Article pubs.acs.org/est
Modeling of Heavy Nitrate Corrosion in Anaerobe Aquifer Injection Water Biofilm: A Case Study in a Flow Rig Karine Drønen,*,‡ Irene Roalkvam,⊥ Janiche Beeder,§ Terje Torsvik,‡ Ida H. Steen,⊥ Arne Skauge,† and Turid Liengen∥ †
Uni Research CIPR, Allégaten 41, 5007 Bergen, Norway Department of Biology, Centre for Geobiology, Allégaten 41, 5007 Bergen, Norway § Statoil, Sandsliveien 90, P.O. Box 7200, 5254 Bergen, Norway ‡ Uni Research CIPR, Thormølensgt 53B, 5006 Bergen, Norway ∥ Statoil, Porsgrunn, Hydrovegen 67, 3933 Porsgrunn, Norway ⊥
S Supporting Information *
ABSTRACT: Heavy carbon steel corrosion developed during nitrate mitigation of a flow rig connected to a water injection pipeline flowing anaerobe saline aquifer water. Genera-specific QPCR primers quantified 74% of the microbial biofilm community, and further 87% of the community of the nonamended parallel rig. The nonamended biofilm hosted 6.3 × 106 SRB cells/cm2 and the S35sulfate-reduction rate was 1.1 μmol SO42−/cm2/day, being congruent with the estimated SRB biomass formation and the sulfate areal flux. Nitrate amendment caused an 18-fold smaller SRB population, but up to 44 times higher sulfate reduction rates. This H2S formation was insufficient to form the observed Fe3S4 layer. Additional H2S was provided by microbial disproportionation of sulfur, also explaining the increased accessibility of sulfate. The reduced nitrate specie nitrite inhibited the dominating H2-scavenging Desulfovibrio population, and sustained the formation of polysulfide and Fe3S4, herby also dissolved sulfur. This terminated the availability of acetate in the inner biofilm and caused cell starvation that initiated growth upon metallic electrons, probably by the sulfurreducing Desulf uromonas population. On the basis of these observations we propose a model of heavy nitrate corrosion where three microbiological processes of nitrate reduction, disproportionation of sulfur, and metallic electron growth are nicely woven into each other.
■
INTRODUCTION
Corrosion is an electrochemical process involving the transfer of electrons in the presence of an electrolyte through oxidation reactions at the anode (Fe 2 + (metal) → Fe2+(electrolyte)) and reduction reactions at the cathode (e−(free in metal) → e−(dissolved species)). The flow of metallic cations to the electrolyte will indirectly influence the flow of electrons from the metal surface and vice versa, adjusting the two rates at an equal level. Thus, the anodic reaction only proceeds if the corresponding free electrons are transferred away from the metal surface of the cathode. Only an electronic conductor or ions or molecules at nanoscopic scale can scavenge the free electrons at the metal surface. H+ ions are the best known e− recipient, where hydrogen is formed according to the formula 2H+ + 2e−→ H2. Availability of protons or electron-acceptors near the metal surface is suggested as the limiting step in corrosion.6,7 The availability
Water injection helps reservoir pressure maintenance, and hence helps to sustain desired oil production rates, during the secondary phase of oil production. In the North Sea oil production, mainly deoxygenated seawater has been injected. The high concentrations of sulfate in seawater and the O2 removal, however, pave the way for sulfate-reducing bacteria (SRB) and their production of hydrogen sulfide (H2S). The sulfide production may lead to reservoir souring, plugging of filters, and corrosion.1 The Utsira aquifer is a cratonic basin 1000 m below seafloor. The water is anaerobic, saline (4 wt %/wt), low in sulfate (0−5 mM), and fully saturated with CH4 and CO2.2 When this water was used as injection water at some oil fields in the Norwegian sector of the North Sea a strong corrosion pattern was experienced despite the low sulfate content.2 Nitrate has been used successfully to inhibit SRB activity and corrosion in oil fields practicing seawater injection.3−5 However, addition of nitrate to the aquifer water injection system increased the corrosion rate dramatically when tested in sidestream rigs.2 © XXXX American Chemical Society
Received: February 18, 2014 Revised: May 29, 2014 Accepted: June 27, 2014
A
dx.doi.org/10.1021/es500839u | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
Table 1. Relevant Microbial Processes and Their Growth Yield and Rates Ysubstrate g dry weight per mol substrate
growth rate μ h−1
Pelobacter Pelobacter Acetobacterium Acetobacterium
3.6 (8.75)a 1.9 5.6 6.1
0.09
Pelobacter Acetobacterium Acetobacterium
3.9 8.2
22 20 20
Methanolobus Methanolobus
7 3
23, 24 22, 23
Desulfobacter Desulfovibrio Desulf uromonas
4.3−4.8 2 4.2 9.8
Arcobacter Sulfurospirillum
0.75 1.5
unknown
0.1b
31
Arcobacter Sulfurospirillum Arcobacter Sulfurospirillum Desulfovibrio Acetobacterium
1.2−1.5 2.4−3.0
32 33 32 26 20
