The effect of Pseudomonas fluorescens on buried steel pipeline

Subscriber access provided by University of Rochester | River Campus & Miner Libraries

Article

The effect of Pseudomonas fluorescens on buried steel pipeline corrosion Amy J. Spark, Ivan S. Cole, Adam S Best, David W. Law, and Liam P Ward Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b00437 • Publication Date (Web): 21 Jun 2017 Downloaded from http://pubs.acs.org on June 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 17

Environmental Science & Technology

1

The effect of Pseudomonas fluorescens on buried steel pipeline corrosion.

2

Amy J. Spark1

3

School of Engineering, RMIT University, GPO Box 2476, Melbourne VIC 3001 Australia

4

Ivan S. Cole2

5

Manufacturing, CSIRO, Research Way, Clayton VIC 3168 Australia

6

Adam S. Best

7

Manufacturing, CSIRO, Research Way, Clayton VIC 3168 Australia

8

David W. Law

9


10

Liam P. Ward*

11


12

*email: [email protected], Tel: +61 3 9925 1713, Fax: +61 3 9639 0138

13

Abstract

14 15 16 17 18 19 20 21 22 23 24 25 26

Buried steel infrastructure can be a source of iron ions for bacterial species, leading to microbiologically influence corrosion (MIC). Localised corrosion of pipelines due to MIC is one of the key failure mechanisms of buried steel pipelines. In order to better understand the mechanisms of localised corrosion in soil, semi-solid agar has been developed as an analogue for soil. Here, Pseudomonas fluorescens has been introduced to the system to understand how bacteria interact with steel. Through electrochemical testing including open circuit potentials, potentiodynamic scans, anodic potential holds and electrochemical impedance spectroscopy it has been shown that P. fluorescens increases the rate of corrosion. Time for oxide and biofilms to develop was shown to not impact on the rate of corrosion but did alter the consistency of biofilm present and the viability of P. fluorescens following electrochemical testing. The proposed mechanism for increased corrosion rates of carbon steel involves the interactions of pyoverdine with the steel, preventing the formation of a cohesive passive layer, after initial cell attachment, followed by the formation of a metal concentration gradient on the steel surface.

27

Supplementary information

28 29 30

Supplementary information accompanies this manuscript providing a more detailed account of the experimental methodology and calculations used to complete the study. It also contains some additional results.

31

Introduction

32 33 34 35 36

Buried infrastructure such as pipelines, tanks and building foundations are essential to the function of society. Much of this infrastructure, particularly pipelines, are made from iron alloys which are susceptible to corrosion1. In the presence of microbes, corrosion can be accelerated. This phenomenon is known as microbiologically influenced corrosion (MIC)2. While much is known of MIC in a variety of systems, work to date has primarily focused on solution based systems such as marine 1

Current Address: AECOM, Collins Square, Melbourne VIC 3001 Australia Current Address: Research and Innovation, RMIT University, GPO Box 2476, Melbourne VIC 3001 Australia

2

ACS Paragon Plus Environment


37 38

waters where the problems are most prevalent. However, in soil there remains a lot still to be understood.

39 40 41 42 43 44 45 46 47 48

Many bacterial species rely on iron for a range of metabolic processes including building proteins and as such have developed ways to take up iron ions from the environment3. Iron is the fourth most abundant element on earth and comprises 5% of soil, but due to its high affinity to oxygen it is primarily found in Fe(III) ion form, not as a solid metal4, 5. There are many different ways the dependence on iron and restrictions on availability of iron in the environment can lead to bacteria influencing the progression of corrosion for iron based metals2. Some bacteria are able to take up electrons as an energy source directly from an iron surface, oxidising the iron in the process6 but most scavenge iron through the production of molecules known as siderophores3. Escherichia coli is known to produce siderophores and scavenge iron similar to the Pseudomonas family7. These are both Gram negative aerobic bacteria species that are prevalent in soil and non-stagnant water.

49 50 51 52 53 54 55 56 57 58 59

The heterogeneous nature of soil is a complicating factor when it comes to undertaking experiments of soil based corrosion, especially where microbes are involved. Soil is a three phase system (solid, liquid, gas) that is porous8. The size of the soil pores influences both the rate of metal corrosion within the soil and the ability of bacteria to move through the soil. Bacteria are restricted to pores within soil because they are only able to move with liquid, they cannot move through the soil particles or gas particles9. Bacteria can move most readily in loose soil with large, interconnected pores. The corrosion rate is related to pore size through the formation of triple phase boundary (TPB) zones on the metallic surface8. As the length of total TPB zones on a metal surface increases, the liquid dispersion across the surface increases and the corrosion current density increases in a linear fashion. The geometry of these TPB zones is affected by soil type, soil compression and water content.

