Sulfate-dependant microbially induced corrosion of mild steel in the deep sea: a 10-year microbiome study

Abstract

Background: Metal corrosion in seawater has been extensively studied in surface and shallow waters. However, infrastructure is increasingly being installed in deep-sea environments, where extremes of temperature, salinity, and high hydrostatic pressure increase the costs and logistical challenges associated with monitoring corrosion. Moreover, there is currently only a rudimentary understanding of the role of microbially induced corrosion, which has rarely been studied in the deep-sea. We report here an integrative study of the biofilms growing on the surface of corroding mooring chain links that had been deployed for 10 years at ~2 km depth and developed a model of microbially induced corrosion based on flux-balance analysis.

Methods: We used optical emission spectrometry to analyze the chemical composition of the mooring chain and energy-dispersive X-ray spectrometry coupled with scanning electron microscopy to identify corrosion products and ultrastructural features. The taxonomic structure of the microbiome was determined using shotgun metagenomics and was confirmed by 16S amplicon analysis and quantitative PCR of the dsrB gene. The functional capacity was further analyzed by generating binned, genomic assemblies and performing flux-balance analysis on the metabolism of the dominant taxa.

Results: The surface of the chain links showed intensive and localized corrosion with structural features typical of microbially induced corrosion. The microbiome on the links differed considerably from that of the surrounding sediment, suggesting selection for specific metal-corroding biofilms dominated by sulfur-cycling bacteria. The core metabolism of the microbiome was reconstructed to generate a mechanistic model that combines biotic and abiotic corrosion. Based on this metabolic model, we propose that sulfate reduction and sulfur disproportionation might play key roles in deep-sea corrosion.

Conclusions: The corrosion rate observed was higher than what could be expected from abiotic corrosion mechanisms under these environmental conditions. High corrosion rate and the form of corrosion (deep pitting) suggest that the corrosion of the chain links was driven by both abiotic and biotic processes. We posit that the corrosion is driven by deep-sea sulfur-cycling microorganisms which may gain energy by accelerating the reaction between metallic iron and elemental sulfur. The results of this field study provide important new insights on the ecophysiology of the corrosion process in the deep sea.

Keywords: Corrosion; Deep sea; MIC; Mild steel; SRB.

Fig. 1

A Map of the study site, indicating the KR18-15 cruise route and dive point of the KAIKO 812 dive to collect the mooring chain links after 10 years of deployment. B Image of the mooring chain location immediately before retrieval (white arrow). JAMSTEC (2019) KAIREI KR18-15 Cruise Data. JAMSTEC. doi:10.17596/0001281 (accessed 2020-12-20)

Fig. 2

Mooring chain links demonstrating heavy localized corrosion after 10 years of exposure to deep-sea conditions. A General overview of a chain link after removal of the corrosion products. B SEM image of the corrosion products. C Intensive pitting observed at the ends of the chain link. D‑E Cross-section analyses of pitting penetration to the base material on the ends of mooring chain link, polished and etched. F Wide, localized corrosion on the chain link sides. G‑H Cross-section analyses of pitting penetration to the base material, polished, and etched surface

Fig. 3

A Alpha diversity metrics calculated based on Shannon’s diversity index, Faith’s phylogenetic diversity, and evenness based on 16S rRNA amplicon sequencing. Red refers to the mooring chain inner layer, blue refers to the mooring chain outer layer, and orange refers to the sediment. B Microbial community differences as visualized with PCoA using the unweighted UniFrac distance metrics. Axes 1 and 2 explain 37.05% and 19.63% of the variance, respectively. Red refers to the mooring chain inner layer, blue to the mooring chain outer layer and orange to the sediment. C The relative abundance of microbial species detected on the mooring chain inner and outer layers or in the sediment (depths 2‑4 cm, 4‑7 cm, and 7‑9 cm)

Fig. 4

A Comparison of binned MAGs across samples comparing mean coverage, genome completion, GC-content, and genome length. The summary was generated and visualized using “anvi-summarize.” B Genomic feature summary and taxonomic identification of MAGs in the microbial community from the deep-sea mooring chain link corrosion products. Genomic features are summarized below for each MAG including draft quality, length, number of contigs, N50, percent GC content, completeness, and contamination estimated by CheckM. Putative taxonomies were identified with GTDB-Tk including 7 main ranks (domain, phylum, class, order, family, genus, and species). The species columns were all null and not shown. MAGs are sorted by percent completeness. (C) Comparison of the percent recruitment of each MAG between samples. For each bin, a paired sample t test was calculated to compare the percent recruitment difference between paired samples from the inner and the outer layer of each mooring chain. Bold text indicates a statistically significant difference with a p value less than 0.05. D PCoA plot with Bray-Curtis dissimilarity using the percent recruitment matrix. Result revealed that the inner and outer samples clustered separately. Axis 1 explains 72.8% of the variance. Red refers to the mooring chain inner layer and blue refers to the mooring chain outer layer

Fig. 5

FBA visualization of the uptake/secretion of metabolites using iPATH. Positive uptake from the extracellular compartment is highlighted in blue. Negative uptake from the extracellular compartment is highlighted in red. The dot size is proportional to the absolute value of the rate of uptake/secretion. The community metabolic model was generated by merging 9 separate metabolic models (with MAG completeness > 80%). The highest secretion rates were observed for H⁺ and CO₂, most likely because of their role in oxidative phosphorylation, methane metabolism, energy metabolism, and urea cycling. Sulfate uptake was observed as part of the sulfur metabolism

Fig. 6

Schematic model of abiotic (black arrows) and biotic (white arrows) processes inducing corrosion of steel on the mooring chain under deep-sea conditions. In the upper panel, the abiotic production of H₂ might provide a chemical cue for the initial biofilm colonization