SPE SPE-173709-MS Investigation of Microorganisms in a West Texas Oilfield Using Growth and Genetic Testing

SPE SPE-173709-MS

Investigation of Microorganisms in a West Texas Oilfield Using Growth and Genetic Testing John Kilbane, SPE, Intertek, 6700 Portwest Drive, Houston TX 77024, [email protected] 713-479-8522 Jonathan Wylde, SPE, Clariant Oil Services, 2750 Technology Forest Blvd., The Woodlands TX 77381, [email protected] 832-663-3925 Andy Williamson, Occidental Petroleum Corp., 5 Greenway Plaza, Suite 110, Houston TX 77046, [email protected] 713-552-8554

Copyright 2015, Society of Petroleum Engineers This paper was prepared for presentation at the SPE International Symposium on Oilfield Chemistry held in The Woodlands, Texas, USA, 13–15 April 2015. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract

Increasing corrosion issues at a sour west Texas oilfield in the Permian Basin suggested that microbiologically

influenced corrosion (MIC) could be the cause. However, multiple water and biofilm samples failed to show

significant concentrations of microorganisms using traditional microbial growth media for the cultivation of APB,

GHB and SRB. Microbial growth tests indicated the highest concentration of microorganisms as 103 cells/ml,

while the results of genetic testing using qPCR indicated that microbial concentrations of up to 1000-fold higher

were actually present. Additionally, biochemical testing for ATP (adenosine triphosphate) indicated that

microorganisms were present in water samples at higher concentrations than were detected in growth tests. It

was speculated that the type of microorganisms present in the samples from this saline (6 to 10%) thermophilic

(600C) field may not grow well in the standard microbial growth media. Additionally, isotopic analysis of sulfur in

the hydrogen sulfide gas and in sulfate from formation water are consistent with microbiological sulfate

reduction being responsible for reservoir souring at this location.

To characterize the microbial community composition of these samples more thorough DNA sequencing was

performed on selected samples. The results of DNA sequence analyses show that the percentage of unclassified

DNA sequences (unidentifiable at the genus level because no similar DNA sequences are present in the global

database) ranged from 1.8% to 25.8%. Of the bacteria that could be identified, a high percentage of halophilic

(salt loving) bacteria, and of SRB were observed in all samples.

These results demonstrate that genetic testing can provide data about microorganisms in some oilfield samples,

even when standard microbial growth tests fail to indicate the presence of microorganisms. Moreover, standard

microbial growth recipes are not well suited for the growth of microbes from all locations and there is a need for

new formulations of microbial growth media for use in some locations, such as the Permian Basin. Using

standard microbial growth tests in this case would have seriously under-estimated the integrity risk caused by

the presence of high numbers of SRB that were demonstrated to be present by genetic analyses. Armed with the

knowledge that significant concentrations of microbes were indeed present, but had different nutritional

2 SPE SPE-173709-MS

requirements than those provided in standard microbial growth media, new formulations of microbial growth

media were developed by Intertek that are tailored to the requirements of these Permian Basin samples.

Introduction

To maintain reservoir pressure in this Permian Basin field the produced water is treated minimally to recover oil,

and is then reinjected into the formation. Some fresh water is also added to the produced water to make up for

the volume of oil removed (about 30% of total production) prior to the reinjection of the water. Therefore, the

same water from this formation is constantly pumped from and reinjected into this formation during the life of

this oilfield. This Permian Basin oilfield has been in production for decades and has experienced increasing

concentrations of hydrogen sulfide. Besides being a health and safety concern, hydrogen sulfide can contribute

to corrosion decreasing the productivity of the oilfield (Tabari 2011). An increased corrosion failure rate of the

rod pumping well pipes and equipment in recent years (see Figure 1) has caused an investigation to determine

the root cause of the hydrogen sulfide, and to see if anything can be done to prevent or control future hydrogen

sulfide production.

Figure 1. Through-wall corrosion of a pipe from a Permian Basin well that had only been in service for 3 months.