60 61 62 63 64 65 66 67 68 69 70 71 72

Traditionally, soil corrosion and MIC studies are conducted in solutions or in soil samples taken from a site of interest1, 10. Both of these systems have significant drawbacks as a method of studying the mechanisms of corrosion in soil11. Solution systems are unable to replicate the physical properties of soil and soil samples are hampered by their need to be removed from the ground which changes the compression of the soil (and thus the pore structure) and makes moisture levels difficult to maintain. It is also difficult to obtain reproducible results from soil samples. As a consequence of these limitations, agar has been investigated as a soil analogue to enable fundamental corrosion studies where the physical characteristics of soil as well as the chemical characteristics are incorporated in the system while also allowing for control of the system12. Previous studies have optimised this system with semi solid agar and investigated the effect of varying concentrations of peptide nutrients on steel13. The semi-solid agar system has been found to be electrochemically stable within a given potential range, give reproducible results, is able to be used at a range of temperatures and behaves in a manner representative of clay based soils.

73 74 75 76 77 78 79

This paper discusses electrochemical tests on carbon steel using the semi-solid agar system with and without P. fluorescens present. P. fluorescens is easily cultured, forms a biofilm in isolation and has been previously associated with increased corrosion of steel14. Two sets of experiments were conducted. In the first, open circuit potential (OCP) experiments of 3, 24 and 48 hours were conducted followed by potentiodynamic scans (PDS). Electrochemical impedance spectroscopy (EIS) scans were undertaken immediately upon the agar being set and following the OCP hold. In the second, OCP experiments of 0, 24 and 48 hours were conducted followed by 48 hour potential holds


Page 2 of 17

Page 3 of 17

80 81


at -600mV vs SCE. Through this study further knowledge of the interactions of bacteria with steel in semi-solid agar and the advantages of semi-solid agar as an analogue for soil is gained.

82 83 84

Materials and Methods

85 86 87 88

Hot rolled HA1 carbon steel sheet 1mm thick was cut into 25mm2 coupons and polished to a final finish with 1µm diamond paste. The composition and polishing process for the steel is given in previous publications12, 13. Immediately prior to setting up an experiment, the exposed face of the coupons was sterilised with 30 minutes exposure to UV.

89 90 91 92 93 94 95 96 97

Gram negative P. fluorescens, strain ATCC49642, culture was taken directly from a nutrient agar plate and mixed with distilled water and matched against MacFarland Equivalence Turbidity Standard no. 4 to give an estimated cell concentration of 12.0x108ml-1 so that a consistent concentration of bacteria was used for each experiment15. The bacteria suspension was swabbed across the surface of the sterilised carbon steel samples to ensure the presence of bacteria on the metal surface at the start of the electrochemical experiments. The samples were allowed to air dry in a sterile biological safety cabinet for approximately 10 minutes before being attached to a standard flat cell with 1cm2 of the swabbed surface exposed. More details of the flat cell arrangement and bacterial culture preparation are provided in the supporting information.

98 99 100 101

Agar with 12.5g/L of peptide nutrients and 2.5g/L NaCl, prepared as described in the supporting information, was poured into the completed flat cell as a solution and then allowed to set (approximately 2.5 hours) before commencing experiments. Triplicate experiments were conducted under all conditions.

102 103 104 105 106 107 108 109

For the first set of experiments (Figure SI-S1), an EIS scan from 1MHz to 20 MHz with a sinus amplitude of ±50mV vs OCP was conducted with three measurements per frequency. The OCP was then measured for 3, 24 or 48 hours. After the OCP measurement, a second EIS scan was conducted with identical parameters. This was followed by a PDS conducted between ±400mV vs OCP at a scan rate of 0.167mV/s16. Tafel plots taken from the PDS were used to determine the corrosion potential and corrosion currents for each experiment. Data from the EIS were used to calculate properties of the biofilms present on the surface and the calculations used are described in the supporting information.

110 111 112 113 114 115 116 117 118 119 120

In the second set of experiments, a series of scans was conducted as outlined in Table SI-S1. Potential holds (PH) were conducted over 48 hours commencing when the agar had just set, after 24 hours of OCP and after 48 hours of OCP. Here, the current response was measured as a function of time at a PH value of -600mV vs SCE. These were compared to the same steel without bacteria exposed to a potential hold of -600mV vs SCE under the same conditions immediately upon the agar being set13. Samples were subsequently examined using scanning electron microscopy (SEM). To examine the biofilms and oxide layers present on the surface, samples were removed from the flat cell with agar remaining on the surface, fixed and chemically dehydrated with the process detailed in the supporting information. After drying the steel coupons were mounted on stubs and sputter coated with gold approximately 200 angstroms thick. The bare metal surface without any bioflm present, was also examined with SEM. These samples were removed from the flat cell and rinsed



121 122

with ethanol to remove the biofilm and agar present. They were then air dried. All samples were examined at 30kV in a FEI Quanta 200 Environmental SEM used in high vacuum mode.

