Thank you for visiting Nature.com. You are using a browser version with limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In addition, to ensure ongoing support, we show the site without styles and JavaScript.
Displays a carousel of three slides at once. Use the Previous and Next buttons to move through three slides at a time, or use the slider buttons at the end to move through three slides at a time.
In freshwater environments, accelerated corrosion of carbon and stainless steels is often observed. A 22-month fresh water tank diving study was conducted here using nine grades of steel. Accelerated corrosion was observed in carbon and chromium steels and cast iron, while in stainless steel no visible corrosion was observed even after 22 months. An analysis of the microbial community showed that during general corrosion, Fe(II)-oxidizing bacteria were enriched at the early stage of corrosion, Fe(III)-reducing bacteria, at the stage of corrosion development, and sulfate-reducing bacteria, at the corrosion stage. stage in the final stage of product corrosion. On the contrary, Beggiatocaea bacteria were especially numerous in steel with 9% Cr subjected to localized corrosion. These compositions of microbial communities also differed from those in water and bottom sediment samples. Thus, as corrosion progresses, the microbial community undergoes dramatic changes, and iron-dependent microbial energy metabolism creates an environment that can enrich other microorganisms.
Metals can deteriorate and corrode due to various physical and chemical environmental factors such as pH, temperature and ion concentration. Acidic conditions, high temperatures and chloride concentrations particularly affect the corrosion of metals1,2,3. Microorganisms in natural and built environments often influence the wear and corrosion of metals, a behavior expressed in microbial corrosion (MIC)4,5,6,7,8. MIC is often found in environments such as indoor pipes and storage tanks, in metal crevices, and in soil, where it appears suddenly and develops rapidly. Therefore, monitoring and early detection of MICs is very difficult, so MIC analysis is usually carried out after corrosion. Numerous MIC case studies have been reported in which sulfate-reducing bacteria (SRB) were frequently found in corrosion products9,10,11,12,13. However, it remains unclear whether SRBs contribute to the initiation of corrosion, since their detection is based on post-corrosion analysis.
Recently, in addition to iodine-oxidizing bacteria21, various iron-degrading microorganisms have been reported, such as iron-degrading SRB14, methanogens15,16,17, nitrate-reducing bacteria18, iron-oxidizing bacteria19 and acetogens20. Under anaerobic or microaerobic laboratory conditions, most of them corrode zero-valent iron and carbon steel. In addition, their corrosion mechanisms suggest that iron-corrosive methanogens and SRBs promote corrosion by harvesting electrons from null-valent iron using extracellular hydrogenases and multiheme cytochromes, respectively22,23. MICs are divided into two types: (i) chemical MIC (CMIC), which is indirect corrosion by microbially produced species, and (ii) electrical MIC (EMIC), which is direct corrosion by electron depletion of the metal24. EMIC facilitated by extracellular electron transfer (EET) is of great interest because microorganisms with EET properties cause faster corrosion than non-EET microorganisms. While the rate-limiting response of CMIC under anaerobic conditions is H2 production via proton reduction (H+), EMIC proceeds via EET metabolism, which is independent of H2 production. The mechanism of EET in various microorganisms is related to the performance of microbial cellular fuel and electrobiosynthesis25,26,27,28,29. Because the culture conditions for these corrosive microorganisms differ from those in the natural environment, it is not clear whether these observed microbial corrosion processes reflect corrosion in practice. Therefore, it is difficult to observe the MIC mechanism induced by these corrosive microorganisms in the natural environment.
The development of DNA sequencing technology has facilitated the study of the details of microbial communities in natural and artificial environments, for example, microbial profiling based on the 16S rRNA gene sequence using new generation sequencers has been used in the field of microbial ecology30,31. ,32. Numerous MIC studies have been published that have detailed microbial communities in soil and marine environments13,33,34,35,36. In addition to SRB, enrichment in Fe(II)-oxidizing (FeOB) and nitrifying bacteria in corrosion samples, eg FeOB, such as Gallionella spp. and Dechloromonas spp., and nitrifying bacteria, such as Nitrospira, has also been reported. spp., in Carbon and copper-bearing steels in soil media33. Similarly, in the marine environment, rapid colonization of iron-oxidizing bacteria belonging to the classes Zetaproteobacteria and Betaproteobacteria has been observed for several weeks on carbon steel 36 . These data indicate the contribution of these microorganisms to corrosion. However, in many studies, the duration and experimental groups are limited, and little is known about the dynamics of microbial communities during corrosion.
Here, we investigate the MICs of carbon steel, chromium steel, stainless steel, and cast iron using immersion studies in an aerobic freshwater environment with a history of MIC events. Samples were taken at 1, 3, 6, 14 and 22 months and the corrosion rate of each metal and microbial component was studied. Our results provide insight into the long-term dynamics of microbial communities during corrosion.
As shown in Table 1, nine metals were used in this study. Ten samples of each material were immersed in a pool of fresh water. Process water quality is as follows: 30 ppm Cl-, 20 mS m-1, 20 ppm Ca2+, 20 ppm SiO2, turbidity 1 ppm and pH 7.4. The dissolved oxygen (DO) concentration at the bottom of the sampling ladder was approximately 8.2 ppm and the water temperature ranged from 9 to 23°C seasonally.
