Characterization of bacterial community and iron corrosion in drinking water distribution systems with O3-biological activated carbon treatment

https://doi.org/10.1016/j.jes.2017.10.009Get rights and content

Abstract

Bacterial community structure and iron corrosion were investigated for simulated drinking water distribution systems (DWDSs) composed of annular reactors incorporating three different treatments: ozone, biologically activated carbon and chlorination (O3-BAC-Cl2); ozone and chlorination (O3-Cl2); or chlorination alone (Cl2). The lowest corrosion rate and iron release, along with more Fe3O4 formation, occurred in DWDSs with O3-BAC-Cl2 compared to those without a BAC filter. It was verified that O3-BAC influenced the bacterial community greatly to promote the relative advantage of nitrate-reducing bacteria (NRB) in DWDSs. Moreover, the advantaged NRB induced active Fe(III) reduction coupled to Fe(II) oxidation, enhancing Fe3O4 formation and inhibiting corrosion. In addition, O3-BAC pretreatment could reduce high-molecular-weight fractions of dissolved organic carbon effectively to promote iron particle aggregation and inhibit further iron release. Our findings indicated that the O3-BAC treatment, besides removing organic pollutants in water, was also a good approach for controlling cast iron corrosion and iron release in DWDSs.

Introduction

Cast iron pipes have been used in drinking water distribution systems (DWDSs) for over five centuries (McNeill and Edwards, 2001). The corrosion of cast iron pipes is the most common problem in DWDSs, which can cause pipe failures and the deterioration of potable water quality (Nawrocki et al., 2010, Liu et al., 2014). The chemical stability of iron corrosion scales and biofilm microbial communities in DWDSs can have a large influence on the release of iron and other corrosion products, which may result in ‘red water’ issues, particularly under the situation of source water switching (Yang et al., 2014).

Iron corrosion in DWDSs is influenced by many chemical factors in the finished water: pH, alkalinity and the concentrations of oxygen, sulfate, chloride, disinfectants, and natural organic matter (NOM) (McNeill and Edwards, 2001, Imran et al., 2005). Recently, microbiologically influenced corrosion (MIC) has attracted great interest due to its complicated role in the corrosion process, where it can either accelerate or inhibit corrosion (Li et al., 2014). Some research found that sulfate-reducing bacteria (SRB) could increase the corrosion rate (Yuan et al., 2013). However, an increasing number of studies have begun to show the protective effects of some bacterial biofilms on metal against corrosion. For example, some reports show that iron-reducing bacteria (IRB) could inhibit corrosion by reducing ferric ions to ferrous ions and enhancing the consumption of oxygen (Dubiel et al., 2002). Interestingly, the nitrate-dependent Fe(II) oxidation by some nitrate-reducing bacteria (NRB) was found to accelerate the formation of α-FeOOH, whereas the reduction of Fe(III) by other NRB promotes the formation of Fe3O4 to inhibit further corrosion and iron release (Zhang et al., 2005, Etique et al., 2014, Wang et al., 2015).

We found that the bacterial communities in DWDSs differed with different disinfectants (Wang et al., 2012, Zhu et al., 2014), which led to different cast iron corrosion patterns. Moreover, it was reported that cast iron corrosion was also different between DWDSs using ground water and DWDSs using surface water, because bacterial communities and dissolved organic carbon (DOC) concentrations were significantly different in the two water types (Li et al., 2015). As noted by Hong Wang et al., the relatively low concentrations of total organic carbon in feed water from biofiltration resulted in variations between microbial communities, such as a decrease of Alphaproteobacteria and Acidobacteria Gp3 and an increase of Gammaproteobacteria and Sphingobacteria in the biofilms of the simulated DWDSs (Wang et al., 2013). Many previous studies found that the presence of fulvic acid and humic acid (the majority components of NOM) could greatly affect the attachment or deposition of microbes on surfaces, especially metal-oxide coated surfaces (Yang et al., 2012a). Therefore, DOC could significantly influence the bacterial community in drinking water, causing differences in the further corrosion of cast iron pipes. The concentration and characteristics of DOC in distributed water depend on the source water and treatment processes of the drinking water treatment plants (DWTPs). Biological activated carbon (BAC) filtration has been used for many years in drinking water treatment, usually after ozonation (O3), and has been proven to be capable of significantly removing NOM, as well as taste and odor compounds (Simpson, 2008). Under normal operating conditions, BAC can be colonized by microorganisms, which can eventually establish active biofilms, and the formation of such biofilms can contribute to NOM removal by biodegradation (Gibert et al., 2013). Meanwhile, BAC colonized by microbes could also influence microbial numbers and the bacterial community structure in water through the release of detached biofilm cells (Han et al., 2013). O3/BAC coupling has been widely used as the main advanced drinking water treatment process prior to chlorination (Cl2) to improve water quality in DWTPs, because of its relatively low cost and high removal capacity for organic pollutants. Given the above, O3/BAC coupling may be a more promising process for improving water quality and maintaining the biological stability of water in DWDSs. However, to best of our knowledge, the previous studies have been limited to the effects of these advanced treatments on the organic matter changes in bulk water, and few studies have been conducted, thus far, to characterize their effects on bacterial communities and corrosion processes in DWDSs.