123 124 125 126

Results and Discussion

127

Open circuit potentials

128 129 130 131 132 133 134 135 136 137

OCP curves over 3, 24 and 48 hour periods prior to PDS with and without bacteria present are shown in Figure 1. The purpose was to establish the free standing corrosion behaviour and any influence associated with biofilm and / or corrosion product formation. Two key effects on the OCP curves are seen with the addition of bacteria to the semi-solid agar system; a slight negative shift in potential and a reduction in electrochemical stability demonstrated by increased fluctuations in the data obtained. With increasing hold time there are no significant changes in potential, and the trend established over three hours is continued over 24 and 48 hours. This data indicates that with the addition of bacteria, the thermodynamic potential for corrosion to occur is increased and this is likely to be due to a biological or non uniform effect. Similar trends were seen in the OCP scans for the potential hold samples (not shown).

138 139 140

Figure 1: Open circuit potential curves over 3 hours (blue), 24 hours (red) and 48 hours (green) with and without bacteria present. Darker shades are used for samples without bacteria.

141

Electrochemical impedance spectroscopy (EIS)

142 143

EIS was undertaken both before and after hold at OCP for each of the conditions to determine the impact of the different OCP times and the presence of bacteria on the interface between the steel


Page 4 of 17

Page 5 of 17

144 145 146 147 148 149 150 151 152


working electrode and the semi-solid agar. The results of these scans are shown in Figure 2. Both before and after hold at OCP, the steel surface was behaving as a simple resistor-capacitor system indicating the formation of a corrosion and/or biofilm interface17. Before OCP, the Nyquist plot for each of the conditions was very similar. After OCP a distinct trend in the resistance and impedance values measured was established. With increasing length of OCP, a larger semi-circle was formed indicating a more resistive layer formed with longer hold times. There was an additional increase in impedance and resistance measured with the addition of bacteria to the system. As with the OCP results, there was a greater amount of variation between experimental runs with bacteria than without shown by the resistance standard deviations given in Table SI-S2.



153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175

Results of calculations based on the EIS data after OCP, as shown in Table SI-S2, demonstrate that the addition of bacteria increases resistance and decreases conductivity at the steel-agar interface, indicating that the bacteria are enhancing the protective nature of the oxide/ nutrient layer formed or are forming an insulating layer. This correlates with the cathodic behaviour of the samples seen in the PDS, but is contradictory to the OCP values recorded at the end of each hold period. Examining the capacitance values calculated suggests that the charge storing capability of the oxide layer without bacteria is independent of the OCP time whereas with bacteria present capacitance increases with increasing hold time. After 3 hours the capacitance is 4.6x10-5F, which increases to 5.6x10-5F after 24 hours and 6.8x10-5F after 48 hours. The increase in capacitance is an indication of an increase in extracellular polymeric substance being produced by the P. fluorescens which is known to bind metal ions, forming metal concentration cells on the surface and increasing charge storage18. The calculations also show there is a decrease in biofilm thickness, seen as hold time as bacteria increases, that is not seen in the samples without bacteria, indicating that with the addition of bacteria the biofilm at the steel-agar interface is being compressed and becoming more compact over time, which could be due to the secretion of EPS. The 48 hour samples are the only samples where the biofilm thickness measured with bacteria is thinner than without, suggesting that after 48 hours the biofilm has stabilised and that at 3 hours and 24 hours while the biofilm is developing it is open and patchy with ready flow of oxygen and other electrolytes to the steel surface. This would increase the rate of corrosion of the steel. It also suggests that for those samples with bacteria competitive layer formation is evident, with biofilm formation dominating over oxide scale formation. The interplay of oxide scale, nutrients and biofilm over time changes the dominant contributor to corrosion but the overall effect on corrosion remains constant with the addition of bacteria.

