64 MATERIALS PERFORMANCE May 2013 NACE International, Vol. 52, No. 5

Investigation of Microbiologically

Influenced Corrosion in

Pipeline Hydrotesting Using Seawater

Kaili Zhao and Tingyue gu, Ohio University, Athens, Ohioivan CruZ, Saudi Aramco, Dhahran, Saudi Arabia

ardjan KopliKu, BP America, Inc., Houston, Texas

Hydrotesting is a common prac-tice to assess pipeline integrity before service. Different from pneumatic testing that is used

only for leak testing, hydrotesting is ap-plied to test for both leaks and strength. During hydrotesting, a pipeline is filled with a liquid and pressurized to a pressure (usually 10%) greater than the anticipated future operating pressure.

BackgroundIn general, hydrotesting itself lasts only

eight to 10 h. In the oil and gas industry, however, it is often the case that water is left in the system afterward for many months before the system is actually com-missioned. During this holding time or when the pipeline is first exposed to an aqueous environment like wet lay-up, corrosion due to microbiologically influ-enced corrosion (MIC) can commence.1

When the system makes contact with the ground2 or is even exposed to air,3 there are further possibilities for micro-bial contamination. Reuse of water also increases chances for MIC. Improper hydrotesting practices can cause MIC pitting attacks and also black powder problems.4 MIC pitting during hydrotest itself may not be a big problem because of the limited hydrotest time frame. The biofilms left behind during hydrotest, however, may present a serious threat once the pipelines become operational, because fluids transported in pipelines may contain sufficient nutrients for bio-films to flourish and a pipeline is often expected to be operational for several decades.

Seawater in PipelinesSeawater is routinely used in the hy-

drotesting of subsea pipelines. Occasion-ally, other water sources may be used and they mainly come from aquifer water and/or produced water. Any water source for hydrotesting can contain mi-

This work investigated the microbiologically

influenced corrosion (MIC) threat in pipeline

hydrotesting using offshore seawater samples for

coupon tests in anaerobic vials. Longer-term sulfate-

reducing bacteria pitting was predicted using a

MIC prediction software program calibrated with

short-term pitting data from the tests.

NACE International, Vol. 52, No. 5 May 2013 MATERIALS PERFORMANCE 65

M A T E R I A L S S E L E C T I O N & D E S I G N

croorganisms. Natural seawater contains viruses, prokaryotes, protists (mainly flagellates), and algae.5 Water used in hydrotesting is sometimes treated with biocides. Even treated water can be a source of sulfate-reducing bacteria (SRB) inoculum.6 Two other methods to treat the hydrotest water are adjusting pH and using water sources without sulfate.7 The pH adjustment (within a basic range), however, could increase the possibility of mineral scale formation on the pipe sur-face, and using a large amount of water without sulfate is usually costly and incon-venient when hydrotesting takes place offshore. Furthermore, the method of pipeline laying or water filling makes water treatment very difficult, if possible at all.

Sulfate-Reducing Bacteria Metabolism

It has been known that some SRB are able to utilize hydrocarbons or even live on carbon dioxide-hydrogen (CO2-H2) autotrophically,8 which means they can live without organic carbon intakes. Ross-moore9 found that a variety of bacteria have the capability to reduce in size, de-creasing energy consumption during starvation and residing in smaller pores. These bacteria can then wait to thrive when the appropriate environmental

conditions are met. This unique feature of bacteria makes predicting and prevent-ing the MIC in hydrotesting difficult. Steel corrosion in seawater sometimes has been misdiagnosed as attack induced only by conventional chloride corrosion. Bo-renstein1 found that microorganisms contained in a stagnant chloride-bearing medium can cause steel failure much faster than in conventional chloride crev-ice corrosion alone. This increased cor-rosion rate may come from sulfate and other nutrients in the seawater, which cause souring and pipeline corrosion due to SRB activities.

