SEARCH

SEARCH BY CITATION

Keywords:

  • microbiological corrosion;
  • pitting corrosion;
  • seawater;
  • stainless steel

Abstract

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

The potential effects of seawater ingress into 316L lined pipes during subsea tie-in operations on corrosion performance were investigated. Immersion and accelerated corrosion tests were conducted on 316L in different mixtures of treated seawater. In particular, we examined the effect of oxygen and microorganisms in seawater on the performance of the alloy at the different mixtures of treated seawater to assess the risk of localized corrosion in the event of seawater ingress into pipelines. Results showed that oxygen has a negative impact on the biocidal and oxygen scavenging efficiency of the chemical treatments and a detrimental effect on pitting corrosion.

1 Introduction

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

Commissioning deep-water pipelines is a standard part of any subsea installation. Once a pipeline or infield flowline is laid offshore, it must be tested to ensure that it will perform as designed and that it was not damaged during installation. Pipeline commissioning typically involves flooding with treated seawater, cleaning, gauging, and hydrostatic testing. Hydrostatic testing is a routine practice to verify that pressure equipment does not leak or have manufacturing flaws. Treated seawater is routinely used in the hydrotesting of subsea pipelines. In the oil and gas industry, it is often the case that hydrostatic test water is left in the system for many months before the system is actually commissioned. Natural seawater contains viruses, prokaryotes (bacteria and archaea), protists, and algae. During this holding time, the activity of residual microorganisms can increase, as the effectiveness of the preservation chemicals decays, and also if there was any ingress of raw seawater that mixes with the volumes of treated fluids. In addition, this stagnant water may permit debris such as sand, marine life, and bacteria introduced by poor water treatment or during tie-in operations to settle and form biofilms. Under-deposit corrosion and/or microbiologically influenced corrosion (MIC) may then occur [1-6]. These biofilms may present a serious threat once the pipelines become operational, because fluids transported in pipelines may contain sufficient nutrients for bacteria to flourish [7, 8].

Dissolved oxygen (DO) in the seawater is one of the most aggressive species towards corrosion of metallic materials. The degree to which DO influences corrosion is dependent on the metal or alloy [9]. Corrosion resistant alloys (CRAs) are becoming important structural materials for offshore applications because they possess an appropriate combination of strength and corrosion resistance in marine environments. These materials have negligible general corrosion in aerated water due to a protective, predominantly chromium oxide, film which forms immediately on their surface with exposure to oxygen [9]. High oxygen content is favorable for passive film-forming alloys because it delays the corrosion initiation at surface defects and penetration through oxide protective films. However, in aerated seawater, surface deposits on passive film-forming alloys can create oxygen concentration cells, which can cause pitting and/or crevice corrosion at localized sites [10]. Once pitting is initiated, the propagation rate is accelerated with increasing DO content [11].

Corrosion caused by any of these mechanisms may reduce pipeline and equipment service life. The severity of the problem depends upon the chemistry and microbiology of the water that is used, the length of time that the water remains in the line and the temperature of the system. In order to protect against these adverse effects, seawater used for hydrotesting requires an appropriate preservation treatment [12, 13]. Seawater treatments include filtration and appropriate dosages of chemical treatments. Filtration reduces the amount of sediment and nutrients entering the pipeline, decreasing the risk of MIC and under-deposit corrosion. Chemical treatments may include an oxygen scavenger, a biocide and a corrosion inhibitor. Tetrakishydroxymethyl phosphonium sulfate (THPS) is the preferred biocide since it is effective against microorganisms while having a favorable environmental profile that allows easy disposal offshore [13]. Ammonium bisulfite (ABS) is the recommended oxygen scavenger for hydrotesting waters [12]. Biocide and oxygen scavengers dosages are provided by chemical vendors and depend upon predicted times of seawater containment in the system.