176


Page 6 of 17

Page 7 of 17


177 178

Figure 2: EIS data for all conditions with and without P. fluorescens a) as set, b) after OCP (3 hrs, 24 hrs, 48 hrs)

179

Potentiodynamic scans

180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195

PDS tests were conducted to establish the response of the steel samples to accelerated corrosion conditions and the influence of any biofilm and / or oxide layer formation after varying OCP hold times. PDS taken following OCP for the three different times (Figure 3) show that the addition of P. fluorescens and increased hold time have some influence on the cathodic and anodic behaviour of the steel. In the anodic region, higher current values were measured with bacteria present than without bacteria, though above -550mV vs SCE in the 48 hour hold sample this trend is reversed. The 3 hour and 48 hour hold samples showed a marginal decrease in corrosion potential with the addition of bacteria, but this was not seen in the 24 hour samples. The slight negative shift in corrosion potential combined with the increase in corrosion current experienced at different anodic potentials with the addition of P. fluorescens, suggest that the bacteria are increasing both the thermodynamic potential and the kinetic potential for corrosion to occur on carbon steel in semisolid agar. P. fluorescens has been previously associated with increased corrosion of steel, though it was seen to have a minor effect when studied as a monoculture14 which is reflected by the marginal changes seen in these results. However, the changes seen in the level of this effect with varying hold times suggests that the build up of oxide and biofilm on the surface is also playing a role in the process of corrosion as discussed in more detail with the EIS results below.

196 197 198 199 200 201 202 203 204 205

The PDS show a negative shift in corrosion potential with a longer OCP prior to the scan as well as a decrease in cathodic current recorded independent of the presence of bacteria. This suggests that by increasing the length of OCP prior to PDS, more nutrients are able to build up on the surface and provide a protective layer against corrosion through the formation of an oxide/nutrient layer. Previous studies conducted in semi-solid agar have determined that the peptide nutrients used in this experiment have an inhibitive effect on the corrosion of carbon steel. In the cathodic region, there is a marginal shift with the addition of bacteria to a lower cathodic current. This suggests that the bacteria forming a biofilm on the surface is also contributing to the formation of a protective layer. The interplay between the effect of a nutrient oxide layer with and without bacteria present and the corrosivity of the bacteria is of great interest.



Page 8 of 17

206 207 208

Figure 3: Averaged potentiodynamic scans of samples exposed to P. fluorescens compared with those which were not. a) 3 hours OCP, b) 24 hours OCP, c) 48 hours OCP, d) all conditions

209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224

The addition of P. fluorescens to the system led to a slight increase in the corrosivity of the steel tested under all of the conditions. This is shown by the scans in Figure 3 as well as the averaged OCP values and the averaged corrosion potential and corrosion current values listed in the supporting information. The OCP with bacteria is consistently 10 to 15mV more negative than without bacteria. This is irrespective of the time for which the OCP was held suggesting that the presence of a biofilm has little bearing on this value and it is the presence of the bacterial cells affecting the OCP. Averaged corrosion potential values, as shown in Table 1, are consistently more negative and averaged corrosion current measurements are consistently higher with the addition of bacteria further supporting the finding that P. fluorescens has a corrosive effect on steel in the semi-solid agar environment. The least variation in corrosion potential and corrosion current is seen in the samples which underwent a 24 hour OCP. The standard deviations for these values with bacteria are also greater compared to the 3 hour OCP and 48 hour OCP samples. This lack of variation between bacteria and without bacteria after 24 hours for a biofilm to form suggests that at this time point the bacteria are having less interaction with the steel. This suggests that there are time periods during biofilm development where the bacteria are more or less active and have more or less impact on their surrounding environment.

225 226

Table 1: Average (± standard deviation) open circuit potential, corrosion potential and corrosion current values for samples with and without P. fluorescens present

3 hours

Length of OCP 24 hours


48 hours

Page 9 of 17


OCP (mV) Ecorr (mV) icorr (µA)

No bacteria Bacteria No bacteria Bacteria No bacteria Bacteria

-715 ± 3 -728 ± 4 -782 ± 1 -801 ± 8 0. 50±0.03 0.70±0.08

-712 ± 2 -728 ± 10 -839 ± 21 -843 ± 21 0.31±0.06 0.34±0.13

-719 ± 13 -729 ± 2 -822 ± 18 -853 ± 2 0.21±0.11 0.59±0.11

227 228

Potential hold measurements

229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244

PH measurements were taken after varying OCP hold times to establish the tenacity of any biofilm and / or passive layer formation on the steel substrate and influence on the corrosivity of the system. Potential hold data for all of the samples are shown in Figure 4. It is clear that the samples exposed to bacteria are experiencing higher levels of current than the sample not exposed to bacteria. The amount of current appears to be independent of the time given for biofilm to develop as all three samples with bacteria show similar amounts of current with the exception of the first 1030 minutes of the potential hold where different peak values are seen, decreasing with increased OCP hold time. Figure 4 also indicates the standard deviation for each OCP hold time. The sample which allowed 48 hours for the biofilm to develop has the smallest standard deviation across the potential hold of the samples exposed to bacteria. The sample not exposed to bacteria showed the least deviation of all. It is hypothesised that the increase in variation between experiments is due to uneven electrochemical activity across the steel surface with the action of bacteria. The standard deviation is greatest in the bacteria sample without hold time as there has been no time for a biofilm to form between cells to give electrochemical connection between cells. With increased hold-time there is more time for biofilm to form, which provides more electrochemical connection and thus less variation between experiments.