Use of Oxygen ScavengersIn the field, oxygen scavengers are

usually added to the hydrotesting water to prevent oxygen-caused corrosion. This provides an anaerobic environment for anaerobic bacteria such as SRB. MIC occurs when several favorable factors are present simultaneously, such as suitable water chemistry, temperature, nutrients (organic and inorganic), microorganisms, and pressure. The majority of SRB can thrive at pH ranges from 5 to 9, and ex-cept for thermophiles, are unable to thrive at temperatures >45 °C. Avail-ability of a carbon source is usually con-sidered to be the most important factor for SRB growth; SRB growth will be se-

verely restricted if utilizable carbon in organic nutrients in the form of volatile fatty acids such as formate, acetate, and propionate, is <20 ppm.10 Pots, et al.10 also indicated that SRB growth would be the most prominent if the ratio of carbon to utilizable nitrogen was 10:1. Synergis-tic microorganisms can enrich the nutri-ents (such as organic carbons) in the local environment and thus promote SRB growth and accelerate the MIC process even though the initial environmental conditions are not suitable for SRB growth. Fermentative acid-producing bacteria (APB) should be considered in MIC forensics, especially in zero-sulfate and low-sulfate environments.

Laboratory TestingPerforming MIC tests in a laboratory

setting for hydrotest has always been a challenge. Pipeline fluids (especially those in subsea pipelines) can be at very high local pressures. Barophilic SRB are adapted to this kind of pressure.8 In a laboratory, however, it is difficult and cost prohibitive to perform many tests in high-pressure reservoir simulators. It is possible that laboratory tests at one at-mosphere may be able to simulate SRB growth at a high pressure because it has been reported that barophilic SRB iso-lated from a high-pressure oil reservoir

TABLE 1

Major element comparison between typical natural seawater and untreated GoM seawaterCa2+ (ppm) Na+ (ppm) Cl– (ppm) F– (ppm) SO4

2– (ppm) K+ (ppm) TOC (ppm)

Typical natural seawater

400 to 412 10,500 to 10,770

18,800 to 19,300

1.2 to 1.3 2,655 to 2,715 380 to 390 <1 to 2

GoM seawater

421 10,800 19,700 1.41 2,655 398 Not detected <1

TABLE 2

Major element comparison between typical natural seawater and Qurrayah seawater(A)

Na+ (ppm) SO42– (ppm) TOC (ppm)

Typical natural seawater 10,500 to 10,770 2,655 to 2,715 <1 to 2

Qurrayah seawater (seawater) 16,580 4,330 498(A)The assay for Na+ and SO4

2– in Qurrayah seawater was done by ENC Labs (Albuquerque, NM), and TOC was assayed by San Antonio Testing Laboratory, Inc. (San Antonio, Texas).

66 MATERIALS PERFORMANCE May 2013 NACE International, Vol. 52, No. 5

M A T E R I A L S S E L E C T I O N & D E S I G N Investigation of Microbiologically Influenced Corrosion in Pipeline Hydrotesting Using Seawater

grew well at one atmosphere and 30,000 kPa in a laboratory.8

Experimental MethodsAnaerobic 125-mL vials filled with

100-mL liquid were used in the tests. A glovebox deoxygenated with N2 gas pro-vided an anaerobic environment. X65 carbon steel coupons were used. These coupons had typical dimensions of 47.6 by 10.9 by 1.6 mm. Prior to use, the coupon surfaces were polished succes-sively with 200 and 400 grit SiC abrasive papers, rinsed with alcohol, and then sonicated in a beaker filled with alcohol. The ratio of coupon surface to liquid volume was close to that in 0.30-m (12-in) inside diameter (ID) pipes.

All liquids in the tests were de-oxygenated using N2 sparging for at least 30 min before use to reflect oxygen scav-enger use in the field. Planktonic SRB bacterial count was determined by manual counting under an optical micro-scope at 400X using a hemacytometer. Only motile SRB were counted. If needed, a Rodine hydrochloric acid (HCl) solution was applied to remove any films on the coupon surfaces. Scanning electron microscopy (SEM) and energy dispersive spectrometry (EDS) were em-

†Trade name.

Planktonic SRB growth in enriched artificial seawater and full nutrient medium at different temperatures.

FIGURE 1

ployed to perform surface analyses. A CHEMets† test kit was used to test the oxygen concentration in the experimental vials.

Desulfovibrio alaskensis (ATCC 14563) was used in this work as a laboratory strain of SRB. Some experimental results as indicated were obtained by enriching artificial seawater and natural seawater samples with 1 g/L yeast extract, 3.5 g/L sodium lactate, and 200 ppm Fe2+.