During installation of equipment for offshore gas fields, flowlines filled with treated seawater need to have end caps removed to tie-in with subsea equipment. This operation will require the connection point on the flowline to be opened, thus exposing the flowline ends directly to seawater. It is expected that the majority of these operations are within 12 h exposure time. However, several feasible scenarios such as vessel position loss and vessel breakdown, tie-in tooling failure or damage to subsea hardware or unforseen unfavorable weather conditions can result in significantly longer exposure and it is possible that a flowline may be exposed to seawater for durations in excess of 24 h. In the case of breakdown of dive systems or poor weather conditions, e.g. cyclones, it is possible to experience durations of >1 week exposure for the worst-case scenario.

This study aimed to investigate the potential effects of the ingress of raw untreated seawater into 316L lined pipes containing hydrostatic test water during subsea tie-in operations on the pitting resistance of the alloy. This used a combination of immersion tests, accelerated corrosion tests, and 3D optical surface imaging. In particular, we examined the effect of oxygen and sulfate-reducing bacteria (SRB) on the corrosion performance of the steel. The presence of SRB is of great concern to oil and gas industry worldwide due to its widely recognized involvement in the corrosion of materials [14-17]. Microbiological analyses were conducted to evaluate biocide efficiency and to assess the likelihood of MIC in the event of seawater ingress into pipelines. Results from this study will allow a risk assessment on the likelihood of localized corrosion events related to tie-in operations and better definition of the time window of tolerance to seawater ingress to the pipeline in the event of any upsets during tie-in operations.

2 Materials and methods

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

2.1 Specimens preparation

316L stainless steel coupons used for electrochemical analysis and immersion tests were cut into small coupons having 5 and 6 cm2 surface active area, respectively. The composition (in wt%) of this material was: Cr 16.92, Ni 10.11, Mo 2.05, N 0.05; Mn 1.38, C 0.016, S 0.001, Fe bal. Prior to exposure, coupons were wet ground using silicon carbide papers of 120, 360, 600, and 1200 grit consecutively. The polished specimens were washed with Milli-Q water, degreased with ethanol and dried with nitrogen gas. For exposure testing, coupons were sterilized by immersion in ethanol 70% for 1 h.

2.2 Seawater solutions

Seawater samples were collected at 20 m depth off Rottness Island (Western Australia) in the Indian Ocean. The chemical composition of the seawater is given in Table 1. Hundred percent treated seawater corresponded to 100 ppm of ABS oxygen scavenger and 550 ppm of tetrakis (hydroxymethyl) phosphonium sulfate (THPS) in 60 µm filtered seawater. From this mixture, 20, 40, 60, and 80% mixtures of treated seawater were prepared. The oxygen scavenger was applied first and allowed to react for 30 min prior to addition of biocide to the system to avoid incompatibilities between chemical treatments [18]. Raw untreated natural seawater was used as experimental control.

Table 1. Analysis of the natural seawater used in this study
AnalysisComposition
Salinity (PSU)35.58
DO (mL/L)5.06
Conductivity (mS/cm)48.79
pH8.2
Chloride (mg/L)18500
Magnesium (mg/L)1340
Sodium (mg/L)11100
Sulfate (mg/L)2700

2.3 Electrochemical testing

For electrochemical testing, the Gamry Intruments Flexcell™, a crevice-free pitting evaluation system was used to evaluate general and pitting corrosion on 316L in the various concentrations of treated seawater. A double junction Ag/AgCl electrode and a platinum coated mesh were used as reference and counter electrode, respectively. Details on the experimental set-up to investigate pitting corrosion using this cell is described elsewhere [19]. Solution temperature was achieved internally using a recirculating water bath connected to the cell through a Teflon-coated copper coil. A 150 rpm agitation rate was sustained using a Teflon rotator. A nitrogen blanket was maintained throughout the test to avoid oxygen into the system and to help stabilize the open circuit potential (OCP).

The linear polarization resistance (LPR) method was used to monitor corrosion rates over 5 h. LPR measurements were performed by applying anodic voltage scans at the rate of 0.1 mV/s over a range of ±10 mV around stabilized OCP. The Gamry DC105 software was used to calculate corrosion rates from LPR data. A linear fit of the current versus voltage data to a standard model yields an estimate of the polarization resistance (Rp). Rp is then used to calculate Icorr and corrosion rate.

