Microbially Enhanced Oil Recovery Through Stimulation of Indigenous Oil Field Microflora with Nitrate Or Introduction of Rhamnolipid Producers

University of Calgary PRISM: University of Calgary's Digital Repository

Graduate Studies The Vault: Electronic Theses and Dissertations

2013-01-03 Microbially enhanced oil recovery through stimulation of indigenous oil field microflora with nitrate or introduction of rhamnolipid producers

Kryachko, Yuriy

Kryachko, Y. (2013). Microbially enhanced oil recovery through stimulation of indigenous oil field microflora with nitrate or introduction of rhamnolipid producers (Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/26902 http://hdl.handle.net/11023/380 doctoral thesis

University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca UNIVERSITY OF CALGARY

Microbially enhanced oil recovery through stimulation of indigenous oil field

microflora with nitrate or introduction of rhamnolipid producers

Yuriy Kryachko

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF BIOLOGICAL SCIENCES

CALGARY, ALBERTA

DECEMBER, 2012

Sandpack model columns filled with two kinds of heavy oil were used to study the influence of nitrate injections on oil production. Columns containing trapped oil and nitrate consistently produced more oil than columns without nitrate in anaerobic conditions under counter-diffusion-induced gas drive. The presence of nitrate likely led to an increase of biomass and amounts of produced biosurfactants, allowing a reduction of oil-water interfacial tension.

Gas-driven recovery of oil rich in alkanes (EO) was enhanced to a greater extent through nitrate injection than recovery of oil rich in polyaromatic hydrocarbons (LO). A water flood following incubation of columns containing nitrate did not lead to trapped oil production indicating that microbial growth was not sufficient to plug high permeability zones. Nitrate reducers Azoarcus and

Delftia, the former being from the field of origin of EO, were found to consume heptane as a sole source of carbon and energy.

Bioinformatics analysis of pyrosequenced 16S rDNA of the microbial communities adhering to EO and associated with an aqueous phase of a produced water-EO mixture showed that the former community was less diverse than the latter. Many known hydrocarbon degraders, e.g.

Thalassolituus, Rhodococcus and Sphingomonas, were found to adhere to EO.

In contrast, all identified genera of the Deltaproteobacteria were found to reside predominantly in the aqueous phase, likely because their substrates were mostly water-soluble. Consistently with the fact that hydrogen is more soluble in oil than in water, hydrogenotrophs, e.g. Methanolobus, Methanobacterium and

Acetobacterium, had higher representation in the microbial community adhering to EO than in the community associated with an aqueous phase.

The application of a cell-containing supernatant from the culture of a native rhamnolipid producer led to a greater enhancement of water-flood-driven

LO recovery than the application of chemical surfactants. The application of cell-containing supernatants from cultures of a recombinant strain carrying genes responsible for rhamnolipid production and a strain with repressed transcription of rhamnolipid production genes led to enhancement of counter- diffusion-induced-gas-driven EO recovery, suggesting that very low surfactant concentrations are required for MEOR under solution gas drive.

iii

Acknowledgements

This work was supported by funding from Aramco Services, Baker

Hughes Incorporated, British Petroleum, Intertek/CML, the Computer Modeling

Group Limited, ConocoPhillips Company, YPF SA, Shell Canada Limited,

Suncor Energy Developments Inc., and Yara International ASA, Alberta

Innovates – Energy and Environment Solutions, Genome Canada, Genome

Alberta, the Government of Alberta and Genome BC.

I am thankful to my supervisor, Dr. Gerrit Voordouw for giving me an opportunity to run exciting experiments, for insightful discussions, flexibility and patience, members of my supervisory committee, Dr. Doug Storey and Dr. Lisa

Gieg for numerous helpful comments and suggestions, Dr. Harvey Yarranton and Dr. Tom Jack for their recommendations regarding my research and participation in supervisory committee meetings, Dr. Christoph Sensen, Dr.

Elmar Prenner, Johanna Voordouw, Patrick Lai, Safia Nathoo and Xiaoli Dong for collaboration during preparation of our articles for publication, Sheng-Hung

Wang for helping analyze DNA pyrosequencing data, Dr. Akhil Agrawal, Dr.