Presentation of Data and Results

SPE SPE-173709-MS 3

A likely possible cause of reservoir souring is microbiological conversion of sulfate to hydrogen sulfide

(Schwermer 2008, Stewart 1998, Tabari 2011). Microbiological testing has been performed multiple times over

the years by multiple service providers, however; previous microbiological testing has been largely restricted to

the use of standard microbial growth media, and has consistently yielded data that the microbial community

concentration at this location was low to undetectable. Previous efforts to use different formulations of

microbial growth media yielded similar results indicating microbes at low to undetectable concentrations. The

temperature of the reservoir is about 140oF (60oC), which is within the range that microorganisms can grow. So

it is possible that microbial conversion of sulfate to hydrogen sulfide is responsible for reservoir souring at this

location, but microbial growth tests fail to detect a significant concentration of microorganisms. Accordingly, a

more in depth investigation using other analytical techniques was employed. More specifically, sulfur isotopic

ratio data, and genetic testing data were employed to determine if microbiological processes are responsible for

reservoir souring at this location.

There are several isotopes of sulfur with 32S and 34S being the most abundant, and 32S being the primary sulfur

isotope. Chemical processes generally do not discriminate between different isotopes whereas biochemical

processes often will preferentially metabolize one isotope versus another resulting in what is termed isotopic

fractionation (Habicht 1997). In microbiological sulfate reduction those sulfate molecules that contain 32S are

metabolized preferentially versus sulfate molecules containing the 34S isotope. This results in the formation of

hydrogen sulfide molecules that have lower amounts of 34S than was present in the sulfate. The amount of

reduction of the 34S content of hydrogen sulfide gas, as compared with the sulfate source, that is caused by

microbiological cultures ranges from no detectable change to -40 ‰(Detmers 2001, Habicht 1997, Stam 2010).

However, for mixed cultures of Sulfate Reducing Bacteria typically found in oil industry samples the amount of

isotopic fractionation (reduced amount of 34S in the H2S gas) is about -20 ‰(Balliger 2001). The concentration of

sulfur isotopes is generally stated as compared with a standard and in the case of sulfate the reference standard

is VCDT (Vienna Canyon Diablo Troilite) and the 34S content of sulfate in produced water from this Permian Basin

site was 19.5 ‰, while the concentration in the H2S averaged -1‰, or a total reduction of -20.5 ‰. This

chemical analysis of the sulfur isotopes of sulfate versus H2S provides an independent source of information that

is consistent with microbiological production of H2S at this Permian Basin location. Therefore, it is puzzling that

microbial growth tests didn’t indicate the presence of significant concentrations of microorganisms so another

source of information concerning the microbiology of the system is needed.

4 SPE SPE-173709-MS

Table 1 shows that the 34S concentration in the hydrogen sulfide is lower than the concentration in the sulfate by

-20.5 ‰.

In recent times the study of microbiology generally has benefitted from the use of genetic techniques that

overcome some of the limitations of traditional microbial growth tests and have led to the conclusion that many

microbial species present in environmental samples can’t be grown in traditional microbial growth tests

(Haveman 2002). The myriad species of microorganisms in nature can occupy a wide range of environments

such that different microbes use different nutrients (biochemical pathways) for growth. Therefore, the nutrients

present in traditional microbial growth media may not be suitable for all types of microbial species. Moreover,

even if the correct combination of nutrients are present the growth conditions such as temperature, salinity,

and redox potential also need to be matched to the needs of each type of microbial community (Fichter 2015).

This constant reinjection of water into the Permian Basin formation, combined with the presence of high sulfate

concentrations, and a reservoir temperature compatible with microbial growth make it highly likely that

microorganisms are growing in the subsurface. Perhaps the composition of the produced water at the Permian

Basin site can provide some insight into this situation and Table 2 lists the concentrations of the most abundant

components of water obtained from a wellhead and from a water collection/treatment location after petroleum

had been removed from the water and the water was prepared for re-injection into the formation.

Values in Table 2 are in units of ppm.