As shown in Figure 1, after 1 month of immersion in ASTM A283, ASTM A109 Condition #4/5, ASTM A179, and ASTM A395 cast iron environments, brown corrosion products were observed on the carbon steel surface in the form of generalized corrosion. The weight loss of these specimens increased with time (Supplementary Table 1) and the corrosion rate was 0.13–0.16 mm per year (Fig. 2). Similarly, general corrosion has been observed in steels with low Cr content (1% and 2.25%) with a corrosion rate of about 0.13 mm/yr (Figures 1 and 2). In contrast, steel with 9% Cr exhibits localized corrosion that occurs in gaps formed by gaskets. The corrosion rate of this sample is about 0.02 mm/year, which is significantly lower than that of steel with general corrosion. In contrast, stainless steels type-304 and -316 display no visible corrosion, with estimated corrosion rates of <0.001 mm y−1. In contrast, stainless steels type-304 and -316 display no visible corrosion, with estimated acceleration rates of <0.001 mm y−1. Напротив, нержавеющие стали типов 304 и 316 не проявляют видимой коррозии, при этом расчетная скорость коррозии составляет <0,001 мм/год. In contrast, Types 304 and 316 stainless steels show no visible corrosion, with an estimated corrosion rate of <0.001 mm/yr.相比之下，304 和-316 型不锈钢没有显示出可见的腐蚀，估计腐蚀速率<0.001 mm y−1。相比之下，304 和-316 型不锈钢没有显示出可见的腐蚀，估计腐蚀速率<0.001 mm y−1。 Напротив, нержавеющие стали типа 304 и -316 не показали видимой коррозии с расчетной скоростью коррозии <0,001 мм/год. In contrast, type 304 and -316 stainless steels showed no visible corrosion with a design corrosion rate of <0.001 mm/yr.
Shown are macroscopic images of each sample (height 50 mm×width 20 mm) before and after descaling. 1 meter, 1 month; 3 meters, 3 months; 6 meters, 6 months; 14 meters, 14 months; 22 meters, 22 months; S, ASTM A283; SP, ASTM A109, condition 4/5; FC, ASTM A395; B, ASTM A179; 1C, steel 1% Cr; 3C steel, 2.25% Cr steel; steel 9C, steel 9% Cr; S6, 316 stainless steel; S8, type 304 stainless steel.
The corrosion rate was calculated using weight loss and immersion time. S, ASTM A283, SP, ASTM A109, hardened 4/5, FC, ASTM A395, B, ASTM A179, 1C, steel 1% Cr, 3 C, steel 2.25% Cr, 9 C, steel 9% Cr, S6, type 316 stainless steel; S8, type 304 stainless steel.
On fig. 1 also shows that corrosion products of carbon steel, low Cr steel and cast iron develop further after immersion for 3 months. The overall corrosion rate gradually decreased to 0.07 ~ 0.08 mm/year after 22 months (Figure 2). In addition, the corrosion rate of 2.25% Cr steel was slightly lower than other corroded specimens, indicating that Cr can inhibit corrosion. In addition to general corrosion, according to ASTM A179, localized corrosion was observed after 22 months with a corrosion depth of about 700 µm (Fig. 3). The local corrosion rate, calculated using the corrosion depth and immersion time, is 0.38 mm/yr, which is about 5 times faster than general corrosion. The corrosion rate of ASTM A395 alloy can be underestimated as corrosion products do not completely remove scale after 14 or 22 months of water immersion. However, the difference should be minimal. In addition, many small pits were observed in the corroded low chromium steel.
Full image (scale bar: 10 mm) and localized corrosion (scale bar: 500 µm) of ASTM A179 and 9% Cr steel at maximum depth using a 3D viewing laser microscope. The red circles in the full image indicate the measured localized corrosion. A full view of the 9% Cr steel from the reverse side is shown in Figure 1.
As shown in fig. 2, for steel with 9% Cr, no corrosion was observed within 3-14 months, and the corrosion rate was practically zero. However, localized corrosion was observed after 22 months (Figure 3) with a corrosion rate of 0.04 mm/yr calculated using weight loss. The maximum localized corrosion depth is 1260 µm and the localized corrosion rate estimated using the corrosion depth and immersion time (22 months) is 0.68 mm/yr. Because the exact point at which corrosion starts is not known, the corrosion rate may be higher.
In contrast, no visible corrosion was observed on stainless steel even after 22 months of immersion. Although a few brown particles were observed on the surface prior to descaling (Fig. 1), they were weakly attached and were not corrosion products. Since the metal reappears on the stainless steel surface after the scale is removed, the corrosion rate is practically zero.
Amplicon sequencing has been performed to understand the differences and dynamics of microbial communities over time in corrosion products and biofilms on metal surfaces, in water and sediments. A total of 4,160,012 reads were received, with a range of 31,328 to 124,183 reads.