245 246 247 248 249 250

Figure 4: Time vs current plot over 48 hours in response to a potential of -600mV vs SCE. Three samples had bacteria at the steel surface with a variety of biofilm development times; they show a marked increase in current when compared to the sample without bacteria. They are a) no bacteria as set, b) bacteria as set, c) bacteria following a 24 hr OCP, d) bacteria following a 48 hr OCP, and e) the four graphs overlaid.

251

Scanning electron microscopy

252 253 254 255 256 257 258 259 260 261

Figure 5 shows two sets of samples (fixed biofilm on the left and metal surface with biofilm removed on the right) prepared for SEM imaging following the potential hold experiments. These in turn were compared to SEM of the samples from an earlier nutrients study without bacteria13. Based on the SEM images of the metal surface with fixed biofilm (Figure 5 a,c,e), it is believed there is a high level of EPS present, especially in the 48 hour samples. The biofilm changes with time, after 0 hours (Figure 5 a) it appeared to be loose with very little lattice structure. At 24 hours (Figure 5 c) it is quite compact with bacteria enmeshed with fibrous particles. By 48 hours (Figure 5 e) it is less compact, but is highly ordered and appeared to have incorporated oxides as well as having many bacterial cells and EPS fibres present. This would both increase the resistivity and capacitance of the biofilm, as suggested by the EIS data. An earlier study of the corrosion effects of P. fluorescens found low


Page 10 of 17

Page 11 of 17

262 263 264 265 266


levels of EPS produced after four weeks of submersion in a culture broth at 37°C14. The variation seen between these samples and the earlier studies could be a result of the support provided by agar as compared to the solution used in the earlier studies. Additionally, P. fluorescens is an environmental bacteria species that is optimally grown at 25-30°C19 and would thus not be as viable at 37°C and be less likely to produce EPS at such a high temperature.



267 268 269 270 271

Figure 5: SEM of samples exposed to bacteria following potential hold a) fixed 0 hrs OCP b) metal surface 0 hrs OCP c) fixed 24 hrs OCP d) metal surface 24 hrs OCP e) fixed 48 hrs OCP f) metal surface 48 hrs OCP. The highlighted section from c) depicts groups of bacteria congregating on the surface; these are seen in the circled regions. Scale bar = 10µm

272 273 274 275

The samples without bacteria, shown in Figure 6, indicate that corrosion has occurred and there is an etching effect. There appears to be some oxide layer present on the surface, but back scatter imaging (Figure 6 b) suggests that the bulk of the surface is iron metal. This is indicated by the lack of contrast seen and mostly light surface.


Page 12 of 17

Page 13 of 17


276 277 278

Figure 6: SEM images of sample with no bacteria following potential hold with 0 hours OCP a) secondary electron image b) back scattered image. Scale bar = 10µm

279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296

A similar etching effect is seen in the samples with bacteria that were rinsed with ethanol prior to SEM (Figure 5 b,d,f). Grains of the steel were clearly visible and the etching effect is more pronounced suggesting it is further advanced. This supports the PDS and potential hold findings that with bacteria, higher corrosion rates are seen. An additional feature seen in the bacteria samples not seen in the samples without bacteria or the PDS samples are gouges within the grains (crevice like features in the SEM images of the exposed surface). These appear to be in clumps and not follow a pattern. With increasing time for biofilm formation they appear to be getting deeper. Examining the samples with biofilm fixed on the surface (Figure 5 a,c,e) indicate localised corrosion is occurring and that this is contributing to the accelerated corrosion occurring in the presence of bacteria. All of the biofilms appear to be cracking due to the drying process along similar lines to the grain boundaries seen in the exposed samples. The image of the 24 hour sample biofilm (Figure 5 c) shows several groups of bacteria clumping together. These clumping patterns are similar to the gouge patterns seen in the exposed steel images suggesting that the bacteria are causing those effects. Earlier work on steel found increased etching in the presence of P. fluorescens at OCP when exposed for four weeks and that the bacteria appeared to clump around haematite crystals 14. As the corrosion initiated by the agar environment is occurring primarily along grain boundaries, the majority of oxides are produced at this point. Thus the bacterial cells, which require Fe(III) to replicate, will congregate at these areas and accelerate corrosion in their immediate vicinity.