Table 1. 34

S Isotopic Analysis of Sulfate and Sulfide

Sulfate H2S Differential

18.3 3.5

20.7 1.3

-4.5

-4.2

Avg. 19.5 -1 -20.5

34S ‰ relative to VCDT (Vienna

Canyon Diablo Troilite)

Table 2. Water Chemistry Summary

Water

collection &

treatment

Wellhead

water seawater

Ca 3,434 3,912 410

Mg 818 1,087 1,300

K 385 557 370

Na 20,957 29,262 11,200

acetate 3.4 49.6 0

Cl 46,388 57,963 19,500

SO4 2,450 2,751 2660

pH 7.42 7.20 7.9

TDS 78,635 105,866 42,000

SPE SPE-173709-MS 5

It can be seen from Table 2 that the water at this location contains significant concentrations of acetate, or

acetic acid, that is a byproduct of microbial biodegradation of petroleum hydrocarbons, and is a further

indication that microorganisms are present in this oilfield. The data in Table 2 also illustrates that the water at

this location is significantly saltier than seawater (shown in Table 2 for comparison). The wellhead water is

saltier than the water collection/treatment sample because some fresh water is added to the produced water to

increase the volume of fluid used in re-injection for the maintenance of pressure in the reservoir. The high level

of salinity can definitely influence the ability of microorganisms to grow, but there are salt-tolerant and salt-

loving microorganisms that are known to be capable of growing in water with even higher salinity than is shown

in Table 2 (Stam 2010). So, this Permian Basin site has a higher salinity, and a higher temperature, than many of

the samples from other oilfield locations and this can be a factor that might help to understand why the

traditional microbial growth tests failed to detect a microbial population at this location.

Commercially available microbial growth media failed to show more than about 100 cells/mL (cm2) in any

sample, and there were no apparent differences between water and biofilm samples in microbial growth tests

using standard commercially available microbial growth media used for the quantification of corrosion

associated bacteria in oil industry samples. However, the sulfur isotope fractionation data and the presence of

acetic acid are consistent with microbial conversion of sulfate to hydrogen sulfide, so biochemical and genetic

testing was used to explore the microbiology of these samples more thoroughly. Biochemical testing was

performed on water samples to detect and quantify ATP (adenosine triphosphate) that is the biochemical source

of energy found in bacterial, fungal, plant and animal cells. ATP testing indicated that microorganisms were

present at higher concentrations than the microbial growth tests indicated (data not shown). Genetic testing

was performed using the quantitative polymerase chain reaction (qPCR) (Klindworth 2012, Zhu 2005) to quantify

the total concentration of microbes, as well as the concentration of some types of microorganisms such as

Sulfate Reducing Bacteria (SRB), Acid Producing Bacteria (APB), and Archeabacteria (such as thermophiles as

methanogens). The qPCR reactions work by targeting specific DNA sequences that are unique to groups of

bacteria, or are unique to conserved regions of genes that encode for enzymes in specific biochemical pathways

such as sulfate reduction, or the formation of organic acids. These genetic tests do not require any growth of

microorganisms and instead rely on the analysis of DNA (deoxyribose nucleic acid) present in the sample. The

presence and the quantity of specific DNA sequences can be determined in qPCR assays to yield data similar to

microbial growth tests and can provide data on total bacteria, SRB, APB, NRB, Archaea, and other types of

bacteria. Permian Basin water samples showed 100-to-1000 total microbial cells/mL by qPCR, while biofilm

samples contained up to 1,000,000 total microbial cells/cm2. Bacteria prefer to grow as biofilms attached to

surfaces such as the walls of pipelines or the pores rocks in geological formations, and the concentration of

bacteria in biofilms is typically orders of magnitude higher than the concentration of bacteria in a water sample

from the same location (Stewart 1998). Therefore the genetic testing/qPCR results showed that significant

concentrations of microbes were present in these Permian Basin samples even though standard microbial

growth tests failed to detect most of these microbes. Furthermore, the genetic/qPCR results for these Permian

Basin samples confirmed that microbial concentrations in biofilms are significantly higher than the microbial

concentration in water samples from the same location.

The failure of commercially available microbial growth media to perform well with these Permian Basin samples

could be because of the differences of Permian Basin samples from samples from other locations in the oil & gas

industry (Fichter 2015). For example, many oil industry water samples have salinities similar to seawater (3%)

and are derived from locations that have temperatures around 90oF (32oC), whereas the Permian Basin samples

have salinities of about 10% and the reservoir temperature is about 140oF (60oC).