Cyclic potendiodynamic polarization (CPP) scans tests were conducted to determine pitting potential (Epit) and repassivation potential (Erep) and their relationship to the OCP of 316L in the various mixtures of treated seawater. Scans were conducted using a forward and reverse scan rate of 0.167 mV/s. Epit was identified as the potential where the anodic current indicated the onset of stable pitting and Erep was identified as the potential for which the forward and reverse scans intersect, which is where repassivation of pits is considered to take place. CPP tests were conducted in duplicate for each condition.

2.4 Immersion tests

316L coupons were exposed to treated seawater and experimental conditions as described in Table 2. Immersion test A was intended to evaluate the performance of the steel in different concentrations of treated seawater in the event of high levels of oxygen ingress into the system. For immersion test B and D, a nitrogen blanket was formed on top of the exposure cells by continuous injection of high rates of nitrogen gas via a gas flow meter to prevent oxygen contamination and evaluate the efficiency of the specified treated seawater mixtures in the absence of oxygen. The outlet flow of nitrogen from the system was monitored by a second flow meter to ensure there were no leaks in the system. Immersion test B was particularly aimed to evaluate the efficiency of 100% treated seawater on preventing MIC in the event of high loading of SRB into the system and on protecting against pitting corrosion in the absence of oxygen.

Table 2. Summary of the test conditions to investigate the effect of various levels of seawater ingress into 316L lined pipes
TestExposure solution Treated seawater/raw untreated seawaterConditionsExposure timeAnalyses
  1. LPR, linear polarization resistance; CPP, cyclic potentiodynamic polarization scan; SRB, sulfate-reducing bacteria.

Accelerated corrosion tests100/0Stirring; 20 °C5 hLPR, CPP
 80/20No O2  
 60/40   
 40/60   
 20/80   
 0/100 (control)   
Immersion test A100/0 + SRBStagnant; 20 °C7, 10, 14, 21, 28 daysWeight loss
 80/20 + SRBFree O2 ingress Surface/pitting analysis
 60/40 + SRB  DAPI staining for bacterial adhesion
 40/60 + SRB   
 20/80 + SRB   
 0/100 + SRB (control)   
Immersion test B100/0 + SRBStagnant; 20 °C7, 14, 28 daysSurface/pitting analysis
  No O2 DAPI staining for bacterial adhesion
Immersion tests C100/0Stagnant; 20 °C7, 14, 28 daysSurface/pitting analysis
  ∼2 ppm O2 DAPI staining for bacterial adhesion
Immersion tests D80/20Stagnant; 20 °C7, 14, 28 daysSurface/pitting analysis
  No O2 DAPI staining for bacterial adhesion

Immersion test C was aimed to evaluate the efficiency of the chemical treatments and the performance of the alloy in the event of low concentrations of oxygen ingress into the system which also allows evaluating the effect of the residual biocide on controlling MIC. For this test, the desired DO concentration was controlled to reach maximum values of 2 ppm. The solution was initially flushed with sterile air for 30 min. Then, no gas was bubbled through the water and an air blanket was formed on top of the cells by injection of high rates of air via a gas flow meter for 1 h. The system was then sealed to allow the oxygen in the gas phase to dissolve into the solution. The DO was monitored continuously until its concentration was close to 2 ppm (approximately 2 days). At this point, a nitrogen blanket was formed on top of the cell as described above for tests B and D. No gas was bubbled through the water during the entire exposure to maintain stagnant conditions. All exposure cells were maintained inside an incubator set at 20 °C. Several analyses were conducted after defined exposure periods (Table 2). DO levels were measured using an Orbisphere 3655 oxygen analyzer from Hach Company before and after the exposure periods and pH of testing solutions was monitored using an Orion 5-star plus portable multimeter from Thermo Fisher Scientific, Inc.

2.5 SRB inoculum

The SRB population used as inoculum was isolated from natural seawater using the anaerobic nutrient-rich Starkey medium contained in anaerobic jars. Starkey medium contains (g/L): KH2PO4 0.5; NH4Cl 1; Na2SO4 1; CaCl2 · 2H2O 0.1; MgSO4 · 7H2O 2; sodium lactate 50% 10 mL; filter-sterilized seawater 1 L. The pH of the medium was adjusted to 7.5. SRB were grown at 20 °C. Active SRB cells were obtained by transferring grown cultures to fresh Starkey medium every 72 h. The bacterial count in the inoculum was estimated by serial dilution method [20] which indicated a population of 108 cell/mL. Treated seawater was inoculated with 10% SRB inoculum when necessary (Table 3). The numbers of SRB in the raw natural seawater used for immersion tests as estimated by standard serial dilution method was 103 cell/mL.