297

Detailed discussion

298 299 300 301 302 303 304 305 306 307 308

OCP, PDS and potential hold data indicate that the P. fluorescens present is increasing both the thermodynamic potential for corrosion of the steel to occur and its rate of corrosion. There is a minor increase in thermodynamic potential for corrosion to occur with increased time for the biofilm to form as shown by the more negative OCP values and PDS, but this is not reflected in the potential hold experiments as the rate of corrosion (0.04mA to 0.06mA) appears to be very similar across all samples. The EIS data suggests that with longer hold times and the addition of bacteria, the biofilm formed following OCP is more resistive and less conductive. However, with the addition of bacteria for the 3 hour samples and 24 hour samples the capacitance of the biofilm is less than without bacteria. The question here becomes, what is the mechanism by which the bacteria contributes to the corrosion of steel. Biofilms and EPS producing bacteria, such as the Pseudomonads, are commonly believed to cause corrosion through interference with passive layers and the creation of



309 310 311 312 313

metal concentration cells18, 20. The Pseudomonads are known to produce EPS and isolates of P. fluorescens from soil have been shown to form complex, heterogeneous biofilm structures without other species present21. Biofilm forming metal concentration cells is proposed as a potential mechanism for the corrosion seen in these experiments, particularly for the 48 hour OCP samples which would have formed a stable biofilm prior to PDS.

314 315 316 317 318 319 320 321 322 323 324 325 326

The peptide based nutrients used in this experiment have been shown to have an inhibitive effect on steel due to the amine and carboxyl groups present in the peptides13. The corrosive effect of the bacteria seen here may be due to nutrients being consumed by the bacteria, reducing the inhibition of the steel corroding and increasing corrosion rates22. A distinct change in corrosion rate sometime after the commencement of an MIC experiment is often due to a change in nutrient conditions. In this instance it is unlikely to be the dominant contributor to the increase in corrosion seen, given that increasing the time for biofilm to form, and thus the length of time for nutrients to be consumed, is not leading to a difference between corrosion rates measured over the period of the potential hold experiments. However, it may be contributing to the decrease in OCP seen with increasing time for biofilm to form. There is also the potential for localised nutrient consumption to be occurring at the grain boundaries where the bacteria have been shown to be congregating. This could be leading to the formation of localised galvanic corrosion cells and thus to accelerated localised corrosion.

327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343

P. fluorescens has previously been implicated in the increase of corrosion of steel14. In this study, clumps of cells were found around iron oxide crystals as well as etching of the steel surface with gouge like effects seen in the potential hold samples (Figure 5). It was hypothesised that the enhanced corrosion seen with the addition of P. fluorescens could be attributed to the action of metabolic products. Pseudomonads produce an iron chelating enzyme known as pyoverdine which binds to Fe(III) and allows the cell to take up the Fe(III) and convert it to Fe(II) which is required for the production of proteins and cell replication3. P. fluorescens particularly produces pyoverdine in low iron environments, such as that experienced when initially exposed to the semi-solid agar7. The peptide nutrients used do contain some iron23, but when halved the concentration of iron is low, especially compared to the nutrient agar the bacteria were grown on. This could explain why the samples which underwent the potential hold as soon as the agar was set showed the highest spike in current produced initially as the bacteria would be producing the most pyoverdine at that time point having just experienced a change in iron concentration. Abundant Fe(III), the insoluble final product of iron corrosion under aerobic conditions, has been shown to increase the growth rate of P. fluorescens but also reduces the amount of pyoverdine secreted by the cells24. Understanding the ways in which P. fluorescens interacts with iron assists in determining a mechanism for the observed increase in corrosion.

344 345 346 347 348 349 350 351 352

The higher thermodynamic and kinetic potential for corrosion of steel seen in the presence of P. fluorescens, across all three OCP times, can be explained through a combination of these mechanisms. Initially, high levels of pyoverdine would have been secreted by the bacteria cells which would have bound the Fe(III) produced by the non-biotic corrosion occurring in the system. The removal of Fe(III) from the oxide layers would have both prevented a cohesive passive film from forming and caused a shift in chemical equilibrium of the corrosion process leading to an increase in reduction of the iron metal occurring. With increased OCP time, excess Fe(III) would have been produced leading to a decrease in pyoverdine synthesised and less uptake of the Fe(III). However, concurrently a more stable biofilm would have formed which had a higher level of capacitance


Page 14 of 17

Page 15 of 17


353 354 355 356 357

indicating the formation of metal concentration cells. Metal concentration cells would also increase the corrosion levels. The shift in corrosion mechanism of P. fluorescens from the action of metabolites following a 3 hour OCP and 24 hour OCP to being due to the biofilm formed after a 48 hour OCP also gives a possible explanation for the observed change in anodic behaviour above 600mV vs SCE.