For biofilm observations under SEM, unless mentioned specifically, coupons were pretreated according to the follow-ing procedures: coupons were removed from vials and were immediately treated with 4% w/w glutaraldehyde for around 1 h (to immobilize the biofilm), and then were dehydrated with 30% (v/v), 50, 75, and 100% alcohol in sequence. Before observing the biofilm, the coupons were first treated using a Bal-Tec CPD 030† critical point dryer and then coated with a gold film.

Results and Discussion

Gulf of Mexico SeawaterTable 1 shows that the Gulf of Mexico

(GoM) seawater had a similar chemical composition to that of typical natural seawater. The total organic carbon

(TOC) in the first GoM sample was <1 ppm compared to <1 to 2 ppm TOC for typical seawater while the TOC of a second GoM sample was 4.6 ppm. The GoM seawater sample analyzed using polymerase chain reaction (PCR) was actually very clean. It had a total bacterial concentration of only 13.3 cells/mL, and its SRB cell count was below the detection limit of 1 to 3 SRB cells/L. The sample was taken from an offshore platform.

When Hardy11 measured seven sea-water samples from two similar locations of the North Sea, he obtained SRB num-bers from 0 to 90 cells/mL, the average being 22 SRB/mL. Lee, et al.,12 using the most probable number enumeration method, detected ~10 and 100 SRB/mL in Persian Gulf and Florida Key West seawaters, respectively. These two water samples came from 1.2 to 1.5 m deep and near-shore (within 100 m) locations that could be contaminated by sewage, agri-cultural run-off, or other waste streams. Table 2 shows Na+, SO4

2–, and TOC in a comparison between typical natural seawater and Qurrayah seawater in Saudi Arabia. It is clear that Na+ and SO4

2– concentrations in Qurrayah seawater are ~1.6 times higher than in typical seawa-ter, and the TOC concentration, which is very important for microbial growth, can be 500 times higher.

Temperature EffectFigure 1 shows how temperature af-

fected planktonic SRB growth, where 37 °C is the optimum growth tempera-ture for the lab strain SRB. Compared to the full nutrient medium (ATCC 1250 modified Baar’s medium), the enriched artificial seawater with limited nutrients is an acceptable environment for SRB growth, especially at 37 °C, and those added chemicals provided adequate nu-trients for SRB growth. In general, me-sophilic SRB grow well at 37 °C. Ther-mophilic SRB prefer even a higher

NACE International, Vol. 52, No. 5 May 2013 MATERIALS PERFORMANCE 67

M A T E R I A L S S E L E C T I O N & D E S I G N

temperature, but 37 °C is likely suffi-ciently high for pipelines in a shallow seabed in a hot climate. This means in-creased SRB growth with increasing temperature is generally expected in practical situations. It should be pointed out that planktonic cell counts may be used to help indicate the sessile cell health in laboratory tests, but the planktonic cell counts should not be used to correlate with sessile cell counts.

Microbial GrowthNo microbial growth was detected

after one month and six months in vials containing untreated GoM seawater. After cleaning with Clark’s solution, SEM images showed roughness on the entire surface of a coupon with one-month ex-posure to the seawater in a vial at 37 °C, and also of a coupon with six-month ex-posure at 25 °C. Due to lack of microbial activities and a hydrogen sulfide (H2S) smell at the end of the test, the roughness was likely not caused by SRB. Similar roughness was also observed in tests using heat sterilized GoM seawater.

Quarrayah SeawaterThe Qurrayah seawater from the

Persian Gulf is much saltier than the GoM seawater, as seen in Table 2. In-house quantitative PCR analysis did not detect SRB in the Qurrayah seawater. Figures 2(a1) and (a2) show that a mineral layer covered the coupon surface after a three-month exposure at 37 °C. Figures 2(b1) and (b2) show scattered pits after the coupon surfaces were cleaned. They were likely due to factors such as a trace amount of oxygen leaking through the capped rubber septum rather than micro-bial activities. Oxygen leakage was not a problem, however, in the tests for vials that were a few weeks long. Some three-month vials were discarded in tests be-cause of visible oxygen rust. A wax seal around the aluminum cap was subse-

SEM images (a) before acid cleaning, (b) after acid cleaning, and (c) the EDS analysis before acid cleaning for a coupon after three-month exposure to the untreated Qurrayah seawater at 37 °C.

FIGURE 2

quently used but it did not completely eliminate oxygen egress. An anaerobic chamber would be the last solution other than using an oxygen scavenger to pre-vent oxygen egress in long-term tests.