The Shannon indices of water samples taken from water intakes and ponds ranged from 5.47 to 7.45 (Fig. 4a). Since reclaimed river water is used as industrial water, the microbial community can change seasonally. In contrast, the Shannon index of bottom sediment samples was about 9, which is significantly higher than that of water samples. Similarly, water samples had lower calculated Chao1 indices and observed operational taxonomic units (OTUs) than sediment samples (Fig. 4b, c). These differences are statistically significant (Tukey-Kramer test; p-values < 0.01, Fig. 4d), indicating that the microbial communities in the sediment samples are more complex than those in the water samples. These differences are statistically significant (Tukey-Kramer test; p-values ​​< 0.01, Fig. 4d), indicating that the microbial communities in the sediment samples are more complex than those in the water samples. Эти различия статистически значимы (критерий Тьюки-Крамера; значения p <0,01, рис. 4d), что указывает на то, что микробные сообщества в образцах донных отложений более сложны, чем в образцах воды. These differences are statistically significant (Tukey-Kramer test; p values ​​<0.01, Fig. 4d), indicating that the microbial communities in sediment samples are more complex than in water samples.这些差异具有统计学意义（Tukey-Kramer 检验；p 值< 0.01，图4d），表明沉积物样本中的微生物群落比水样中的微生物群落更复杂。这些 差异 具有 统计学 （tukey-kramer 检验 ； p 值 <0.01 ， 图 4d） 表明 沉积物样本 中 的 微生物 中 中 的 群落更。。。。。。。。。 Эти различия были статистически значимыми (критерий Тьюки-Крамера; p-значение <0,01, рис. 4d), что позволяет предположить, что микробные сообщества в образцах донных отложений были более сложными, чем в образцах воды. These differences were statistically significant (Tukey-Kramer test; p-value <0.01, Fig. 4d), suggesting that microbial communities in sediment samples were more complex than in water samples. Since the water in the overflow basin is constantly renewing and sediments settle to the bottom of the basin without mechanical disturbance, this difference in microbial diversity should reflect the ecosystem in the basin.
a Shannon index, b Observed operational taxonomic unit (OTU), and c Chao1 uptake index (n=6) and basin (n=5) Water, sediment (n=3), ASTM A283 (S: n=5), ASTM A109 Temper #4/5 (SP: n=5), ASTM A179 (B: n=5), ASTM A395 (FC: n=5), 1% (1 C: n=5), 2.25% (3 C: n = 5) and 9% (9 C: n = 5) Cr-steels, as well as type 316 (S6: n = 5) and -304 (S8: n = 5) stainless steels are shown as box-shaped and whisker charts. d p-values ​​for the Shannon and Chao1 indices obtained using ANOVA and Tukey-Kramer multiple comparison tests. The red backgrounds represent pairs with p-values < 0.05. The red backgrounds represent pairs with p-values ​​< 0.05. Красные фоны представляют пары со значениями p <0,05. Red backgrounds represent pairs with p-values ​​< 0.05.红色背景代表p 值< 0.05 的对。红色背景代表p 值< 0.05 的对。 Красные фоны представляют пары с p-значениями <0,05. Red backgrounds represent pairs with p-values ​​<0.05. The line in the middle of the box, the top and bottom of the box, and the whiskers represent the median, 25th and 75th percentiles, and the minimum and maximum values, respectively.
The Shannon indices for carbon steel, low chromium steel, and cast iron were similar to those for water samples (Fig. 4a). In contrast, the Shannon indices of the stainless-steel samples are significantly higher than those of the corroded steels (p-values < 0.05, Fig. 4d) and similar to those of the sediments. In contrast, the Shannon indices of the stainless-steel samples are significantly higher than those of the corroded steels (p-values ​​< 0.05, Fig. 4d) and similar to those of the sediments. Напротив, индексы Шеннона образцов из нержавеющей стали значительно выше, чем у корродированных сталей (значения p <0,05, рис. 4d), и аналогичны индексам отложений. In contrast, the Shannon indices of stainless steel specimens are significantly higher than those of corroded steels (p-values ​​< 0.05, Fig. 4d) and are similar to deposit indices.相比之下，不锈钢样品的香农指数明显高于腐蚀钢的香农指数（p 值< 0.05，图4d），与沉积物相似。相比之下，不锈钢样品的香农指数明显高于腐蚀钢的香农指数（p 值< 0.05，图4d），与沉积物〸 Напротив, индекс Шеннона образцов из нержавеющей стали был значительно выше, чем у корродированной стали (значение p <0,05, рис. 4d), как и у отложений. In contrast, the Shannon index of the stainless steel specimens was significantly higher than that of the corroded steel (p value < 0.05, Fig. 4d), as was the deposit. In contrast, the Shannon index for steels with 9% Cr ranged from 6.95 to 9.65. These values ​​were much higher in non-corroded specimens at 1 and 3 months than in corroded specimens at 6, 14 and 22 months (Fig. 4a). Furthermore, the Chao1 indices and observed OTUs of the 9% Cr steels are higher than those of the corroded and water samples and lower than those of the non-corroded and sediment samples (Fig. 4b, c), and the differences are statistically significant (p-values < 0.01, Fig. 4d). Furthermore, the Chao1 indices and observed OTUs of the 9% Cr steels are higher than those of the corroded and water samples and lower than those of the non-corroded and sediment samples (Fig. 4b, c), and the differences are statistically significant (p-values ​​< 0.01, Fig. 4d). In addition, the Chao1 and observed OTU of steels with 9% Cr are higher than those of corroded and aqueous samples and lower than those of non-corroded and sedimentary samples (Fig. 4b, c), and the differences are statistically significant. (p-значения <0,01, рис. 4d). (p-values ​​<0.01, Fig. 4d).此外，9% Cr 钢的Chao1 指数和观察到的OTU 高于腐蚀样品和水样，低于未腐蚀样品和沉积物样品（图4b，c），差异具有统计学意义（p 值< 0.01，图4d）。此外 ， 9% CR 钢 Chao1 指数 和 观察 的 的 rtu 高于 腐蚀 样品 水样 ， 低于 腐蚀 样品 和 沉积物 （图 图 4b ， c） 差异 统计学 意义 （p 值 <0.01 图 图 图 图 图 图 图 图 ， ， ， ， ， ， ， ， ， 4d）。 Кроме того, индекс Chao1 и наблюдаемые OTU стали с содержанием 9 % Cr были выше, чем у корродированных и водных образцов, и ниже, чем у некорродированных и осадочных образцов (рис. 4b,c), а разница была статистически значимой (p- значение < 0,01, рис. 4г). In addition, the Chao1 index and observed OTU of 9% Cr steel were higher than those of corroded and aqueous samples and lower than those of uncorroded and sedimentary samples (Fig. 4b,c), and the difference was statistically significant (p-value < 0.01, Fig. 4d). These results indicate that the microbial diversity in corrosion products is lower than in biofilms on uncorroded metals.