358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375

P. fluorescens has been previously shown to be negatively affected by the application of an anodic potential, with growth of a biofilm restricted after 8 hours25. The bacteria under these conditions do not produce sufficient extracellular polymeric substance (EPS) and cells clump together rather than distributing evenly across a surface. The clumping seen so clearly in the potential hold 24 hr OCP images (Figure 5 c) correlates with this. Without time for a biofilm to form before the potential is applied, over the 48 hour period an anodic potential is applied here, there is the possibility that the bacteria are killed indicating that an anodic potential is, as previously suggested, a negative indicator for cell life. However, at the potential applied once a biofilm was able to form, it was able to protect the bacteria from death as suggested by the increase in oscillation in PH data seen for the 0 and 24 hr OCP samples which was not seen in the 48 hr sample. The biofilm can be seen to change with increased time for formation, with more cell and EPS matter present in a coherent form at 48 hours than either 0 hours or 24 hours which suggests a reduced amount of EPS contributes to the increase in oscillations seen for the 0 and 24 hour samples. Interestingly, other than increasing the oscillations in the potential hold data, time for biofilm to form has not impacted the amount of current produced. This suggests that the presence of the biofilm is contributing to the increase in corrosion, possibly through the creation of oxygen and/or metal concentration cells on the steel surface. This mechanism has been suggested previously for Pseudomonas species influencing corrosion on steel26.

376 377 378 379 380 381 382 383 384 385 386 387 388

A range of electrochemical tests have been conducted on steel in a semi solid agar environment containing peptide based nutrients with and without P. fluorescens present. It was shown that the addition of P. fluorescens increased the thermodynamic and kinetic potential for corrosion, irrespective of the time for biofilm development. Increased etching with additional gouge like regions was seen on the surface of samples with bacteria present. This can be attributed to clumping of bacterial cells along the grain boundaries where iron oxides form. It is proposed that two mechanisms are contributing to the increase in corrosion seen with the addition of P. fluorescens to the system. At shorter OCP times (0, 3 hours and 24 hours) pyoverdine, a metabolite produced by the bacteria, is binding the Fe(III) in the system and allowing the bacteria to transport the Fe(III) across the cell wall. This results in a chemical drive for more Fe(III) to be produced leading to higher rates of corrosion. At longer OCP times (48 hours) when Fe(III) is readily available and the production of pyoverdine would have decreased, the biofilm is believed to be stable and to have formed a metal concentration gradient leading to an increase in corrosion.



389

390 391 392

Figure 7: proposed mechanism for increased corrosion rate with the addition of P. fluorescens to the agar- steel system.

393

Acknowledgements

394 395 396 397 398

The authors acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the RMIT Microscopy & Microanalysis Facility, at RMIT University. The authors also acknowledge CSIRO and RMIT for the use of their facilities and continued support. They would also like to thank Parveen Sangween and Jacinta White of CSIRO with their assistance in performing the bacterial experiments and preparing samples for SEM.

399

References

400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425

1. Cole, I. S.; Marney, D., The science of pipe corrosion: A review of the literature on the corrosion of ferrous metals in soils. Corrosion Science 2012, 56, 5-16. 2. Little, B. J.; Lee, J. S., Microbiologically influenced corrosion. Hoboken, N.J. : WileyInterscience: Hoboken, N.J., 2007. 3. Cornelis, P., Iron uptake and metabolism in pseudomonads. Appl Microbiol Biotechnol 2010, 86, (6), 1637-1645. 4. Neilands, J. B., Iron absorption and transport in microorganisms. Annual Review of Nutrition 1981, 1, (1), 27-46. 5. Andrews, S. C.; Robinson, A. K.; Rodríguez-Quiñones, F., Bacterial iron homeostasis. FEMS Microbiology Reviews 2003, 27, (2-3), 215-237. 6. Bose, A.; Gardel, E. J.; Vidoudez, C.; Parra, E. A.; Girguis, P. R., Electron uptake by ironoxidizing phototrophic bacteria. Nature Communications 2014, 5, 1-7. 7. Meyer, J.-M., Pyoverdines: pigments, siderophores and potential taxonomic markers of fluorescent Pseudomonas species. Arch Microbiol 2000, 174, (3), 135-142. 8. Jiang, J.; Wang, J.; Wang, W.; Zhang, W., Modeling influence of gas/liquid/solid three-phase boundary zone on cathodic process of soil corrosion. Electrochimica Acta 2009, 54, (13), 3623-3629. 9. Chenu, C.; Stotzky, G., Interactions between microorganisms and soil particles: an overview. In Interactions between soil particles and microorganisms: Impact on the terrestrial ecosystem, Huang, P. M.; Bollag, J.-M.; Senesi, N., Eds. John Wiley & Sons, Ltd: Chichester, England, 2002; pp 340. 10. Usher, K. M.; Kaksonen, A. H.; Cole, I.; Marney, D., Critical review: Microbially influenced corrosion of buried carbon steel pipes. International Biodeterioration & Biodegradation 2014, 93, (0), 84-106. 11. Spark, A.; Ward, L.; Cole, I. S.; Marney, D., Studying the synergistic effects of microbes and environment on carbon steel pipeline corrosion. In Corrosion & Prevention, Australasian Corrosion Association (ACA): Brisbane, 2013; p Paper 57.