Table 3. Enumeration of SRB attached to 316L after exposure to the various levels of seawater at the four immersion tests for 28 days. SRB enumeration was conducted by serial dilution method
Immersion testExposure solution Treated seawater/raw untreated seawaterSRB count (cell/mL)
A100/0None
 80/20103
 60/40<10
 40/60<10
 20/80<10
 0/100104
B100/0None
C100/0None
D80/20None

2.6 Surface analysis by 3D optical microscopy

Surface inspection and pit profile measurements to evaluate localized corrosion of the tested alloys were conducted using an optical infinite focus microscope (IFM G4g system, from Alicona Imaging GmbH).

2.7 Evaluation of bacterial adhesion and SRB enumeration

To evaluate bacterial attachment, 316L surfaces were stained with 4,6-diamidino-z-phenylidole (DAPI), a fluorescent stain that binds strongly to DNA. Coupons were removed from the reaction vessel, rinsed with sterile water, stained with DAPI (2 µg/mL) and incubated in the dark at room temperature for 15 min. DAPI-stained samples were examined with an epifluorescence microscope (Axio Imager.A1; Carl Zeiss, Germany) equipped with a Plan-Neofluar objective.

Enumeration of sessile SRB (attached to coupons) was conducted only on 28 days exposure coupons. To detach sessile SRB, coupons were removed from the reactions bottles, immersed in sterile seawater containing filter-sterilized Tween 20 solution (0.1% w/v final concentration) and sonicated (series of 90-s sonication steps). Suspensions were then filtered using 0.22 µm membrane filters. Membrane filters with concentrated SRB cells were then placed into SRB enumeration medium (Postgate C medium) [21]. SRB numbers were estimated by the serial dilution method [20]. The growth of bacteria is indicated by the deposition of a black precipitate in the dilution vial as a result of bacterial metabolic activity.

3 Results and discussion

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

3.1 Electrochemical testing

Electrochemical testing was conducted to investigate uniform and localized corrosion on 316L in the various mixtures of treated seawater. Average corrosion rates from LPR measurements are shown in Fig. 1. It can be seen that the corrosion rates were very low and trended to decrease with time for all the levels of seawater. Corrosion rates in seawater were very similar regardless of the concentration of the chemical treatments. The slight decrease in the corrosion rate observed with longer testing is considered to be due to the enrichment of Cr in the passivating surface layers. These low corrosion rates indicate that uniform corrosion is insignificant but it is not a measure of the pitting tendency of the alloy.

image

Figure 1. Corrosion rates as a function of time of 316L exposed to different concentrations of treated seawater at 20 °C calculated from LPR measurements

Download figure to PowerPoint

CPP tests were conducted to determine the susceptibility to localized corrosion of 316L in the various levels of treated seawater [22]. Average corrosion potential (OCP), Epit and Erep obtained from duplicate CPP tests of 316L exposed to the various mixtures of treated seawater at 20 °C are shown in Fig. 2. The gap OCP-Epit indicates that a high anodic potential is required for 316L to undergo stable pitting. In practice, pitting is often observed after a very long time (months, years) at potentials lower than Epit. In this region of potential, metastable pits can be formed. It appears that during aging periods, the protectiveness of the passive film on the alloys deteriorates and transition to stable pitting occurs. Crevice corrosion that occurs more readily than pitting can be expected at potentials close to the Erep. Erep is a very important critical potential indicating the range of potentials below which pitting will not occur [23]. Pitting and crevice corrosion are usually considered to take place if the corrosion potential of a metal in a given environment surpasses the crevice repassivation potential [24]. It has been demonstrated that when OCP>Erep pitting is expected and when OCP<Erep the material is protected against pitting. Based on the above statements, results from the present study indicate that pitting could easily initiate if a slight anodic potential (Erep-OCP) is attained during exposure. It can be seen that the localized corrosion resistance of 316L SS in seawater is very similar regardless of the levels of chemical treatments. However, these results must be carefully interpreted for assessing the possibility of pitting initiation considering the influence of environmental factors and aging effects on the pitting initiation mechanisms.