Dae-Kyun Ro, Dr. Lisa Gieg and Trinh-Don Nguyen for assistance with GCMS analysis, Shiping Lin for help with assembling sandpack columns, Dr. Brij Maini and Paul Stanislav for help with the engineering aspects of my work, Ryan

Ertmoed and Adriana Cavallaro for providing samples of oil and produced water, Dr. Doug Storey for providing the strains of Pseudomonas aeruginosa,

Dr. Sean Caffrey for organizational efforts and recommendations on the applications of statistical methods, Dr. Hyung Soo Park, Dr. Adewale Lambo,

Cameron Callbeck and Jaspreet Mand for assistance with conducting HPLC analysis, Morgan Khan for assistance with MALDI-TOF analysis, Cameron

Callbeck and Jane Fowler for help with DGGE analysis, Dr. Rhonda Clark, Dr.

Indranil Chatterjee and Dr. Dongshan An for their coordinating contributions, all sponsors of the Petroleum Microbiology Research Group, and all members of the Dr. Voordouw’s and Dr. Gieg’s laboratories at the University of Calgary with whom I worked and all of whom are my friends. Playing soccer and basketball with some of my colleagues helped me maintain myself in a good physical and mental condition during my PhD program, and I am thankful to all of them for being wonderful partners and rivals.

Table of Contents

Section Page

Abstract ii Acknowledgements iv Table of Contents vi List of Tables x List of Figures xi List of Abbreviations xiv

Chapter 1: Overview of principles, successes and problems of MEOR- technology

1.1. Introduction 1 1.2. Alteration targets 2 1.2.1. Oil-water IFT and the role of biosurfactants in its 2 reduction 1.2.2. Wettability changes 8 1.2.3. Plugging of high permeability zones 11 1.2.4. Microbial solvent production in situ 12 1.3. Successful MEOR applications and tests 13 1.3.1. Conditions of success 13 1.3.2. Overview of some significant achievements in the history 18 of MEOR 1.4. Major problems of MEOR 20 1.5. Microbial enhancement of solution-gas-driven oil recovery 22 1.5.1. The principles of solution-gas-driven oil recovery 22 1.5.2. Enhancement of solution-gas-driven oil recovery 24 with surfactants in high and low concentrations

Chapter 2: Research objectives and hypotheses

2.3. Research objectives 29 2.4. Hypotheses 29

Chapter 3: MEOR effects of nitrate injections in experiments with miniature model columns

3.1. Introduction 32 3.2. Materials and methods 34 3.2.1. Oil samples 34 3.2.2. Enrichment cultures with EO or LO as sources of carbon 34 and energy 3.2.3. Column assembly 35 3.2.4. Saturating columns with SPW 37 3.2.5. Saturating columns with oil 37 3.2.6. Eluting oil with SPW 38 3.2.7. Experiments with columns containing trapped oil 38 3.2.8. Determination of the amounts of produced oil and water 44

3.2.9. Oil composition analysis 44 3.2.10. Determination of concentrations of gases produced during 46 long-time incubations 3.2.11. Enrichment cultures with heptane as a sole source of 46 carbon and energy 3.2.12. DNA pyrosequencing 47 3.2.13. Analysis of pyrosequencing data 48 3.3. Results 49 3.3.1. Oil production during long-time incubation of columns with 49 trapped EO and LO 3.3.2. Gas production during long-time incubation of columns with 50 trapped EO and LO 3.3.3. Nitrate and nitrite concentrations at the ends of long-time 55 incubation rounds 3.3.4. SPW-flood- and immiscible-gas-flood-driven oil production 55 following long-time incubation rounds 3.3.5. Compositions of EO and LO before and after long-time 56 incubation of columns with trapped oil 3.3.6. Biodegradation of heptane by NRB in enrichment cultures 64 3.4. Discussion 65 3.4.1. Mechanisms of oil production during long-time 65 incubation of columns with trapped oil 3.4.2. The mechanism of nitrate-mediated enhancement of gas- 75 driven oil recovery 3.4.3. Possible substrates for microbial growth during long-time 77 incubation of columns with trapped oil 3.4.4. NRB capable of heptane biodegradation 78 3.4.5. Long-time incubation of columns with trapped oil was not 80 accompanied by plugging of high permeability zones 3.4.6. Increased nitrate concentration did not help further enhance 81 gas-driven oil recovery 3.5. Conclusions and implications 82