Page 16 of 17

Page 17 of 17


426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460

12. Spark, A. J.; Cole, I.; Law, D.; Marney, D.; Ward, L., Investigation of agar as a soil analogue for corrosion studies. Materials and Corrosion 2016, 67, (1), 7-12. 13. Spark, A.; Cole, I.; Law, D.; Ward, L., The effect of peptide based nutrients on the corrosion of carbon steel in an agar based system. Corrosion Science 2016, 110, 174-181. 14. Moreno, D. A.; Ibars, J. R.; Beech, I. B.; Gaylarde, C. C., Biofilm formation on mild steel coupons by Pseudomonas and Desulfovibrio. Biofouling 1993, 7, (2), 129-139. 15. Jadhav, S.; Shah, R.; Bhave, M.; Palombo, E. A., Inhibitory activity of yarrow essential oil on Listeria planktonic cells and biofilms. Food Control 2013, 29, (1), 125-130. 16. Gu, K.; Lv, L.; Lu, Z.; Yang, H.; Mao, F.; Tang, J., Electrochemical corrosion and impedance study of SAE1045 steel under gel-like environment. Corrosion Science 2013, 74, 408-413. 17. Mansfeld, F., Electrochemical impedance spectroscopy. In Analytical Methods In Corrosion Science and Engineering, CRC Press: 2005; pp 463-505. 18. Beech, I. B., Corrosion of technical materials in the presence of biofilms—current understanding and state-of-the art methods of study. International Biodeterioration & Biodegradation 2004, 53, (3), 177-183. 19. Lens, P., Biofilms in medicine, industry and environmental biotechnology. IWA Publishing: 2003. 20. Olivares, G. Z.-.; Mejia, G. M.; Caloca, G. G.; Lopez, I. G.; Dabur, F. R.; Ulloa-Ochoa, C. M.; Esquivel, R. G., Sulfate reducing bacteria influence on the cathodic protection of pipelines that transport hydrocarbons. In CORROSION 2003, NACE International: San Diego, California, 2003. 21. Baum, M. M.; Kainović, A.; O'Keeffe, T.; Pandita, R.; McDonald, K.; Wu, S.; Webster, P., Characterization of structures in biofilms formed by a Pseudomonas fluorescens isolated from soil. BMC Microbiology 2009, 9, (1), 1-13. 22. Little, B. J.; Lee, J. S., Microbiologically influenced corrosion: an update. International Materials Reviews 2014, 59, (7), 384-393. 23. Oxoid Dehydrated Culture Medium: Nutrient Broth No.2. http://www.oxoid.com/UK/blue/prod_detail/prod_detail.asp?pr=CM0067 (13th November), 24. Rachid, D.; Ahmed, B., Effect of iron and growth inhibitors on siderophores production by Pseudomonas fluorescens. African Journal of Biotechnology 2005, 4, (7), 697-702. 25. Busalmen, J. P.; de Sánchez, S. R., Electrochemical polarization-induced changes in the growth of individual cells and biofilms of Pseudomonas fluorescens (ATCC 17552). Applied and Environmental Microbiology 2005, 71, (10), 6235-6240. 26. Abdolahi, A.; Hamzah, E.; Ibrahim, Z.; Hashim, S., Localised corrosion of mild steel in presence of Pseudomonas aeruginosa biofilm. Corrosion Engineering, Science and Technology 2015, 50, (7), 538-546.

461 462

For table of contents only


The effect of Pseudomonas fluorescens on buried steel pipeline

Recommend Documents