image

Figure 2. Average OCP, Epit and Erep of 316L exposed to different concentrations of treated seawater at 20 °C. Potential were identified from cyclic potentiodynamic polarization tests

Download figure to PowerPoint

3.2 Immersion tests

3.2.1 Dissolved oxygen (DO) and pH measurements

DO concentration measured throughout exposure for the different immersion tests is shown in Fig. 3. For immersion tests A, DO concentration remained high regardless of the chemical treatment (Fig. 3a). 316L exposed to raw seawater showed a decrease of DO concentration with time. This oxygen depletion could be due to microbial activity in the raw untreated seawater. Previous studies on the compatibility of THPS and ABS showed that THPS had the ability to deactivate the ABS [18]. It was shown that the detection of THPS and ABS residual values after exposure in the presence of oxygen indicated that the two chemicals had not reacted with each other but that the THPS interferes with the oxygen scavenging process by ABS. The authors concluded that the residence time between addition of ABS (to scavenge oxygen) and its subsequent contact with THPS should be maximized in order to prevent the two molecules from having a negative impact on each other. In the present study, ABS was added to the system 30 min prior to the addition of the THPS, which was expected to provide enough time for the oxygen scavenging process to take place.

image

Figure 3. DO measurements as a function of time for (a) immersion test A; (b) immersion test C. N2 indicates when a nitrogen blanket was set on top of cells to avoid oxygen ingress and to control the desired levels of DO and (c) immersion test B and D. A nitrogen blanket was maintained throughout exposure to avoid oxygen ingress

Download figure to PowerPoint

DO measurements for immersion tests B and D are shown in Fig. 3b. DO levels for test B remained below 20 ppb throughout exposure. DO levels in test D started at 200 ppb and decreased during the first 5 days of exposure to 40–60 ppb. Afterwards, DO levels remained fairly steady until the completion of the exposure indicating that nitrogen blanket provided an effective mechanism to prevent oxygen ingress into the system so that oxygen scavenging took place effectively. For test C, where the desired DO concentration was set to maximum ∼2 ppm, the solution was initially flushed with sterile air until the DO reached 500 ppb (approx. 30 min). The system was sealed to allow the oxygen in the gas phase to dissolve in the solution. When DO levels reached ∼2 ppm, a nitrogen blanket was initiated to avoid any oxygen ingress into the system. When DO levels were below 1000 ppb, the nitrogen blanket was stopped and the exposure cells where properly sealed again. This procedure was maintained throughout exposure in order to achieve the specified oxygen levels (Fig. 3c). Once the nitrogen blanket was stopped, DO levels raised from 0.2 to 2 ppm in 2–3 days.

Figure 4 shows the pH of the seawater in immersion test A. For immersion tests B, C, and D, pH values were very similar to those detected in immersion test A at the respective seawater mixtures evaluated. pH slightly decreased during the first 7 days of exposure and then remained fairly steady throughout exposure time.

image

Figure 4. pH measurements of different concentrations of treated seawater throughout exposure