On fig. 5a shows a Principal Coordinate Analysis (PCoA) plot based on UniFrac unweighted distance for all samples, with three major clusters observed. Microbial communities in water samples were significantly different from other communities. The microbial communities in the sediments also included stainless steel communities, while they were widespread in the corrosion samples. In contrast, the map of steel with 9% Cr is divided into non-corroded and corroded clusters. Consequently, microbial communities on metal surfaces and corrosion products are significantly different from those in water.
Principal coordinate analysis (PCoA) plot based on unweighted UniFrac distances in all samples (a), water (b), and metals (c). Circles highlight each cluster. The trajectories are represented by lines connecting the sampling periods in series. 1 meter, 1 month; 3 meters, 3 months; 6 meters, 6 months; 14 meters, 14 months; 22 meters, 22 months; S, ASTM A283; SP, ASTM A109, condition 4/5; FC, ASTM A395; B, ASTM A179; 1C, steel 1% Cr; 3C steel, 2.25% Cr steel; steel 9C, steel 9% Cr; S6, 316 stainless steel; S8, type 304 stainless steel.
When arranged in chronological order, the PCoA plots of the water samples were in a circular arrangement (Fig. 5b). This cycle transition may reflect seasonal changes.
In addition, only two clusters (corroded and non-corroded) were observed on the PCoA plots of metal samples, where (with the exception of 9% chromium steel) a shift of the microbial community from 1 to 22 months was also observed (Fig. 5c). In addition, since the transitions in corroded samples were greater than in non-corroded samples, there was a correlation between changes in microbial communities and corrosion progression. In steel samples with 9% Cr, two types of microbial communities were revealed: points at 1 and 6 months, located near stainless steel, and others (3, 14, and 22 months), located at points close to corroded steel. 1 month and coupons used for DNA extraction at 6 months were not corroded, while coupons at 3, 14 and 22 months were corroded (Supplementary Figure 1). Therefore, the microbial communities in corroded samples differed from those in water, sediment, and non-corroded samples and changed as corrosion progressed.
The main types of microbial communities observed in water samples were Proteobacteria (30.1–73.5%), Bacteroidetes (6.3–48.6%), Planctomycetota (0.4–19.6%) and Actinobacteria (0 –17.7%), their relative abundance varied from sample to sample (Fig. 6), for example, the relative abundance of Bacteroidetes in pond water was higher than in abstract water. This difference can be influenced by the residence time of the water in the overflow tank. These types were also observed in bottom sediment samples, but their relative abundance differed significantly from that in water samples. In addition, the relative content of Acidobacteriota (8.7–13.0%), Chloroflexi (8.1–10.2%), Nitrospirota (4.2–4.4%) and Desulfobacterota (1.5–4.4%) %) was higher than in water samples. Since nearly all Desulfobacterota species are SRB37, the environment in the sediment must be anaerobic. Although Desulfobacterota possibly influence corrosion, the risk should be extremely low because their relative abundances in the pool water are <0.04%. Although Desulfobacterota possibly influence corrosion, the risk should be extremely low because their relative abundances in the pool water are <0.04%. Хотя Desulfobacterota, возможно, влияют на коррозию, риск должен быть чрезвычайно низким, поскольку их относительное содержание в воде бассейна составляет <0,04%. Although Desulfobacterota may have an effect on corrosion, the risk should be extremely low as their relative abundance in pool water is <0.04%.尽管脱硫杆菌门可能影响腐蚀，但风险应该极低，因为它们在池水中的相对丰度<0.04%。 <0.04%。 Хотя тип Desulfobacillus может влиять на коррозию, риск должен быть крайне низким, поскольку их относительное содержание в воде бассейна составляет <0,04%. Although the Desulfobacillus type can influence corrosion, the risk should be extremely low as their relative abundance in pool water is <0.04%.
RW and Air represent water samples from the water intake and basin, respectively. Sediment-C, -E, -W are sediment samples taken from the center of the bottom of the basin, as well as from the east and west sides. 1 meter, 1 month; 3 meters, 3 months; 6 meters, 6 months; 14 meters, 14 months; 22 meters, 22 months; S, ASTM A283; SP, ASTM A109, condition 4/5; FC, ASTM A395; B, ASTM A179; 1C, steel 1% Cr; 3C steel, 2.25% Cr steel; steel 9C, steel 9% Cr; S6, 316 stainless steel; S8, type 304 stainless steel.
At the genus level, a slightly higher proportion (6–19%) of unclassified bacteria belonging to the Trichomonadaceae family, as well as Neosphingosine, Pseudomonas, and Flavobacterium, was observed in all seasons. As minor main components, their shares vary (Fig. 1). . 7a and b). In the tributaries, the relative abundance of Flavobacterium, Pseudovibrio, and Rhodoferrobacter was higher only in winter. Similarly, a higher content of Pseudovibrio and Flavobacterium was observed in the winter water of the basin. Thus, microbial communities in water samples varied depending on the season, but did not undergo drastic changes during the study period.
a Intake water, b Swimming pool water, c ASTM A283, d ASTM A109 temperature #4/5, e ASTM A179, f ASTM A395, g 1% Cr, h 2.25% Cr, and i 9% Cr steel , j Type-316 and stainless steel K-304.