The objective of this study was to investigate the effects of O3-BAC-Cl2 on the corrosion of cast iron coupons and bacterial communities in DWDSs using annular reactors (ARs) with O3-Cl2 and Cl2 treatments as references. In the three ARs, the corrosion products and corrosion process were characterized by electrochemical and physicochemical measurements. High-throughput Illumina HiSeq DNA sequencing was used to monitor changes in the composition of the bacterial community, including corrosion-related bacteria. The quantitative real-time polymerase chain reaction (PCR) was used to monitor changes in the microbial community according to specific groups: total bacteria based on the 16S rRNA gene and denitrifiers based on the functional genes nirS, nirK and nosZ. Most probable number enumerations of NRB, IRB and nitrate-dependent Fe(II)-oxidizing microorganisms were carried out under different conditions for biofilms on corrosion scales. DOC was characterized using a total organic carbon analyzer and high-performance size exclusion chromatography (HPSEC). The relationships among organic matter, bacterial communities and corrosion were discussed.

Section snippets

Materials, preparations and model distribution systems

The elemental composition (W/W) of cast iron coupons in this study was C 3.24%, O 1.64%, Si 2.24%, P 0.08%, S 0.10%, Fe 90.47%, Cu 0.75%, Mn 0.71%, and Zn 0.77%. Prior to use, the coupons were exposed to ultraviolet (UV) light for 30 min and weighed. Granular activated carbon was washed with deionized water in an ultrasonic water bath to remove all the carbon fines, dried at 105°C for 2 days and cooled in a desiccator.

In the experiments, three annular reactors (ARs) (Model 1320LJ, BioSurface

Water quality changes under different treatments

Table 1 shows mean DOC, BDOC and AOC in water samples prior to the three ARs under different treatments. The DOC concentration in raw water (RW) was initially 2.11 mg/L, and then it increased slightly after ozonation (2.40 mg/L) due to the oxidation reaction, and decreased sharply to 1.09 mg/L from adsorption and biodegradation in the BAC filter. The same change processes were observed in the BDOC and AOC concentrations because ozone has the capacity to generate more BDOC, including AOC, and

Conclusions

Our results verified that the lowest corrosion rates and total iron concentration release occurred consistently throughout the experiment in AR-OBC compared to AR-C and AR-OC. It was found that O3-BAC pretreatment influenced the bacterial communities of DWDSs greatly, resulting in a higher rate of NRB occurrence. Moreover, the NRB in AR-OBC promoted more active Fe(III) reduction coupled to the Fe(II) oxidation redox cycle, resulting in increased Fe3O4 formation to inhibit iron corrosion and

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Nos. 51308529, 51290281).

References (54)

  • M.S. Rahman et al.

    Bench-scale evaluation of drinking water treatment parameters on iron particles and water quality

    Water Res.

    (2014)
  • D.R. Simpson

    Biofilm processes in biologically active carbon water purification

    Water Res.

    (2008)
  • H.B. Wang et al.

    Effects of microbial cycling of Fe(II)/Fe(III) and Fe/N on cast iron corrosion in simulated drinking water distribution systems

    Corros. Sci.

    (2015)
  • H.B. Wang et al.

    Effects of disinfectant and biofilm on the corrosion of cast iron pipes in a reclaimed water distribution system

    Water Res.

    (2012)
  • H.B. Wang et al.

    Effects of microbial redox cycling of iron on cast iron pipe corrosion in drinking water distribution systems

    Water Res.

    (2014)
  • H. Wang et al.

    Effect of GAC pre-treatment and disinfectant on microbial community structure and opportunistic pathogen occurrence

    Water Res.

    (2013)
  • P. Westerhoff et al.

    Nitrate removal in zero-valent iron packed columns

    Water Res.

    (2003)
  • X. Xu et al.

    Bioreduction of vanadium (V) in groundwater by autohydrogentrophic bacteria: mechanisms and microorganisms

    J. Environ. Sci.

    (2015)
  • H.Y. Yang et al.

    Influence of humic acid on the transport behavior of bacteria in quartz sand

    Colloids Surf. B

    (2012)
  • F. Yang et al.

    Morphological and physicochemical characteristics of iron corrosion scales formed under different water source histories in a drinking water distribution system

    Water Res.

    (2012)
  • F. Yang et al.

    Effect of sulfate on the transformation of corrosion scale composition and bacterial community in cast iron water distribution pipes

    Water Res.

    (2014)
  • C.Q. Ye et al.

    EIS analysis on chloride-induced corrosion behavior of reinforcement steel in simulated carbonated concrete pore solutions

    J. Electroanal. Chem.

    (2013)
  • S.J. Yuan et al.