Download figure to PowerPoint

3.2.2 Surface analysis of microbial adhesion by DAPI fluorescent dye

4′6-diamidino-2-phenylindole (DAPI) fluorescent dye was used to stain bacteria adhered to 316L at the different exposure times. The aim of this analysis was to study the biocidal efficiency against microbial adhesion of the various mixtures of treated seawater and the different exposure conditions. Killing the adhered microorganisms does not necessarily mean their removal from the surface. However, this method can be used to evaluate the first steps of bacterial adhesion therefore providing information on the early killing efficiency of a biocide. Figures 5 and 6 show images of DAPI stained 316L observed under an epifluorescence microscope. Images revealed that microorganisms attached to 316L exposed to the different mixtures of seawater in immersion test A regardless of the exposure time (Fig. 5). Bacterial cells on 316L were typical rod-shaped bacterial cells mostly forming patches of colonies on the surface. The formation of typical biofilms [25] (cells embedded in self-produced extrapolymeric substance (EPS)) was not evident. It was noted that bacterial cells on 316L exposed to raw seawater differed in size and fluorescence to cells attached to coupons exposed to treated seawater which can be related to some inhibition effects by residual chemical treatments. Microbial adhesion slightly increased with exposure time for all the different levels of treated seawater in immersion test A. For immersion tests B, DAPI stained 316L coupons revealed microorganisms attached to the surface on 7 and 14 days exposure coupons but microorganisms were not observed on the 28 days coupons. For immersion test B-28 day, whole microorganisms were not observed on DAPI stained coupons but instead, DNA remnants such as microbial debris were observed on the surface. This may indicate microorganisms were affected and disintegrated by the chemical treatment with exposure time (Fig. 6). For immersion tests C and D, microorganisms were not observed on the surface regardless of the exposure time.

image

Figure 5. Images of DAPI stained 316L coupons under an epifluorescence microscope for immersion test A. Images show bacterial attached to coupons exposed to (a) raw seawater and (b) 100% treated seawater, for up to 7 days. Bacteria were found attached to coupons from 7 to 28 days exposure regardless of the concentration of the chemical treatment

Download figure to PowerPoint

image

Figure 6. Images of DAPI stained 316L coupons under an epifluorescence microscope for immersion test B. (a) bacteria attached to 316L exposed to 100% treated seawater for up to 7 days. (b) debris and DNA remains on 316L exposed to 100% treated seawater for 28 days

Download figure to PowerPoint

3.2.3 Enumeration of sessile SRB by standard serial dilution technique

Enumeration of viable sessile SRB attached to 316L exposed to the various levels of seawater immersion test A, B, C, and D for 28 days was conducted using SRB culture enumeration medium and the standard serial dilution method. Results are summarized in Table 3. For immersion test A, viable and cultivable sessile SRB were detected on 316L exposed to raw seawater and 20–80% treated seawater but were not detected on 316L exposed to 100% treated seawater. Viable SRB were not detected on coupons from immersion test B, C, and D.

The SRB counts on 316L exposed to raw seawater (control) were 104 cell/mL. This indicates the SRB numbers inoculated at the beginning of the test (108 cell/mL) decreased with time. In seawater, sulfate is reduced to sulfide by the SRB and this process is coupled to the oxidation of an electron donor. In this study, the SRB populations used to inoculate the various seawater mixtures were growing on lactate as carbon source and electron donor for sulfate reduction. As the test solutions were maintained under stagnant conditions, microorganisms could have consumed a significant proportion of nutrients during the first days of exposure so that nutrients became depleted with exposure time. Primary inhibition of SRB growth could be due to exhaustion of organic and inorganic substrates in the seawater with time. Inhibition could have also resulted from the toxicity of accumulated sulfide in the solution. Sulfide inhibition towards different trophic groups is widely reported in literature [26, 27]. Oxygen and pH of test solutions also could have limited SRB growth. SRB are recognized as anaerobic microorganisms so the presence of oxygen in the system could also have restricted their growth in the different seawater mixtures. However, it has been reported that SRB possess various self-protection enzymes that facilitate survival during periods of oxygen exposure [28]. This is supported by the fact that oxygen levels in the raw seawater remained very low compared to the mixtures of seawater treatments. Since seawater pH was lowered by additions of chemical treatments, pH could have played a major role on the inhibition of SRB growth even if the biocide efficiency was affected by oxygen in the system. For 100% treated seawater, pH was lowered to values below 5 which are expected to be very aggressive for SRB given that SRB were growing in culture medium at pH 7.0 before mixing with chemical treatments.

It is important to underline that this enumeration technique is restricted to cultivable microorganisms and usually underestimates the real bacterial numbers in the system [29]. The application of molecular tools has shown that bacteria growing in culture media often represent a minor part of the microbial community: around 99% of microorganisms existing in nature are unable to be cultured by selective enrichment cultures, and they will therefore, be excluded when enumerated with growth media [30]. This can explain why low numbers of sessile SRB were detected by the serial dilution method whereas abundant bacterial cells were observed on DAPI stained coupons.