6 SPE SPE-173709-MS

The sulfur isotopic ratio data, the presence of acetic acid in produced water, and the qPCR results all indicate

that microbial populations are present in this Permian Basin location, but these analyses do not provide insight

as to why traditional microbial growth tests did not work. To gain additional information about the composition

of the microbial community in these samples DNA sequence analysis was performed to determine the DNA

sequences of a highly conserved gene, the 16S ribosomal RNA gene, that is uniquely present in all

microorganisms, but not in other types of cells such as human or plant cells (Gittel 2009, Klindworth 2012). qPCR

analyses allow for the detection and quantification of specific DNA sequences, but does not provide detailed

information about the genus and species of microbial cultures that are present. The DNA sequence analysis does

provide detailed information and by comparing the DNA sequences obtained from environmental samples to

computer databases of all known/characterized microbial species it is possible to obtain more specific

information regarding the composition of a microbial community (Gittel 2009).

Figure 2. Microbial community composition of the water collection/treatment sample at the Permian Basin site.

The microbial community composition of the sample from the water collection/treatment system indicates that

69% of microbes present are Aerobacter (aerobic Enterobacteriaceae that can degrade hydrocarbons and acetic

acid), 10% Thiomicrospira (SRB), 7% Desulfuromonas (SRB), 5.6 % unclassified, and lesser concentrations of

halophilic (salt loving) anaerobic bacteria that includes Acid Producing Bacteria like Clostridium and diverse SRB

cultures. It should be pointed out that the technique used to obtain DNA sequence data as reported in Figures 2

and 3 targeted only Eubacteria, and did not allow the detection of Archaeabacterial DNA sequences. So the

actual microbial community compositions of these Permian Basin samples are even more complex results

reported in Figures 2 and 3 because qPCR analyses show that significant concentrations of Archaeabacteria are

present as well.

Frequently, the most readily accessible samples for microbiological testing are water samples, and while biofilm

samples are better for microbiological testing they may not be available. The microbiological testing previously

performed at this Permian Basin site exclusively used water samples, and as Figure 2 shows, the microbes

present are primarily not the anaerobic Acid Producing Bacteria and Sulfate Reducing Bacteria that are typically

present in oilfield samples, so it is not surprising that the commercially available APB and SRB microbial growth

media did not do a good job in quantifying microorganisms present in water samples from this site. The high

concentration of aerobic bacteria in this water sample indicates that oxygen corrosion is a potential concern at

this location, and that the microbial community in the above-ground water handling system is fundamentally

SPE SPE-173709-MS 7

different than the microbial community present in the subsurface, as will be described in greater detail in Figure

3 below.

Figure 3. Microbial community composition of a biofilm sample from a Permian Basin production well.

In contrast to the composition of the microbial community found in the water collection/treatment sample,

there were no aerobic bacterial species detected in the biofilm sample obtained from a Permian Basin wellhead,

and only anaerobic bacteria were identified. The vast majority of microbial cultures identified in the biofilm

sample were SRB, along with thermophilic (heat loving) and halophilic anaerobic cultures. It is also important to

note that more than 16% of the DNA sequences in this sample are listed as unclassified and could not be

identified because there was no match in the DNA sequence database. This high concentration of unclassified

microbial species demonstrates that this microbial community is indeed rather different than microbial

communities found in other oilfield samples, and it further explains why traditional microbial growth media did

not work well to detect and quantify the microorganisms present in these Permian Basin samples.

The biofilm sample obtained from a Permian Basin wellhead, as shown in Figure 3, is a better sample as regards

characterizing the subsurface microbial community, and that microbial community contains numerous

uncharacterized microbial species and could have different nutritional requirements than microbes from other

oilfield environments. Accordingly, Intertek prepared and tested a variety of microbial growth media

formulations to identify media that would work well for these Permian Basin samples. The results shown in

Table 3 illustrate the results obtained with both qPCR and microbial growth tests using media optimized for

Permian Basin samples.

8 SPE SPE-173709-MS

Values in Table 3 are in units of cells/cm2

It can be seen in Table 3 that the concentration of total bacteria as measured by qPCR in 15 biofilm samples

from the Permian Basin ranges from 104 to 106 cells/cm2, and SRB concentrations range from 103 to 106

cells/cm2. So the qPCR results indicate that the biofilm samples with the highest microbial populations are wells

15, 3, 13, and 10 in that order. The growth test results using media optimized for Permian Basin samples indicate

that the wells with the highest microbial populations are 3, 13, 10, and 6 in that order. So both methods

identified the same three wells (3, 13, and 10) as having the highest microbial concentrations out of the group of

fifteen biofilm samples analyzed. Well #15 scored high in qPCR testing but low in microbial growth testing.