The EDS analysis of the surface in Figure 2(c) indicates the absence of the sulfur element, which means that SRB activity was likely absent.

Due to the lack of native viable mi-crobes and the lack of nutrients, no MIC pitting was observed in untreated seawa-ter samples. To simulate a contaminated hydrotest fluid and to speed up laboratory testing, worst-case scenario tests were carried out by enriching seawater samples and spiking them with the laboratory

SRB strain. Figure 3(a) shows the SEM image of the biofilm on a one-week old coupon. Kidney bean-shaped SRB cells are clearly visible. Pits characteristic of MIC attack were revealed after acid cleaning of the coupon surface, as seen in Figures 3(b1) and (b2). An EDS analysis shown in Figure 3(c) indicates the pres-ence of iron sulfide (FeS).

Kinetics-Based Mechanistic Model

Recently, Gu, et al.13 introduced an electrochemical kinetics-based mechanis-tic model for MIC using a new biocata-lytic cathodic sulfate reduction (BCSR) theory. It assumes that a corrosive SRB

68 MATERIALS PERFORMANCE May 2013 NACE International, Vol. 52, No. 5

biofilm is present on an iron surface, caus-ing the following reactions to go forward due to biocatalysis:

Anodic: 4Fe → 4Fe2+ + 8e– (iron oxidation) (1)

Cathodic: SO42– + 9H+ + 8e– →

HS– + 4H2O (BCSR) (2)

By using charge transfer and mass transfer theories and electrochemical ki-netics, a mechanistic model was devel-oped and solved numerically.14 The software based on the model is known as MICORP. It incorporates BCSR, proton reduction, and organic acid reduction to account for low pH at a pit bottom due to organic acids. Figure 4 shows the model prediction and experimental data obtained in this work. The model was calibrated with a single pit depth data to predict long-term pitting.

Tetrakis hydroxymethyl phosphonium sulfate (THPS) is a biodegradable biocide that is most often proposed for hydrotest fluid treatment. A minimum dosage is needed to prevent biofilm establishment. Tests were carried out in anaerobic vials to evaluate the THPS degradation profiles in artificial seawater, GoM seawater, and Qurrayah seawater. A mechanistic model of THPS degradation under alka-line pH was obtained and reported elsewhere.15

ConclusionsThis work provided a framework for

laboratory testing of MIC in hydrotest-ing. Arguments were made for laboratory testing at one atmosphere instead of a high pressure expected in a subsea pipe-line during hydrotesting. Clean offshore seawater samples from the Gulf of Mex-ico and the Persian Gulf were found to lack native viable microbial activities. The seawater samples were enriched and spiked with a laboratory strain SRB to simulate contaminated seawater used in hydrotest. Biofilms and MIC pits were observed in the accelerated tests. A soft-ware package based on the BCSR theory was used to predict the pit growth.

M A T E R I A L S S E L E C T I O N & D E S I G N Investigation of Microbiologically Influenced Corrosion in Pipeline Hydrotesting Using Seawater

SEM images (a) before acid cleaning, (b) after acid cleaning, and (c) EDS analysis before acid cleaning for a coupon after one-week exposure to the enriched Qurrayah seawater spiked with SRB at 37 °C.

BCSR model prediction of two-week and one-month pit depths using biofilm aggressiveness calibrated from the one-week pit depth (18 μm).

FIGURE 3

FIGURE 4

NACE International, Vol. 52, No. 5 May 2013 MATERIALS PERFORMANCE 69

M A T E R I A L S S E L E C T I O N & D E S I G N

AcknowledgmentsThe authors would like to thank Saudi

Aramco and BP America for their finan-cial support of this research and for their permission to present the results.

References1 W.S. Borenstein, P.B. Lindsay, “MIC

Failure of 304L Stainless Steel Piping Left Stagnant after Hydrotesting,” MP 41, 6 (2002): pp. 70-73.

2 H.A. Videla, Manual of Biocorrosion (Boca Raton, FL: Lewis Publishers, 1996).

3 A. Parra, J.J. Capio, I. Martinez, “Microbial Corrosion of Metals Exposed to Air in Tropical Marine Environment,” MP 35, 10 (1996).

4 R.M. Baldwin, “Black Powder in the Gas Industry—Sources, Characteristics and Treatment,” Southwest Research Insti-tute, Mechanical and Fluids Engineering Div., May 1998.