Chapter 4: Microbial communities associated with oil and water in a mesothermic oil field

4.1. Introduction 84 4.2. Materials and methods 85 4.2.1. Isolation of biomass 85 4.2.2. DNA isolation 88 4.2.3. DGGE 89 4.2.4. DNA pyrosequencing 89 4.2.5. Analysis of pyrosequencing data 90 4.3. Results 91 4.3.1. DGGE community analysis 91 4.3.2. Alpha- and beta-diversity 91 4.3.3. Characterization of the OCP and the ACP 92 4.3.3.1. Phylum and class levels 92 4.3.3.2. Order and genus levels 99 4.4. Discussion 107

vii

4.4.1. Microorganisms associated with the oil phase 107 4.4.1.1. Gammaproteobacteria 107 4.4.1.2. Alphaproteobacteria 107 4.4.1.3. Betaproteobacteria 108 4.4.1.4. Actinobacteria 109 4.4.1.5. Mollicutes and Clostridia 109 4.4.2. Positioning of methanogens 110 4.4.3. Other microorganisms associated with the aqueous phase 112 4.4.3.1. Deltaproteobacteria 112 4.4.3.2. Alphaproteobacteria 112 4.4.3.3. Bacteroidetes and Synergistetes 113 4.4.4. Representation of diazotrophs 113 4.5. Conclusions and implications 114

Chapter 5: Native and recombinant rhamnolipid producers as MEOR- agents

5.1. Introduction 116 5.2. Materials and methods 119 5.2.1. Constructing recombinant E. coli and isolating CSs 119 5.2.2. Rhamnolipid extraction 122 5.2.3. MALDI-TOF analysis 123 5.2.4. Surface pressure measurements 123 5.2.5. TLC analysis 124 5.2.6. Hexadecane emulsification test 124 5.2.7. Monitoring oil and water production from model columns 125 5.2.7.1. Oil samples 125 5.2.7.2. Column assembly 125 5.2.7.3. Saturating columns with SPW 125 5.2.7.4. Saturating columns with oil 125 5.2.7.5. Eluting oil with water 126 5.2.7.6. Experiments with columns containing trapped oil 126 5.2.7.7. Determination of the amounts of produced oil and water 128 5.2.7.8. Determination of the concentrations of gases produced 128 during long-time incubations 5.3. Results 128 5.3.1. Properties of CSs 128 5.3.2. Recovery of trapped oil from sandpack columns 139 5.3.2.1. Recovery of oil during CS injections and floods 139 5.3.2.2. Recovery of oil during long-time incubations and media 142 replenishments 5.3.3. Gas production during long-time incubations 144 5.4. Discussion 144 5.4.1. Determinants of the surface activity of CSs 144 5.4.2. Effects of short-time incubations and CS/solution floods 145 5.4.3. Effects of long-time incubations 148 5.5. Conclusions and implications 151

viii

Chapter 6: Main conclusions and directions for future research 153

References 157

Appendices 191

List of Tables

Table Page

Table 1. Oil and water production from columns containing trapped oil 52 with and without nitrate during incubations in an anaerobic hood at 30˚C Table 2. N2 and CO2 production from columns containing trapped oil with 53 and without nitrate during long-time incubations in an anaerobic hood at 30˚C Table 3. Concentrations of nitrate and nitrite at the beginnings and the 54 ends of incubation rounds A and B conducted with the set IV columns containing LO Table 4. Relative concentrations of selected low molecular weight 58 hydrocarbons in EO: ratios of areas for different peaks to peak areas for pristane used as an internal standard Table 5. Relative concentrations of selected low molecular weight 61 hydrocarbons in LO: ratios of areas for different peaks to peak areas for phenanthrene used as an internal standard Table 6. Numbers and fractions of reads for genera found in enrichment 73 cultures (microcosms) with heptane as a sole source of carbon and energy Table 7. Alpha-diversity indices calculated for enrichment cultures 74 (microcosms) with heptane as a sole source of carbon and energy Table 8. Numbers of reads for AC-1 – AC-5 and OC-1 – OC-5 identified 95 at the phylum (NP), class (NC) and genus (NG) levels Table 9. Parsimony, weighted unifrac, unweighted unifrac, and AMOVA 95 hypothesis testing results Table 10. Alpha-diversity indices calculated for the OC and the AC 96 Table 11. Numbers and fractions of reads for phyla found in the OCP and 97 the ACP Table 12. Numbers and fractions of reads for classes found in the OCP 98 and the ACP Table 13. Numbers and fractions of reads for genera found in the OCP 100 and the ACP Table 14. Lipid concentrations in CSs 130 Table 15. Production of oil and water from columns containing trapped 141 LO and EO (sets 1, 2 and 3) under different conditions Table 16. Oil, water and gas (N2 and CO2) production from columns 147 containing (bio)surfactant solutions during long-time incubations in an anaerobic hood at 30˚C Table A1. Volumes of solutions injected into columns with trapped EO 191 and LO before the beginnings of long-time incubation rounds. Table A2-1. Rounds of long-time incubations for column set I 192 Table A2-2. Rounds of long-time incubations for column set II 192 Table A2-3. Rounds of long-time incubations for column set III 192 Table A2-4. Rounds of long-time incubations for column set IV 192