Proteobacteria were the main constituents in all samples, but their relative abundance in the corroded samples decreased as corrosion progressed (Fig. 6). In samples ASTM A179, ASTM A109 Temp No. 4/5, ASTM A179, ASTM A395 and 1% and 2.25% Cr, the relative abundance of proteobacteria decreased from 89.1%, 85.9%, 89.6%, 79.5%, 84.8%. , 83.8% are 43.3%, 52.2%, 50.0%, 41.9%, 33.8% and 31.3% respectively. In contrast, the relative abundances of Desulfobacterota gradually increase from <0.1% to 12.5–45.9% with the progression of corrosion. In contrast, the relative abundances of Desulfobacterota gradually increase from <0.1% to 12.5–45.9% with the progression of corrosion. Напротив, относительное содержание Desulfobacterota постепенно увеличивается с <0,1% до 12,5–45,9% по мере развития коррозии. In contrast, the relative abundance of Desulfobacterota gradually increases from <0.1% to 12.5–45.9% as corrosion progresses.相反，随着腐蚀的进展，脱硫杆菌的相对丰度从<0.1% 逐渐增加到12.5-45.9%。相反，随着腐蚀的进展，脱硫杆菌的相对丰度从<0.1% Напротив, относительная численность Desulfobacillus постепенно увеличивалась с <0,1% до 12,5–45,9% по мере развития коррозии. In contrast, the relative abundance of Desulfobacillus gradually increased from <0.1% to 12.5–45.9% as corrosion progressed. Thus, as corrosion progressed, Proteobactereira was replaced by Desulfobacterota.
In contrast, biofilms on uncorroded stainless steel contained the same proportions of different bacteria. Proteobacteria (29.4–34.1%), Planctomycetota (11.7–18.8%), Nitrospirota (2.9–20.9%), Acidobacteriota (8.6–18.8%), Bacteroidota (3.1–9.2%) and Chloroflexi (2.1–8.8%). It was found that the proportion of Nitrospirota in the stainless steel samples gradually increased (Fig. 6). These ratios are similar to those in sediment samples, which corresponds to the PCoA plot shown in Fig. 5a.
In steel samples containing 9% Cr, two types of microbial communities were observed: 1-month and 6-month microbial communities were similar to those in bottom sediment samples, while the proportion of proteobacteria in corrosion samples 3, 14, and 22 increased significantly. months In addition, these two microbial communities in the 9% Cr steel samples corresponded to split clusters in the PCoA plot shown in Fig. 5c.
At the genus level, >2000 OTUs containing unassigned bacteria and archaea were observed. At the genus level, >2000 OTUs containing unassigned bacteria and archaea were observed. At the genus level, over 2000 OTUs have been observed containing unidentified bacteria and archaea. At the genus level, over 2000 OTUs have been observed containing unspecified bacteria and archaea. Among them, we focused on 10 OTUs with a high population in each sample. This covers 58.7-70.9%, 48.7-63.3%, 50.2-70.7%, 50.8-71.5%, 47.2-62.7%, 38.4 -64.7%, 12.8-49.7%, 17.5-46.8% and 21.8-45.1% in ASTM A179. , ASTM A109 Temp No. 4/5, ASTM A179, ASTM A395, 1%, 2.25% and 9% Cr steels and Type 316 and -304 stainless steels.
A relatively high content of dechlorinated monoliths with Fe(II) oxidizing properties has been observed in corrosion samples such as ASTM A179, ASTM A109 Temp No. 4/5, ASTM A179, ASTM A395 and steels with 1% and 2.25% Cr. early stage of corrosion (1 month and 3 months, Fig. 7c-h). The proportion of Dechloromonas decreased over time, which corresponded to the decrease in Proteobacteria (Fig. 6). Furthermore, the proportions of Dechloromonas in the biofilms on the non-corroded samples are <1%. Furthermore, the proportions of Dechloromonas in the biofilms on the non-corroded samples are <1%. Кроме того, доля Dechloromonas в биопленках на некорродированных образцах составляет <1%. In addition, the proportion of Dechloromonas in biofilms on uncorroded specimens is <1%.此外，未腐蚀样品的生物膜中脱氯单胞菌的比例<1%。此外，未腐蚀样品的生物膜中脱氯单胞菌的比例 < 1% Кроме того, доля Dechloromonas в биопленке некорродированных образцов была <1%. In addition, the proportion of Dechloromonas in the biofilm of uncorroded specimens was <1%. Therefore, among the corrosion products, Dechloromonas is significantly enriched at an early stage of corrosion.
In contrast, in ASTM A179, ASTM A109 tempered #4/5, ASTM A179, ASTM A395 and steels with 1% and 2.25% Cr, the proportion of SRB Desulfovibrio species finally increased after 14 and 22 months (Fig. 7c–h) . Desulfofibrion was very low or not detected in the early stages of corrosion, in water samples (Fig. 7a, b) and in non-corroded biofilms (Fig. 7j, j). This strongly suggests that Desulfovibrio prefers the environment of the formed corrosion products, although they do not affect corrosion in the early stages of corrosion.