3.3 Surface analysis by 3D optical microscopy

3D optical surface imaging and pit profile measurements were conducted to confirm pitting corrosion and to measure pit depths. Pitting corrosion was confirmed on 316L after CPP tests at the different levels of treated seawater. Figure 7 shows pits observed on 316L in 100% treated seawater after completion of the CPP test. Pits depths were very similar for 316L regardless of the concentration of seawater treatment. Table 4 summarizes the surface analysis of 316L exposed to the different levels of treated seawater at the different immersion conditions. In immersion test A, pitting was observed on coupons exposed to raw untreated seawater as well as to all levels of treated seawater from the 7th day to the 28th day of exposure with only few exceptions. Figure 8 shows an optical image and pit profile measurement of a pit on 316L exposed to raw seawater in immersion test A. Overall, pit density tended to decrease with exposure time for 316L exposed to all seawater levels. In most cases, pit density did not exceed 1 pits/cm2. For immersion test A, pit depths ranged from 3 to 22 µm, maximum pit density was 2.98 pits/cm2 and maximum pit depth found was ∼21 µm. Most of the pits on 316L exposed to immersion test A had pit depths in the range from 0 to 10 µm. The frequency of the deepest pits decreased with time and was the highest at 7 days of exposure. This finding could indicate that exposure time tended to favor repassivation rather than propagation processes. This is also supported by the fact that the deepest pits observed were wide and shallow, with pit widths typically more than 10 times pit depths. Overall, pitting was randomly observed on 316L exposed to the various levels of seawater for immersion test A and it is not possible to establish any association between seawater treatment concentration and pit depth/density.

image

Figure 7. 3D optical surface image of typical pits found on 316L after cyclic polarization tests

Download figure to PowerPoint

Table 4. Summary of surface analysis results – immersion tests
Immersion testExposure time (days)Seawater treatment (concentration)Pit density (pits/cm2)Average pit depth (µm)Maximum pit depth (µm)
A7Raw (control)2.9811.6121.39
  20%0.856.7814.73
  40%0.674.425.45
  60%0.856.7111.65
  80%0.854.866.15
  100%2.547.8117.46
 10Raw (control)1.356.8112.64
  20%0.358.1612.36
  40%0.344.405.23
  60%0.337.399.43
  80%0.1717.6217.62
  100%1.748.1616.04
 14Raw (Control)0.533.213.56
  20%0.686.1614.84
  40%0.174.344.34
  60%0.000.000.00
  80%0.684.355.41
  100%0.525.687.09
 21Raw (control)0.000.000.00
  20%0.5214.3121.45
  40%0.178.438.43
  60%0.1712.5012.50
  80%0.344.916.97
  100%0.1713.9813.98
 28Raw (control)1.046.008.36
  20%0.1714.9614.96
  40%0.174.564.56
  60%0.857.6113.29
  80%1.027.269.89
  100%1.206.349.22
B7100%0.000.000.00
 14 0.000.000.00
 28 0.663.024.80
C7100%0.334.236.33
 14 0.244.646.55
 28 0.242.862.98
D780%0.000.000.00
 14 0.000.000.00
 28 0.000.000.00
image

Figure 8. Optical surface image (left) and pit depth measurement (right) of a typical pit found on 316L exposed to raw seawater at 20 °C for 7 days in immersion test A

Download figure to PowerPoint

For immersion test B, pitting did not take place during the first 14 days of exposure, but a few shallow pits were evident on the 28 days exposure coupons. Since for this test, high numbers of SRB were inoculated, these results could indicate that the biocide activity was not sufficient to supress all microorganisms in the seawater. It is likely that the remaining bacteria could have induced some localized events on the steel surface with time. For immersion test C, pits were observed on the 7, 14, and 28 days exposure coupons. Again, surface analysis of pits indicated that pit widths were about 10 times the pit depths. These wide shallow pits are more likely to undergo repassivation rather than propagation. In addition, the overall decrease in pit density and pit depths in all immersion tests with time may indicate that some repassivation processes could have taken place. Pitting was not detected on immersion test D regardless of the exposure time.