List of Figures Figure Page

Figure 1A. Meeting of a gas bubble with an oil droplet at a high 25 surfactant concentration Figure 1B. Meeting of a gas bubble with an oil droplet when a surfactant 25 is present in a low concentration Figure 2. A partially-assembled 20 mL syringe sandpack column 36 Figure 3A. Water-flooding of a 30 mL syringe sandpack column 39 containing oil using the Mandel-Gilson Minipulse 3 peristaltic pump Figure 3B. A close-up of a water-flooded 30 mL syringe sandpack 39 column with an attached 50 mL polypropylene tube for collecting the effluent Figure 4. EO production from a 20 mL column flooded with SPW at a 40 flow rate of 6.5 mL/hour during several hours in the beginning of the flood Figure 5. EO production from 30 mL columns (3 replicates) when 41 flooded with SPW Figure 6. LO production from 30 mL columns (3 replicates) when flooded 41 with SPW Figure 7A. Addition of DCM to a fraction of the produced oil and water 45 Figure 7B. Absorbance of heavy oil in DCM: a standard curve 45 Figure 8A. Production of oil and water from columns from sets I and II 51 during long-time incubations at 30ºC Figure 8B. Production of water from the 30-mL column with SPW (no oil) 51 during long-time (45-day) incubation at 20ºC Figure 9. GC-MS chromatograms of the samples of original EO and LO 57 Figure 10. Low molecular weight components in the samples of original 57 EO and LO, as determined through GC-MS analysis Figure 11. Low molecular weight hydrocarbons present in the original 59 EO and in columns with EO after incubations Figure 12. Higher molecular weight hydrocarbons present in the original 59 EO and in columns with EO after incubations Figure 13. Low molecular weight compounds in the original EO and EO 60 extracted from the water-flooded (containing trapped oil), but not subjected to a long-time incubation column V-I Figure 14. Contents of some low molecular weight hydrocarbons in LO 62 after incubations (compared to their original contents in LO) Figure 15. Contents of higher molecular weight hydrocarbons in LO after 62 incubations (compared to their original contents in LO) Figure 16. Enrichment culture of NRB capable of consuming heptane as 69 a sole source of carbon and energy Figure 17. Concentrations of nitrate and nitrite in microcosm #1 as 70 functions of time Figure 18. Concentrations of nitrate and nitrite in microcosm #3 as 70 functions of time Figure 19. Concentrations of nitrate and nitrite in microcosm #5 as 71 functions of time