Similarly, well #6 scored fairly high in growth tests, but not in qPCR. A larger data set is needed to more

completely demonstrate the correlation of the results obtained with microbial growth media optimized for the

Permian Basin and qPCR, but it is clear that the new Permian microbial media performs better than traditional

growth media that failed to show significant growth with samples from these locations (Fichter 2015). It is also

important to note that the use of biofilms rather than water samples greatly improves the reliability of the

results obtained in SRB media. This is because water samples from these Permian Basin wells contain significant

concentrations of H2S that can react immediately with the chemicals in SRB growth media yielding a false

positive result, and preventing the detection of low concentrations of SRB. However, when biofilms are sampled

they are collected using cotton swabs and transferred to a sterile sulfide-free liquid. This avoids the false

positives caused by H2S in the inoculum, and the use of biofilms enables the detection of low concentrations of

SRB with confidence. Since there are no convenient field methods for the qPCR analysis of oilfield samples at

this time, the development of microbial growth media optimized for use with Permian Basin samples provides a

convenient way to assay microbial concentrations in the field at this location so that efforts to understand and

qPCR qPCR qPCR qPCR

Well #

Total

Bacteria SRB APB Archaea SRB APB

1 6.1 x 104 3.5 x 104 9.4 x 103 1.4 x 104 102 102

2 4.4 x 104 2.8 x 104 1.4 x 104 6.2 x 103 101 102

3 8.0 x 106 2.6 x 106 1.4 x 106 1.5 x 106 103 106

4 3.3 x 104 1.5 x 104 5.4 x 103 3.0 x 103 101 100

5 4.6 x 104 3.5 x 104 3.7 x 104 3.4 x 104 101 102

6 6.3 x 104

3.1 x 104

< 103

5.8 x 103

102

103

7 4.9 x 104

3.6 x 104

6.6 x 103

2.9 x 104

102

101

8 1.9 x 105

7.1 x 104

5.5 x 103

9.6 x 103

101

100

9 5.4 x 104 2.4 x 104 1.7 x 103 4.8 x 103 101 101

10 6.2 x 105 1.3 x 105 5.0 x 104 3.7 x 104 103 104

11 1.1 x 105 2.2 x 104 1.1 x 104 2.0 x 103 101 103

12 8.0 x 105 6.9 x 104 7.4 x 104 5.7 x 103 103 102

13 6.0 x 105 3.1 x 105 2.2 x 105 1.2 x 105 103 105

14 1.1 x 104 6.4 x 103 2.9 x 103 2.3 x 103 101 101

15 3.3 x 106

2.1 x 106

6.7 x 104

1.4 x 104

101

102

28-day growth

results with Permian

media

Table 3. Comparison of qPCR and microbial growth tests results using new

media optimized for Permian Basin samples.

SPE SPE-173709-MS 9

control reservoir souring can be monitored. ATP quantification can be performed in the field and while it does

not identify the types of microorganisms present it is a useful tool for measuring the overall microbial

concentrations in oilfield water samples.

Knowing that reservoir souring at this Permian Basin location is due to microbial activity helps to define

strategies to deal with the situation. Biocides can be used to reduce the concentration of microorganisms in

injected fluids and in wellbores, and a microbiological monitoring program, focused on biofilm samples, was

implemented. Nitrate injection can be used to influence microbial metabolism in the subsurface to prevent

sulfate reduction, and can even be used to reverse reservoir souring (DeGusseme 2009). The use of sulfide

scavengers, corrosion inhibitors, and reservoir modeling can also be helpful in managing reservoir souring.

Without intervention reservoir souring caused by microorganisms always gets worse, but reservoir souring and

microbiologically influenced corrosion can be prevented or controlled when guided by appropriate data.