Fe(III)-reducing bacteria (RRB), such as Geobacter and Geothrix, were found in corrosion products at the middle stages of corrosion (6 and 14 months), but the proportion of late (22 months) stages of corrosion is higher in them. relatively low (Fig. 7c, eh). The genus Sideroxydans with Fe(II) oxidation properties showed a similar behavior (Fig. 7f), so the proportion of FeOB, IRB, and SRB was only higher in the corroded samples. This strongly suggests that changes in these microbial communities are associated with corrosion progression.
In steel with 9% Cr corroded after 3, 14 and 22 months, a higher proportion of members of the Beggiatoacea family (8.5–19.6%) were observed, which can exhibit sulfur oxidizing properties, and sideroxidans were observed (8.4– 13.7%) (Fig. 1). ). 7i) In addition, Thiomonas, a sulfur oxidizing bacterium (SOB), was found in higher numbers (3.4% and 8.8%) at 3 and 14 months. In contrast, nitrate-reducing bacteria Nitrospira (12.9%) were observed in 6-month-old uncorroded samples. An increased proportion of Nitrospira was also observed in biofilms on stainless steel after dipping (Fig. 7j,k). Thus, the microbial communities of 1- and 6-month-old uncorroded 9% Cr steels were similar to those in stainless steel biofilms. In addition, the microbial communities of 9% Cr steel corroded at 3, 14 and 22 months differed from the corrosion products of carbon and low chromium steels and cast iron.
Corrosion development is usually slower in freshwater than in seawater because the concentration of chloride ions affects the corrosion of the metal. However, some stainless steels may corrode in freshwater environments38,39. In addition, MIC was initially suspected as corroded material had previously been observed in the fresh water pool used in this study. In long-term immersion studies, various forms of corrosion, three types of microbial communities, and a change in microbial communities in corrosion products were observed.
The freshwater medium used in this study is a closed tank for technical water taken from a river with a relatively stable chemical composition and a seasonal change in water temperature ranging from 9 to 23 °C. Therefore, seasonal fluctuations in microbial communities in water samples may be associated with changes in temperature. In addition, the microbial community in the pool water was somewhat different from that in the input water (Fig. 5b). The water in the pool is constantly being replaced due to overflow. Consequently, DO remained at ~8.2 ppm even at intermediate depths between the basin surface and the bottom. On the contrary, the environment of the sediment should be anaerobic, since it settles and remains at the bottom of the reservoir, and the microbial flora in it (such as CRP) should also differ from the microbial flora in the water (Fig. 6). Since the coupons in the pool were further away from the sediments, they were only exposed to fresh water during immersion studies under aerobic conditions.
General corrosion occurs in carbon steel, low chromium steel, and cast iron in freshwater environments (Figure 1) because these materials are not corrosion resistant. However, the corrosion rate (0.13 mm yr-1) under abiotic freshwater conditions was higher than in previous studies40 (0.04 mm yr-1) and was comparable to the corrosion rate (0.02–0.76 mm yr-1) in the presence of microorganisms 1) Similar to freshwater conditions40,41,42. This accelerated corrosion rate is a characteristic of MIC.
In addition, after 22 months of immersion, localized corrosion was observed in several metals under the corrosion products (Fig. 3). In particular, the localized corrosion rate observed in ASTM A179 is about five times faster than general corrosion. This unusual form of corrosion and accelerated corrosion rate has also been observed in corrosion occurring on the same object. Thus, the immersion performed in this study reflects corrosion in practice.
Among the studied metals, 9% Cr steel exhibited the most severe corrosion, with a corrosion depth of >1.2 mm, which is likely MIC because of the accelerated corrosion and abnormal form of corrosion. Among the studied metals, 9% Cr steel exhibited the most severe corrosion, with a corrosion depth of >1.2 mm, which is likely MIC because of the accelerated corrosion and abnormal form of corrosion. Среди исследованных металлов сталь с 9% Cr показала наиболее сильную коррозию с глубиной коррозии> 1,2 мм, что, вероятно, является МИК из-за ускоренной коррозии и аномальной формы коррозии. Among the metals examined, steel with 9% Cr showed the most severe corrosion with a corrosion depth >1.2 mm, which is probably the MIC due to accelerated corrosion and an abnormal form of corrosion.在所研究的金属中，9% Cr 钢的腐蚀最为严重，腐蚀深度>1.2 mm，由于加速腐蚀和异常腐蚀形式，很可能是MIC。在所研究的金属中，9% Cr Среди исследованных металлов наиболее сильно корродировала сталь с 9% Cr, с глубиной коррозии >1,2 мм, скорее всего, МИК из-за ускоренных и аномальных форм коррозии. Among the studied metals, steel with 9% Cr corroded most severely, with a corrosion depth of >1.2 mm, most likely MIC due to accelerated and anomalous forms of corrosion. Because 9% Cr steel is used in high temperature applications, its corrosion behavior has been studied previously43,44 but no MIC has been previously reported for this metal. As numerous microorganisms, except for hyperthermophiles, are inactive in a high-temperature environment (>100 °C), MIC in 9% Cr steel may be ignored in such cases. As numerous microorganisms, except for hyperthermophiles, are inactive in a high temperature environment (>100 °C), MIC in 9% Cr steel may be ignored in such cases. Поскольку многие микроорганизмы, за исключением гипертермофилов, неактивны в высокотемпературной среде (>100 °С), МИК в стали с 9% Cr в таких случаях можно не учитывать. Since many microorganisms, with the exception of hyperthermophiles, are inactive in a high temperature environment (>100°C), the MIC in steel with 9% Cr can be ignored in such cases.由于除超嗜热菌外，许多微生物在高温环境(>100 °C) 中不活跃，因此在这种情况下可以忽略9% Cr 钢中的MIC。 9% Cr 颃(>100 °C) Поскольку многие микроорганизмы, кроме гипертермофилов, не проявляют активности в высокотемпературных средах (>100 °С), МПК в стали с 9% Cr в данном случае можно не учитывать. Since many microorganisms, except for hyperthermophiles, do not show activity in high-temperature environments (>100 °C), the MIC in steel with 9% Cr can be ignored in this case. However, when 9% Cr steel is used in a medium temperature environment, various measures must be taken to reduce the MIC.