Figure 20. Concentrations of nitrate and nitrite in microcosm #6 as 71 functions of time Figure 21. Concentrations of nitrate and nitrite in microcosm #7 as 72 functions of time Figure 22. Concentrations of nitrate and nitrite in microcosm #8 as 72 functions of time Figure 23. Rarefaction curves for enrichment cultures (microcosms) with 74 heptane as a sole source of carbon and energy Figure 24. Separation of produced water and oil (EO) in an anaerobic 87 hood filled with 90% N2 and 10% CO2 Figure 25. 16S rRNA gene fragments of microorganisms belonging to 93 the communities associated with the oil phase (OC-dgge-1 and OC-dgge-2) and the aqueous phase (AC-dgge-1 and AC-dgge- 2), analyzed by DGGE Figure 26. Clustering analysis of pyrosequencing data. (A) Jaccard 94 sample relation tree built using the UPGMA clustering algorithm based on the classical Jaccard values. (B) ThetaYC tree built using the UPGMA algorithm based on Yue and Clayton (2005) theta values Figure 27. Rarefaction curves for the OCP and the ACP 96 Figure 28. Phyla found in the OCP 104 Figure 29. Phyla found in the ACP 104 Figure 30. The most abundant classes found in the OCP 105 Figure 31. The most abundant classes found in the ACP 105 Figure 32. The most abundant genera found in the OCP 106 Figure 33. The most abundant genera found in the ACP 106 Figure 34. Recombinant plasmid pF1bR4 120 Figure 35. Agarose-gel-electrophoresis analysis of digesting pF1bR4 121 plasmid with restriction enzymes Figure 36. MALDI-TOF mass-spectrum for the commercial rhamnolipid 130 solution in the m/z range where most rhamnolipid peaks are expected to be found Figure 37. MALDI-TOF mass-spectrum for P. aeruginosa PA14 CS 131 extract in the m/z range where most rhamnolipid peaks are expected to be found Figure 38. MALDI-TOF mass-spectrum for E. coli pF1bR4 CS extract in 131 the m/z range where most rhamnolipid peaks are expected to be found Figure 39. MALDI-TOF mass-spectrum for P. aeruginosa PDO111 CS 132 extract in the m/z range where most rhamnolipid peaks are expected to be found Figure 40. MALDI-TOF mass-spectrum for E. coli TG2 CS extract in the 132 m/z range where most rhamnolipid peaks are expected to be found Figure 41. Emulsification of hexadecane by P. aeruginosa PA14, E. coli 133 pF1bR4 and E.coli TG2 CSs Figure 42. MALDI-TOF peak for an ion with the m/z ratio equaling 134 361.96 – 361.99 in the commercial rhamnolipid solution Figure 43. MALDI-TOF peak for an ion with the m/z ratio equaling 134 361.96 – 361.99 in the P. aeruginosa PA14 CS extract

xii

Figure 44. MALDI-TOF peak for an ion with the m/z ratio equaling 135 361.96 – 361.99 in the E. coli pF1bR4 CS extract Figure 45. MALDI-TOF peak for an ion with the m/z ratio equaling 135 361.96 – 361.99 in the P. aeruginosa PDO111 CS extract Figure 46. MALDI-TOF peak for an ion with the m/z ratio equaling 136 361.96 – 361.99 in the E. coli TG2 CS extract Figure 47. Results of the TLC analysis of P. aeruginosa PA14, E.coli 137 F1bR4, P. aeruginosa PDO111 and E. coli TG2 CS extracts, and the commercial rhamnolipid solution Figure 48. Surface pressure as a function of the area per molecule: 138 results of monolayer experiments completed with lipid extracts from P. aeruginosa PA14, P. aeruginosa PDO111, E. coli pF1bR4 and E. coli TG2 CSs, the commercial rhamnolipid solution and TY medium Figure 49. EO production from set 3 columns containing one of the three 140 CSs or 1% SDS during long-time incubation Figure A1. Experiments with set I – IV columns containing trapped oil 193 and nitrate-containing or nitrate-deficient media Figure A2. Nitrate-containing and nitrate-deficient media introduced into 193 set I – III columns prior to long-time incubation rounds Figure A3. Nitrate-containing and nitrate-deficient media introduced into 194 set IV columns prior to long-time incubation rounds Figure A4. Experiments involving introduction of CSs or surfactant 195 solutions into columns containing trapped oil