5 M.G. Weinbauer, D.F. Wenderoth, “Microbial Diversity and Ecosystem Functions—the Unmined Riches,” Electronic J. of Biotechnology, April 15, 2002.

6 P.F. Sanders, “Overview of Souring, Corrosion and Plugging Due to Reservoir Organisms,” U.K. Corrosion 98, Paper 15 (Northampton, U.K.: Institute of Corrosion, 1998).

7 R. Prasad, “Chemical Treatment for Hydrostatic Test,” U.S. Patent 6815208 B2, 2004.

8 J.T. Rosnes, T. Torsvik, T. Lien, “Spore-Forming Thermophilic Sulfate-Reducing Bacteria Isolated from North Sea Oil Field Waters,” Applied and Environmental Microbiology 57 (1991): pp. 2302-2307.

9 H.W. Rossmoore, Handbook of Biocide and Preservation Use (Glasgow, U.K.: Blackie Academic & Professional, Chapman & Hall, 1995).

10 B.F.M. Pots, R.C. John, I.J. Rippon, “Improvements on DeWaard-Milliams Corrosion Prediction and Applications to Corrosion Management,” CORRO-SION 2002, paper no. 02235 (Houston, TX: NACE, 2002).

11 J.A. Hardy, W.A. Hamilton, “The Oxygen Tolerance of Sulfate-Reducing Bacteria Isolated from North Sea Waters,” Current Microbiology 6, 259-262, (1981).

12 J.S. Lee, R.I. Ray, “Comparison of Key West and Persian Gulf Seawaters,” CORROSION 2007, paper no. 07518 (Houston, TX: NACE, 2007).

13 T. Gu, K.L. Zhao, S. Nesic, “A Practical Mechanistic Model for MIC Based on a Biocatalytic Cathodic Sulfate Reduction Theory,” CORROSION 2009, paper no. 09390 (Houston, TX: NACE, 2009).

14 K.L. Zhao, “Investigation of Microbio-logically Influenced Corrosion (MIC) and Biocide Treatment in Anaerobic Salt Water and Development of a Mechanis-tic MIC Model” (Ph.D. diss., Ohio Uni-versity, 2008).

15 K.L. Zhao, J. Wen, T. Gu, A. Kopliku, I. Cruz, “Mechanistic Modeling of Anaerobic THPS Degradation Under Alkaline Condition in the Presence of Mild Steel,” MP 48, 8 (2009): pp. 62-66.

This article is based on CORROSION 2010 paper no. 10406, presented in San Antonio, Texas.

KAILI ZHAO is a principal scientist at a coating company in Nanjing, China. He graduated with a Ph.D. in chemical engineering from Ohio University in 2008. He also holds an M.S. degree in chemical engineering from Tianjin University, China.

TINGYUE GU is a professor at Ohio University, Dept. of Chemical and Biomolecular Engineering, Athens, OH 45701, e-mail: gu@ohio.edu. He obtained his Ph.D. in chemical engineering from Purdue University in 1990. He worked for Miller Brewing Co. for one year before joining Ohio University. He is affiliated with the Ohio University Institute for Corrosion and Multiphase Technology. He has published more than 65 journal papers and book chapters as well as two books. He is a nine-year member of NACE.

IVAN CRUZ served as senior engineering consultant to Saudi Aramco, PO Box 6891, Dhahran, 31311, Saudi Arabia. Prior to joining Saudi Aramco in 1992, he was senior engineering advisor at the Mobil Exploration & Producing Technical Center in Dallas, Texas. He has extensive corrosion engineering experience in the oil and gas industry. He has a B.S. degree in chemical engineering and is a certified NACE Corrosion Specialist and Materials Selection/Design Specialist and a NACE instructor. He is a recipient of the 2007 NACE Distinguished Service Award. A member of NACE since 1977, he retired in November 2010.

ARDJAN KOPLIKU is the Corrosion Management& Fabric Maintenance team leader for the Gulf of Mexico Deepwater Operations in BP, 200 Westlake Park Blvd., Houston, TX 77079. He has longtime experience in corrosion management, material selection, and integrity management. Before joining BP in 2004, he served as a materials and corrosion engineer at ENI (formerly Agip) in Italy. He has been a NACE member since 2000.