Various microbial communities and their changes were observed in deposits of uncorroded material and in corrosion products in biofilms compared to water, in addition to accelerated corrosion (Fig. 5-7), strongly suggesting that this corrosion is a microphone. Ramirez et al.13 report a 3-step transition (FeOB => SRB/IRB = > SOB) in a marine microbial ecosystem over 6 mo, wherein hydrogen sulfide produced by secondary enriched SRB may finally contribute to the enrichment of SOB. Ramirez et al.13 report a 3-step transition (FeOB => SRB/IRB => SOB) in a marine microbial ecosystem over 6 mo, when hydrogen sulfide produced by secondary enriched SRB may finally contribute to the enrichment of SOB. Ramirez et al.13 сообщают о трехэтапном переходе (FeOB => SRB/IRB => SOB) в морской микробной экосистеме в течение 6 месяцев, когда сероводород, образующийся при вторичном обогащении SRB, может, наконец, способствовать обогащению SOB. Ramirez et al.13 report a three-stage transition (FeOB => SRB/IRB => SOB) in the marine microbial ecosystem over a period of 6 months, where hydrogen sulfide generated from SRB secondary enrichment can finally contribute to SOB enrichment. Ramirez 等人13 报告了一个超过6 个月的海洋微生物生态系统中的三步转变(FeOB => SRB/IRB => SOB)，其中二次富集SRB 产生的硫化氢可能最终有助于SOB 的富集。 Ramirez 等 人 13 报告 了 个 超过 超过 6 个 月 海洋 微生物 生态 系统 中 的 三 步 转变 转变 转变 转变 转变 转变 转变 转变 转变 转变 转变 转变 转变 r srb/IRB) ， 其中 次 富集 srb 产生 硫化氢 可能 最终 有助于 sob 的富集。 Ramirez et al.13 сообщили о трехступенчатом переходе (FeOB => SRB/IRB => SOB) в морской микробной экосистеме в течение 6 месяцев, в котором сероводород, образующийся в результате вторичного обогащения SRB, может в конечном итоге способствовать обогащению SOB. Ramirez et al.13 reported a three-step transition (FeOB => SRB/IRB => SOB) in the marine microbial ecosystem over a period of 6 months, in which hydrogen sulfide produced from SRB secondary enrichment may eventually contribute to SOB enrichment. McBeth and Emerson36 reported primary enrichment in FeOB. Similarly, enrichment of FeOB during the early corrosion phase is observed in this study, but the microbial changes with the progression of corrosion observed in the carbon and 1% and 2.25% Cr steels and cast iron over 22 mo is FeOB => IRB = > SRB (Figs. 7 and 8). Similarly, enrichment of FeOB during the early corrosion phase is observed in this study, but the microbial changes with the progression of corrosion observed in the carbon and 1% and 2.25% Cr steels and cast iron over 22 mo is FeOB => IRB => SRB (Figs. 7 and 8). Точно так же в этом исследовании наблюдается обогащение FeOB на ранней стадии коррозии, но микробные изменения по мере прогрессирования коррозии, наблюдаемые в углеродистых и 1% и 2,25% Cr сталях и чугуне в течение 22 месяцев, представляют собой FeOB => IRB = > SRB (рис. 7 и 8). Similarly, in this study enrichment in FeOB at an early stage of corrosion is observed, but microbial changes as corrosion progresses, observed in carbon and 1% and 2.25% Cr steels and cast iron over 22 months, are FeOB => IRB => SRB (Figures 7 and 8).同样，在本研究中观察到早期腐蚀阶段FeOB 的富集，但在碳和1% 和2.25% Cr 钢以及超过22 个月的铸铁中观察到的微生物随着腐蚀的进展而变化是FeOB => IRB => SRB（图7 和8）。同样 ， 在 本 研究 中 观察 早期 腐蚀 阶段 feob 的 富集 ， 但 碳 和 和 1% 和 2.25% Cr 钢 超过 22 个 的 铸铁 中 到 的 微生物 腐蚀 的 进展 而 变化 FEOB => IRB => SRB（图7和8）。 Аналогичным образом, в этом исследовании наблюдалось обогащение FeOB на ранних стадиях коррозии, но микробиологические изменения, наблюдаемые в углеродистых и 1% и 2,25% Cr сталях и чугуне в течение 22 месяцев, были FeOB => IRB => SRB (рис. 7 и 8). Similarly, FeOB enrichment in the early stages of corrosion was observed in this study, but the microbiological changes observed in carbon and 1% and 2.25% Cr steels and cast iron over 22 months were FeOB => IRB => SRB (Fig. 7 and 8). SRBs can easily accumulate in seawater environments due to high sulfate ion concentrations, but their enrichment in freshwater environments is delayed by low sulfate ion concentrations. SRB enrichment in seawater has been frequently reported10,12,45.
a Organic carbon and nitrogen via Fe(II)-dependent energy metabolism iron oxide (red [Dechloromonas sp.] and green [Sideroxydans sp.] cells) and Fe(III) reducing bacteria (grey cells [Geothrix sp. and Geobacter sp. ]) at an early stage of corrosion, then anaerobic sulfate-reducing bacteria (SRP) and heterotrophic microorganisms enrich the mature stage of corrosion by consuming the accumulated organic matter. b Changes in microbial communities on corrosion-resistant metals. Violet, blue, yellow, and white cells represent bacteria from the families Comamonadaceae, Nitrospira sp., Beggiatoacea, and others, respectively.