xiii

List of Abbreviations

AC microbial community associated with an aqueous phase AC-dgge microbial community associated with an aqueous phase analyzed by DGGE ACP pooled microbial community associated with the aqueous phase AHL N-acyl homoserine lactone CS culture supernatant CMC critical micelle concentration CTAB cetyl trimethyl ammonium bromide DCM dichloromethane DGGE denaturing gradient gel electrophoresis EDTA ethylenediaminetetraacetic acid EO oil from the MHGC (also known as the Enermark) field EOR enhanced oil recovery GC-MS gas chromatography-mass spectrometry HPLC high performance liquid chromatography hNRB heterotrophic nitrate reducing bacteria IFT interfacial tension LO oil from the Loma Alta Sur (LAS) field (Argentina) MALDI-TOF matrix-assisted laser desorption/ionization time-of-flight mass spectrometry MATH microbial adhesion to hydrocarbons MEOR microbially enhanced oil recovery m/z mass to charge ratio of an ion OC microbial community associated with an oil phase OC-dgge microbial community associated with an oil phase analyzed by DGGE OCP pooled microbial community associated with an oil phase OOIP original oil in place OTU Operational Taxonomic Unit PCR polymerase chain reaction PV pore volume qPCR quantitative real time polymerase chain reaction SDS sodium dodecyl sulphate SRB sulphate reducing bacteria soNRB sulfide-oxidizing nitrate reducing bacteria SPW synthetic produced water TLC thin layer chromatography VFA volatile fatty acids Vo volume of oil Voi volume of oil produced during long-time incubation Vot volume of trapped oil Vw volume of water Vwi volume of water produced during long-time incubation WAG water alternating gas process

xiv

Chapter 1: Overview of principles, successes

and problems of MEOR-technology

1.1. Introduction

Modern society heavily depends on crude oil as a source of energy (Hall et al., 2003; Youssef et al., 2007). Currently used production technologies recover only 30 – 50% of oil present in a reservoir (Hall et al., 2003; Planckaert,

2005; Youssef et al., 2007), with the rest remaining entrapped in small pores.

Methods of enhanced oil recovery (EOR), e.g. application of heat, chemicals, miscible gas injection, and microbial processes (Green and Willhite, 1998) are used for the recovery of some of this entrapped oil.

The technology known as microbially enhanced oil recovery (MEOR) relies on the activity of microbial cells and their metabolites (Banat, 1995;

Bryant and Burchfield, 1989; Li et al., 2002; McInerney et al., 2005a). Although the feasibility of MEOR has been critiqued (Gray et al., 2008), well-controlled studies showed that MEOR is feasible and cost-effective (Zhaowei et al., 2008;

Youssef et al., 2007; Banat, 1995; Brown et al., 2002; Bryant et al., 1993;

McInerney et al., 2005a).

According to Lazar et al. (2007), some of the advantages of MEOR, compared to the other EOR technologies, are the following:

- the injected bacteria and nutrients are not expensive; it is not difficult to obtain and handle them in the field;

- they are economically attractive for marginally producing oil fields; therefore, they could be a suitable alternative before the abandonment of marginal wells;

- the implementation of the MEOR process needs only minor modifications of the existing field facilities;

- MEOR processes were shown to be particularly effective for carbonate oil reservoirs, where other EOR technologies could not be applied with good efficiency;

- the effects of bacterial activity within the reservoir may be magnified by their growth, while in other EOR technologies the effects of the additives usually decrease with time and distance;

- MEOR products are environmentally friendly: being biodegradable, they do not accumulate in the environment.

1.2. Alteration targets

Reduction of interfacial tension (IFT), wettability changes, plugging of high permeability zones in oil reservoirs, and production of solvents in situ have been paid much attention to in the literature and may be considered the main targets of MEOR-technology. They will be discussed in detail further in this section.

1.2.1. Oil-water IFT and the role of biosurfactants in its reduction

Residual oil exists mainly as globules dispersed within the pores of a reservoir (Willhite and Green, 1998; Terry, 2001). There is insufficient pressure gradient to move these globules through the surrounding pore throats under existing reservoir conditions (Gray et al., 2008). The globule is under the viscous force promoting flow that is opposed by the capillary force. The capillary number is the ratio of viscous to capillary forces, Nca=uμ/σ, where u is velocity

(m/s), μ is viscosity (Pa*s), and σ is interfacial tension (N/m). Interfacial tension

(IFT) can be defined as the Gibbs free energy per unit area of the interface between two immiscible phases at given temperature and pressure; it occurs because of the difference between molecular interactions near an interface and those within the bulk fluid (Schlumberger, 2012a).