    Surface chemistry and corrosion behaviour of 304 stainless steel in simulated seawater containing inorganic sulphide and sulphate-reducing bacteria

    Corros. Sci.

    (2013)
  • Z.F. Zhou et al.

    Responses of activities, abundances and community structures of soil denitrifiers to short-term mercury stress

    J. Environ. Sci.

    (2012)
  • Y. Zhu et al.

    Characterization of biofilm and corrosion of cast iron pipes in drinking water distribution system with UV/Cl2 disinfection

    Water Res.

    (2014)
  • R.M.W. Amon et al.

    Bacterial utilization of different size classes of dissolved organic matter

    Limnol. Oceanogr.

    (1996)
  • I.B. Beech et al.

    Microbe-surface interactions in biofouling and biocorrosion processes

    Int. Microbiol.

    (2005)
  • Cited by (15)

    • Review on corrosion and corrosion scale formation upon unlined cast iron pipes in drinking water distribution systems

      2022, Journal of Environmental Sciences (China)
      Citation Excerpt :

      Most probable number (MPN) enumerations have been employed to culture the CRB and establish their abundance (Chen et al., 2013; Wang et al., 2014). Pyrosequencing analysis of 16S rRNA genes was performed, with a high-throughput sequencing platform used to establish the microbial community composition, abundance, diversity and species of CRBs commonly (Liu et al., 2017b; Mi et al., 2015b; Prest et al., 2016; Wang et al., 2015b; Xing et al., 2018). The most widely reported CRBs in DWDSs include iron-oxidization bacteria (IOB), iron- reducing bacteria (IRB), nitrate-reducing bacteria (NRB), sulfate-oxidizing bacteria (SOB), SRB, acid-producing bacteria (APB), nitrate-reducing bacteria (NRB), and siderophore-producing bacteria (SPB) (Beech and Sunner. 2004; Li et al., 2015a; Little et al., 2014; Lytle et al., 2005b; Muehe et al., 2009; Perez-Miranda et al., 2007; Ray et al., 2010; Wang et al., 2012; Weber et al., 2006a, 2006b; Yang et al., 2014a).

    • Heavy metal enrichment in drinking water pipe scales and speciation change with water parameters

      2022, Science of the Total Environment
      Citation Excerpt :

      A long HRT allows for various complex physical, chemical, and biological reactions to occur, resulting in the formation of scale in the pipes (Echeverría et al., 2009; Sun et al., 2014a; Cui et al., 2016). When the water quality parameters change (such as a change in water source), the heavy metals in the pipe scale can be released back into the water at significant levels, placing consumers' health at risk (Li et al., 2016; Xing et al., 2018; Dargahi et al., 2016). The drinking water crisis that occurred in Flint, Michigan, USA, has attracted considerable attention, in which high blood lead levels were detected in children due to changes in source water and treatment (Hanna-Attisha et al., 2016; Pieper et al., 2017; Pieper et al., 2018).

    • Microbial corrosion of metals: The corrosion microbiome

      2021, Advances in Microbial Physiology
      Citation Excerpt :

      Step one is feasible. Many molecular studies have described the composition of the microbial community colonizing corroding surfaces based on phylogenetic classifications (Bodtker et al., 2008; Capão, Moreira-Filho, Garcia, Bitati, & Procópio, 2020; Cayford, Jiang, Keller, Tyson, & Bond, 2017; Duan et al., 2008; Duncan et al., 2014; Hirano, Nagaoka, & Matsumoto, 2020; Huttunen-Saarivirta, Rajala, Bomberg, & Carpén, 2017; Jan-Roblero, Romero, Amaya, & Le Borgne, 2004; Kip et al., 2017; Li, O'Moore, et al., 2019; Li et al., 2017; Liang et al., 2018; Liang, Davidova, Hirano, Duncan, & Suflita, 2019; Liduino et al., 2019; Moura et al., 2018; Neria-Gonzalez, Wang, Ramirez, Romero, & Hernandez-Rodriguez, 2006; Okoro et al., 2014; Oliveira et al., 2011; Park et al., 2011; Pavissich, Vargas, González, Pastén, & Pizarro, 2010; Procópio, 2020; Rajala, Raulio, & Carpén, 2019; Salerno, Little, Lee, & Hamdan, 2018; Salgar-Chaparro, Lepkova, Pojtanabuntoeng, Darwin, & Machuca, 2020; Tkachuk, Zelena, Mazur, Lukash, & Lukash, 2020; Vigneron et al., 2016; Vincke, Boon, & Verstraete, 2001; Xing, Wang, Hu, & Liu, 2018; Zhang, Tian, et al., 2019; Zhang, Fang, & Ko, 2003; Zhou et al., 2020; Zhu, Lubeck, & Kilbane 2nd., 2003; Zhu et al., 2020). However, simple correlations between the abundance of specific microbial genera and the degree of corrosion will not be sufficient to determine which microbes have corrosive capabilities or their mechanisms for corrosion.

    View all citing articles on Scopus
    View full text