employed to fit the impedance spectra (San et al., 2014). The equivalent electrical circuits involved parameters R _s (solution resistance), Q _dl (double film capacitance), R _ct (charge transfer resistance), Q _b (biofilm capacitance), and R _b (biofilm resistance). The values of R _ct + R _b are shown in Figure 4G. There is a close correlation between R _ct + R _b value and corrosion rate, where a lower R _ct + R _b value corresponds to a higher corrosion rate. Compared to blank group, the R _ct + R _b values in other groups were similar but obviously higher on Day 1, indicating that the corrosion rates in PASP group, D-Phe group, and P + D group were close but lower than those in blank group. After 6 days-incubation, compared to the blank group, the R _ct + R _b values in D-Phe group were lower, which might be due to a larger exposed electrode area and a thinner FeS-biofilm on coupon surface. These results could be attributed to two reasons. Firstly, the D-Phe impeded the initial attachment of bacteria, thus reducing the biofilm formation (Carniello et al., 2018). Consequently, the coupon in D-Phe group displayed high corrosion resistance on Day 1, which is consistent with results in Figures 4A, B. Secondly, D-Phe induced cell wall deformation and weakened cell-to-cell connections, ultimately resulting in the collapse of the biofilm (Beech and Sunner, 2004; Castaneda and Benetton, 2008; AlAbbas et al., 2013; Bucher et al., 2015). There were high R _ct + R _b values in the presence of PASP, indicating that PASP enhanced the charge transfer resistance of coupons and thus decreased the corrosion rate. It also verified the existence of protective film formed by PASP. Moreover, the R _ct + R _b values were highest at most time during incubation in P + D group, confirming the synergy effect of PASP and D-Phe.

Potentiodynamic polarization tests were also conducted on Day 14 to evaluate the corrosion (Figure 4H), and the parameters related to potentiodynamic polarization tests were listed in Table 1. The polarization curves in PASP group and P + D group exhibited distinct variations in comparison with that in blank group, suggesting disparate reaction kinetics of the anode and cathode process. It might be attributed to the change in the electron transfer process, which was caused by attached SRB and PAPS protective film (Valcarce and Vazquez, 2010; Zhang et al., 2011; Farag and Hegazy, 2013; Zhang P. et al., 2015; Lv et al., 2022). Typically, there is a positive correlation between corrosion current and corrosion rate (Chilkoor et al., 2018). Tafel curves of both anodic and cathodic in P + D group shifted to lower corrosion current at 0.530 × 10⁻⁷ A cm^-2, significantly lower than that in blank group (1.863 × 10⁻⁷ A cm^-2), indicating that the joint application of PASP and D-Phe decreased corrosion rate. For D-Phe group, the corrosion current at 1.644 × 10⁻⁷ A cm^-2 was close to that in blank group, and the polarization curve on Day 14 was also similar to that in blank group, which might be attributed to the weak effect of D-Phe due to its decomposition over 14 days. These results correspond with weight loss measurement.

Based on the preceding discussion and results, the possible synergy mechanism of PASP and D-Phe on corrosion inhibition was summarized in Figure 5. In particular, PASP adsorbed onto the surface of coupons to form a compact protective film. This film acted as a barrier, obstructing the active sites on the coupons and inhibiting direct interaction between SRB and iron on coupon surface (Qian et al., 2013). As a result, the protective film obstructed the electron uptake by SRB from the elemental iron. Nevertheless, as the biofilm forms on the surface of the protective film, the cells within the biofilm instigated a significant challenge to the integrity of this protective film. Upon the rupture of this protective film, the corrosion is initiated, leading to the release of iron ions. The released iron ions at the corrosion sites on the carbon steel surface interacted with PASP to form complexes, effectively hindering the dissolution of metal ions. Additionally, D-Phe effectively suppressed the initial bacterial adhesion, notably extending the period for SRB to establish biofilms on the protective film. During the biofilm formation, D-Phe induced the deformation of cell walls, along with the attenuation of intercellular connections, leading to the collapse of biofilm. Thin and loose biofilms significantly diminished their aggressiveness towards the protective film. Consequently, the presence of D-Phe obviously enhanced both the efficiency and duration of PASP on corrosion inhibition.