process Acetate Formation Ethylene Glycol 2HO(CH2) 2OH → CH3COOH + CH3CH2OH + H2O CH3CH2OH + H2O → CH3COOH + 2H2 2CH3CH2OH + 2HCO3−→ 3CH3COO− + 3H+ + 2H2O 5HO(CH2)2OH + 2HCO3− → 6CH3COOH + 4H2O Ethylene Glycol Monobutyl Ether 4CH3(CH2) 3O(CH2)2OH + H2O → CH3(CH2) 3OH + 2CH3COOH CH3(CH2)3OH + H2O → CH3(CH2)2COOH +2H2 4CH3OH + 2HCO3− → 3CH3COO− + H+ + 4H2O Methane Formation 4CH3OH → 3CH4 + CO2 + 2H2O 4CH3NH2 + 2H2O → 3CH4 + CO2 + 4H4+ Sulfate and Sulfur Reduction 2CH3COOH + H2SO4 → H2S + 4CO2 + 4H2 4H2 + SO42− + 2H+→ H2S + 4H2O CH3COO− + 4S0 + H2O → 2HCO3− + 4H2S + H+ CH3CH2OH + 6S0 + 3H2O → 2CO2 + 6H2S Sulfide Oxidation HS− +NO3− + H+ → NO2− + S0 + H2O HS− +NO3− + H+ → NO2− + S0 + H2O 3HS− + NO2− + 5H+ → NH4+ +3S0 + 2H2O S0 Disproportionation 4S0 + H2O → SO42− + 3HS− + 5H+ H2 oxidation H2 + NO3− → NO2− + H2O 5H2 + NO3− → NH4+ + 3H2O H2 + S0 → HS− + H+ 4H2 + SO42− → H2S + 42H2O + 2OH− 4H2 + 2HCO3− + H+→CH3COO− + 4H2O a
utilizer
2 0.68
0.09
0.03 0.5 0.14
0.06
0.5
ref
19 19, 20 21 20
25 26 27 28 29 30
Coculture with H2 scavenger. bValid for Desulfocapsa.
of H+ depends on the transport through the boundary layer near the metal surface where convection is absent and the mass transfer is by diffusion. Weak acids and buffer species such as HCO3− and H2S have been suggested as potential proton shuttles through the boundary layer and the biofilm in order to increase the proton transport from the more proton rich bulk liquid.7 Acetate in concentrations >1 mM may also be a proton source to give enhanced corrosion in oilfield environments.8 Hydrogen-utilizing bacteria are not argued as a driving force in corrosion following from the mentioned proton-limiting step at the cathode. However, bacteria may behave as electron conductors, scavenging electrons directly from the metal surface omitting the step of H2 formation. Such a mechanism has a high corrosion potential, but in the absence of a measurable chemical marker or known gene expression, the final ample evidence is difficult to obtain. Electrochemical studies of the SRB strains Desulfovibrio caldoniensis, Desulfophila corrodens (tentative), and Desulfovibrio ferrophilus (tentative) indicated however that they grew on iron directly.9−11 Iron sulfide layers may also conduct electrons from the cathode to the outer crust, and Fe° oxidation can be coupled to S° reduction in the presence of an electron conductive layer.12 In many cases the corrosion rate slows or change with time, as pH increases or as corrosion product forms a protective layer. Lee et al.13 observed that SRB-influenced corrosion varied with the ferrous iron concentration in the bulk liquid and presence/absence of an iron sulfide film prior to the
accumulation of biofilm. In the present paper we report ongoing corrosion observed in two side stream rigs for three months. The rigs were flowed with injection water comprising aquifer water, produced water (1% vol/vol), and a corrosion inhibitor (film forming). One of the rigs was also given 0.5 mM nitrate, resulting in heavy corrosion. SRBs were not associated with the heavy corrosion, and a unique case of microbialinfluenced corrosion (MIC) is modeled. For the first time microbial disproportionation of sulfur is documented to be a key reaction in heavy corrosion, a process inhibited by H2.14
■
MATERIALS AND METHODS Experimental Setup and Biological Material. Flow rig design, water qualities, electrochemical measurements, the carbon steel probes/coupons, and injection water flow conditions are described by Beeder et al.2 Corrosion products from two weight loss coupons from each rig were analyzed with X-ray diffraction after 90 days of exposure. The weight loss was determined after purification of coupons in accordance with the ASTM G1 standard. Organic material from two parallel bioprobes from each flow rig was harvested for microbial analysis by sonication (3 × 30 s). Cell pellets were stored at −20 °C. Three bioprobes from each rig were also used directly in S35-radiorespiratoric sulfate-reduction analysis. Reference strains for the quantitative polymerase chain reaction (QPCR) analysis were obtained from Deutsche Sammlung von Mikroorganismen and zellkulturen Gmb B
dx.doi.org/10.1021/es500839u | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
■
Article
RESULTS AND DISCUSSION Corrosion Rates and MIC Potential in the Flow Rigs. Two sidestream rigs were observed for corrosion in a period of 90 days, where one rig was amended with nitrate. The corrosion rate, measured as weight loss, was 1.4 ± 0.8 mm/year in the nitrate-amendment rig, a corrosion rate characterized as heavy according to the NACE RP-07-75(1999) standard. This was far higher than the moderate corrosion rate of 0.06 ± 0.01 mm/year observed in the nonamended rig (Table S3, page S6, SI). In the nitrate-amended rig, ER and LPR probes demonstrated a ∼50-days initial phase with relative stable corrosion rate 1 mM Fe2+, whereas