Conclusions

In summary, while initial testing failed to detect the presence of significant concentrations of microbes in

samples from a Permian Basin location that was experiencing reservoir souring, more detailed investigation

demonstrated that reservoir souring was due to microbiological sulfate reduction. Genetic testing, sulfur isotope

fractionation data, ATP data and the presence of acetate in produced water all indicate that sulfate reducing

bacteria are present and responsible for reservoir souring. The unusual composition of the microbial community

at this Permian Basin location explains why traditional microbial growth media formulations failed, but new

microbial growth media formulations tailored to the nutritional requirements of this community were

developed and shown to be useful to monitor microbial concentrations in Permian Basin samples.

References

1. Balliger, C., M. H. Schroth, S. M. Bernasconi, J. Kleikemper, J. Zeyer. (2001) “Sulfur isotope fractionation

during microbial sulfate reduction by toluene-degrading bacteria” Geochemica et Cosmochemica Acta

65(19):3289-3298.

2. DeGusseme, B., P. DeSchryver, M. DeCooman, K. Verbekan, P. Boeckx, W. Verstraete, N. Boon. (2009)

“Nitrate-reducing, sulfide-oxidizing bacteria as microbial oxidants for rapid biological sulfide removal”

FEMS Microbial Ecology 67:151-161.

3. Detmers, J., V. Bruchert, K. S. Habicht, J. Kuever. (2001) “Diversity of sulfur isotope fractionations by

sulfate-reducing prokaryotes” Applied & Environmental Microbiology 67(2):888-894.

4. Fichter, J., E. Summer, C. Janes, G. Hamblin. (2015) “Direct comparison of microbial population analysis

using ATP quantification, culture based enumeration in bug bottles, epiflourescence microscopy, and

shotgun population analysis by sequencing 16S amplicons” NACE2015 paper C2015-5555.

5. Gittel, A., K. B. Sorensen, T. L. Skovhus, K. Ingvorsen, A. Schramm. (2009) “Prokaryotic community

structure and sulfate reducer activity in water from high-temperature oil reservoirs with and without

nitrate treatment” Applied & Environmental Microbiology 75:7086-7096.

10 SPE SPE-173709-MS

6. Habicht, K. S., D. E. Canfield. (1997) “Sulfur isotope fractionation during bacterial sulfate reduction in

organic-rich sediments” Geochemica et Cosmochemica Acta 61:5351-5361.

7. Haveman, S. A., K. Pedersen. (2002) “Distribution of culturable microorganisms in Fennoscandian Shield

groundwater” FEMS Microbial Ecology 39:129-137.

8. Klindworth, A. et al (2012) “Evaluation of general 16S ribosomal RNA gene PCR primers for classical and

next-generation sequencing-based diversity studies” Nucleic Acids Research pp. 1-11

doi:10.1093/nar/gks808.

9. Schwermer, C. V.,G. Lavik, R. M. M. Abed, B. Dunsmore, T. G. Ferdelman, P. Stoodley, A. Gieseke, D.

deBeer. (2008) “Impact of nitrate on the structure and function of bacterial biofilm communities in

pipelines used for injection of seawater into oil fields” Applied & Environmental Microbiology 74:2841-

2851.

10. Stam, M. C., P. R. D. Mason, C. Pallud, P. Van Cappellen. (2010) “Sulfate reducing activity and sulfur

isotope fractionation by natural microbial communities in sediments of a hypersaline soda lake (Mono

Lake California)” Chemical Geology 278:23-30.

11. Stewart, P. S., L. Grab, J. A. Diemer. (1998) “Analysis of biocide transport limitation in an artificial biofilm

system” Journal of Applied Microbiology 85:495-500.

12. Struchtmeyer C. G., J. P. Davis, M. S. Elshahed. (2011) “Influence of the drilling formulation process on

the bacterial communities in thermogenic natural gas wells of the Barnett Shale” Applied &

Environmental Microbiology 77(14):4744-4753.

13. Tabari, K., M. Tabari, O. Tabari. (2011) “Investigation of reservoirs souring reasons after water flooding”

Australian Journal of Basic & Applied Sciences 5(12):952-954.

14. Zhu, X. Y., J. J. Kilbane. (2005) “Faster and more accurate data collection for microbiologically influenced

corrosion” SPE International Symposium on Oilfield Chemistry, paper no. SPE-93089-PP.