Results from immersion tests and DO measurements indicated that DO plays a critical role in triggering the initiation of pitting on 316L. DO measurements, surface and microbiological analyses indicated that in the presence of oxygen, chemical treatments become ineffective hence facilitating pitting initiation and microbial growth on steel surfaces. As mentioned earlier, the numbers of SRB in the raw natural seawater used for immersion tests as estimated by the standard serial method was 103 cell/mL. Results indicate that, in the absence of oxygen, THPS was efficient in killing SRB in the seawater with ingress of 20% untreated raw seawater. Results also showed that THPS at the recommended dosage for hydrotesting water (100% treatment) is efficient against SRB even at SRB concentrations as high as 108 cell/mL.

The use of preservation treatments for hydrostatic test waters is a crucial practice to provide protection of subsea pipelines against corrosion. In the event of seawater ingress into 316L lined pipes during tie-in operations, there is a high risk of localized corrosion particularly from the oxygen and microbial life coming into the system and mixing with the chemical treatments thus increasing their concentration in the treated seawater.

The ingress of oxygen into the system not only can accelerate pitting initiation by creating oxygen concentration cells but it can also interfere in the efficiency of chemical treatments hence restraining oxygen scavenging and biocidal activity in the system. The entry of seawater into the system can also introduce aggressive species that could affect the efficiency of any residual treatment such as deposits and nutrients that may support bacterial growth and biofilm formation on the steel surface hence increasing the likelihood of MIC. Results from this study indicate that 316L is susceptible to localized corrosion in seawater at 20 °C if oxygen is not effectively removed from the seawater. If oxygen is restricted to minimal levels in the system, 316L is protected against pitting even if the recommended dosages of chemical treatments are mixed with raw seawater in the proportion treated/raw seawater 80:20. However, this result must be carefully interpreted as pitting could be easily initiated under these conditions if the physicochemical and biological conditions in the system are slightly changed during exposure, e.g. ingress of high loading of bacteria and nutrient sources for biofilm growth, increase in exposure temperatures and oxygen ingress into the system, among others.

In the event of long exposure of lined pipes to raw seawater during tie-in operations, the removal of oxygen is crucial to ensure appropriate protection and subsequent preservation of the flowline against pitting. Particular consideration should be given to the potential ingress of deposits or sand into the system, which could increase the likelihood of MIC. Filtration is essential to remove such particulates.

4 Conclusions

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

Cyclic potentiodynamic tests indicated that in the absence of oxygen, an anodic driving potential is required to trigger pitting corrosion on crevice-free 316L in seawater at 20 °C regardless of the concentration of the chemical treatments.

Immersion tests indicated that, in the presence of high concentrations of oxygen, the efficiency of the THPS, regardless of the concentration, is considerably reduced. This was demonstrated by the large bacterial colonization on the steel surface observed at all the levels of chemical treatments. Under these conditions, oxygen scavenging no longer takes place, most likely because of inactivation by THPS, bacteria growth is favored and pitting can easily initiate after as little as 7 days of exposure.

Immersion tests showed that if oxygen is restricted to levels below 20 ppb in the system, 316L is protected against pitting at the recommended dosages of chemical treatment for up to 14 days exposure even at SRB concentrations as high as 108 cell/mL. At concentrations of DO below 60 ppb, which were achieved by mixtures of 80% treated seawater:20% raw seawater, and provided the ingress of oxygen is restricted during exposure, 316L is also protected against pitting corrosion for up to 28 days. A concentration of 2 ppm of DO was sufficient to induce pitting corrosion on 316L from the first 7 days of exposure even at the recommended dosage of chemical treatments.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References

The authors would like to thank Chevron Energy Technology Pty Ltd for the financial support and permission to publish this work. We also acknowledge the support of the Australian and Western Australian Governments and the North West Shelf Joint Venture Partners, as well as the Western Australian Energy Research Alliance (WA:ERA).

References

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Materials and methods
  5. 3 Results and discussion
  6. 4 Conclusions
  7. Acknowledgements
  8. References