Large capillary numbers are required for mobilization of the residual oil and its higher recovery from the reservoir when water flood is applied. The capillary number for most reservoirs under water flood is in the order of 10-7; and it must be increased to 10-5 to 10-4 to mobilize residual oil (Willhite and

Green, 1998; Gray et al., 2008); that is, surfactants, when they are applied in combination with water flood, must be capable of reducing oil-water IFT by at least two orders of magnitude (Gray et al., 2008). Typically, the interfacial tension between hydrocarbons and water is in the order of 30 to 40 mN/m, and, to be considered effective, the surfactant must reduce interfacial tension to at least below 0.4 mN/m. The recommended target interfacial tension for chemical surfactant floods is 0.001 to 0.01 mN/m. For example, core floods with commercial surfactants reducing interfacial tensions to 0.03 and 0.002 mN/m led to mobilization of 50 and 70% of the residual oil, respectively (Willhite and

Green, 1998; Gray et al., 2008).

Chemically synthesized surfactants have been used for the clean-up of oil spills and for enhanced oil recovery. However, because of their toxicity and resistance to degradation, they can cause serious environmental problems

(Mulligan, 2005).

In MEOR, free or cell-bound biosurfactants produced by bacteria may decrease IFT (Banat, 1995; Bryant and Burchfield, 1989; McInerney et al.,

2005a; Yonebayashi and Ono, 1997). However, reducing oil-water IFT to desirable 0.001 – 0.01 mN/m (or even just below 0.4 mN/m) may be more difficult with biosurfactants than with commercial surfactants (Gray et al., 2008).

At the same time, biosurfactants have been known to have lower toxicity and higher biodegradability compared to synthetic chemical surfactants (Rosenberg and Ron, 1999). They can be produced by bacteria or yeasts from sugars, glycerol, oils, hydrocarbons and agricultural wastes (Lin, 1996), but their yields are usually low and production costs are high, so that it is difficult to achieve ex situ production of microbial surfactants on a commercial scale (Mukherjee et al.,

2006). Alternatively to an approach relying on their ex situ production, stimulation of indigenous microorganisms or injection and stimulation of known biosurfactant-producers leading to biosurfactant production in situ can be used.

Glycolipids, lipopeptides, phospholipids, fatty acids, neutral lipids, and polymeric or particulate compounds may carry biosurfactant properties (Desai and Banat, 1997; Wang et al., 2007). A long-chain fatty acid, hydroxyl fatty acid or α-alkyl-β-hydroxyl fatty acid can be the hydrophobic part, and a carbohydrate, amino acid, cyclic peptide, phosphate, carboxylic acid or alcohol can be the hydrophilic part of a biosurfactant molecule.

There are indications that the combination of bacterial cells and surface active metabolites can be more efficient in reducing oil-water IFT than the metabolites alone (Kowalewski et al., 2006). Indeed, in the experiments conducted by Bryant et al. (1989), the combined use of bacterial cells and metabolites enhanced oil recovery to a greater extent than the use of just metabolites.

Bouchez-Naitali and Vandecasteele (2008) studied hydrophilic cells of

Pseudomonas aeruginosa and showed that solubilization of hexadecane by biosurfactants was crucial for its biodegradation by this bacterium. Interestingly,

Bouchez-Naitali and Vandecasteele (2008) found, at the same time, that extracellular biosurfactants produced by another bacterium – Rhodococcus equi

– played only a minor role in hexadecane degradation. Its uptake by hydrophobic cells of Rhodococcus equi occurred primarily through direct attachment to hydrocarbons. Obuekwe et al. (2008) found hydrophobic subpopulations of hydrocarbon-degrading Pseudomonas aeruginosa, which produced no or very small amounts of biosurfactants and were able to directly attach to hydrocarbons, to have greater hydrocarbon-utilizing ability than hydrophilic ones, which needed hydrocarbon solubilization and, therefore, produced greater amounts of biosurfactants.

Interesting experiments on the degradative potential of denitrifying bacteria enriched with crude oil have been conducted by Rabus et al. (1999).