With regard to changes in the microbial community and possible SRB enrichment, FeOB is critical in the early stage of corrosion, and Dechloromonas can obtain their growth energy from Fe(II) oxidation. Microorganisms can survive in media containing trace elements, but they will not grow exponentially. However, the plunge pool used in this study is an overflow basin, with an inflow of 20 m3/h, which continuously supplies trace elements containing inorganic ions. In the early stages of corrosion, ferrous ions are released from carbon steel and cast iron, and FeOBs (such as Dechloromonas) use them as an energy source. Trace amounts of carbon, phosphate and nitrogen required for cell growth must be present in process water in the form of organic and inorganic substances. Therefore, in this freshwater environment, FeOB is initially enriched on metal surfaces such as carbon steel and cast iron. Subsequently, IRBs can grow and use organic matter and iron oxides as energy sources and terminal electron acceptors, respectively. In mature corrosion products, anaerobic conditions enriched with nitrogen should be created due to the metabolism of FeOB and IRB. Therefore, SRB can rapidly grow and replace FeOB and IRB (Fig. 8a).
Recently, Tang et al. reported corrosion of stainless steel by Geobacter ferroreducens in freshwater environments due to direct electron transfer from iron to microbes46. Considering EMIC, the contribution of microorganisms with EET properties is critical. SRB, FeOB, and IRB are the main microbial species in the corrosion products in this study, which should have EET characteristics. Therefore, these electrochemically active microorganisms can contribute to corrosion through EET, and the composition of their community changes under the influence of various ionic species as corrosion products are formed. On the contrary, the microbial community in steel with 9% Cr differed from other steels (Fig. 8b). After 14 months, in addition to enrichment with FeOB, such as Sideroxydans, SOB47Beggiatoacea, and Thiomonas were also enriched (Fig. 7i). This change is markedly different from that of other corrosive materials, such as carbon steel, and can be influenced by chromium-rich ions dissolved during corrosion. Notably, Thiomonas has not only sulfur oxidizing properties, but also Fe(II) oxidizing properties, an EET system, and heavy metal tolerance48,49. They can be enriched due to the oxidative activity of Fe(II) and/or direct consumption of metal electrons. In a previous study, relatively high abundance of Beggiatoacea was observed in biofilms on Cu using a discontinuous biofilm monitoring system, suggesting that these bacteria may be resistant to toxic metals such as Cu and Cr. However, the energy source needed by Beggiatoacea to grow in this environment is unknown.
This study reports changes in microbial communities during corrosion in freshwater environments. In the same environment, microbial communities differed in the type of metal. In addition, our results confirm the importance of FeOB in the early stages of corrosion, as iron dependent microbial energy metabolism promotes the formation of a nutrient rich environment favored by other microorganisms such as SRB. In order to reduce MIC in freshwater environments, FeOB and IRB enrichment must be limited.
Nine metals were used in this study and processed into blocks of 50 × 20 × 1–5 mm (thickness for ASTM 395 steel and 1%, 2.25% and 9% Cr: 5 mm; thickness for ASTM A283 and ASTM A179 : 3 mm). mm; ASTM A109 Temper 4/5 and Type 304 and 316 Stainless Steel, thickness: 1mm), with two 4mm holes. Chromium steels were polished with sandpaper and other metals were polished with 600 grit sandpaper before dipping. All samples were sonicated with 99.5% ethanol, dried and weighed. Ten samples of each metal were used for corrosion rate calculation and microbiome analysis. Each specimen was fixed in a ladder fashion with PTFE rods and spacers (φ 5 × 30 mm, Supplementary Fig. 2).
The pool has a volume of 1100 cubic meters and a depth of about 4 meters. The water inflow was 20 m3 h-1, the overflow was released, and the water quality did not fluctuate seasonally (Supplementary Fig. 3). The sample ladder is lowered onto a 3 m steel wire suspended in the middle of the tank. Two sets of ladders were removed from the pool at 1, 3, 6, 14 and 22 months. Samples from one ladder were used to measure weight loss and calculate corrosion rates, while samples from another ladder were used for microbiome analysis. Dissolved oxygen in the immersion tank was measured near the surface and bottom, as well as in the middle, using a dissolved oxygen sensor (InPro6860i, Mettler Toledo, Columbus, Ohio, USA).
Corrosion products and biofilms on the samples were removed by scraping with a plastic scraper or wiping with a cotton swab, and then cleaned in 99.5% ethanol using an ultrasonic bath. The samples were then immersed in Clark’s solution in accordance with ASTM G1-0351. All samples were weighed after drying was completed. Calculate the corrosion rate (mm/yr) for each sample using the following formula:
where K is a constant (8.76 × 104), T is exposure time (h), A is total surface area (cm2), W is mass loss (g), D is density (g cm–3).
After weighing the samples, 3D images of several samples were obtained using a 3D measuring laser microscope (LEXT OLS4000, Olympus, Tokyo, Japan).