Two distinct growth phases were exhibited by the enrichment culture: during the first phase, no alkanes were utilized from the crude oil and bacteria grew homogeneously in the aqueous phase on C1–C3 alkylbenzenes; during the second phase, utilization of n-alkanes (C5 to C12) from the crude oil was observed and the growth of bacteria occurred along with their adhesion to the crude oil layer and oil emulsification. Thus, microorganisms tended to use water-soluble substrates first and began consuming water-insoluble hydrocarbons only upon their depletion. Azoarcus and Thauera predominated in these enrichment cultures (Rabus et al., 1999). Oil solubilization/emulsification

5 and microbial adhesion to an oil-water interface likely required biosurfactant production.

Lambo et al. (2008) and Agrawal et al. (2012) showed that toluene was a preferred electron donor for microbial nitrate reduction in an oil field. There is evidence that alkanes (especially short-chain alkanes) can be readily consumed by NRB as well (Rabus et al., 2001; Wilkes et al., 2002; Grundmann et al.,

2008).

Some authors who predicted high recovery of crude oil due to biosurfactants, assumed their co-production with polymers and alcohols; the presence of water-soluble polymers would lead to increasing viscosity of the aqueous phase thus reducing the water-oil mobility ratio, while surfactants and alcohols would decrease IFT (Sarkar et al., 1991; McInerney et al., 2005b).

According to Gray et al. (2008), this favourable combination of bioproducts from a microbial culture has never been demonstrated and, therefore, the feasibility of this mechanism has to be deemed low. In addition, if the microorganisms consume hydrocarbons present in the reservoir to produce surfactants, and/or alcohols, and/or viscosifying polymers, then the issue of the resulting possible significant decrease of oil mobility (due to increase of oil viscosity if low molecular weight hydrocarbons are preferably consumed) must be taken into account.

In the experiment conducted by Kowalewski et al. (2005), Dietzia sp.

A14101 was able to reduce IFT from 38 to 0.006 mN/m during aerobic growth at n-dodecane/synthetic sea water interface. Youssef et al. (2007) reported that the average concentration of lipopeptide biosurfactant in the fluids produced from an oil field inoculated with two strains of Bacillus was about 90 mg/L,

6 approximately nine times the claimed minimum concentration needed to mobilize trapped oil from sandstone cores. These two publications are quite rare testaments to the capability of biosurfactants to significantly reduce oil- water IFT.

A number of studies have been conducted to elucidate biological properties and roles of biosurfactants. They often show antimicrobial activity

(Ron and Rosenberg, 2001) and may play important roles in cellular differentiation in some organisms (van Hamme et al., 2006). For example, streptofactin was found to be required for aerial mycelium formation in S. tendae (Richter et al., 1998). Production of a large amount of biosurfactant controlled by quorum-sensing was found to be an essential parameter for initiation of sliding (Harshey, 2003). Davey et al. (2003) created rhlA mutants lacking the rhamnosyltransferase enzyme that is important in rhamnolipid production, and proved that the presence of rhamnolipids was essential for the maintenance of biofilm architecture over time. Interestingly, long-chain AHL (N- acyl homoserine lactone) molecules, known to be quorum sensing signaling agents, were found to function as biosurfactants, which were promoting surface colonization (Daniels et al., 2006).

According to López et al. (2009), biofilm formation is stimulated by a variety of small molecules produced by bacteria, e.g. surfactin produced by

Bacillus subtilis. Such molecules can induce potassium leakage to stimulate the activity of a membrane protein kinase (KinC) governing the expression of genes responsible for biofilm formation. However, surfactin and some other bio- and chemical surfactants were found to inhibit biofilm formation by wild-type

Salmonella enterica grown in urethral catheters and polyvinyl chloride microtiter

7 wells and to promote swarming motility through modifying wettability properties of the surfaces. Similar effects were observed in experiments with Escherichia coli and P. mirabilis grown in urethral catheters (Mireles II et al., 2001).

1.2.2. Wettability changes

A biosurfactant adsorbed to a mineral surface also has the potential to alter its wettability/hydrophobicity. Wettability can be defined as the preference of a solid to contact one liquid or gas (the wetting phase) rather than another

(the non-wetting phase); the wetting phase tends to spread on the solid surface and a porous solid tends to imbibe the wetting phase, while the nonwetting phase is being displaced (Schlumberger, 2012b).

According to Gray et al. (2008), since known biosurfactants applied in combination with water floods in most cases do not reduce IFT enough to mobilize oil directly, a mechanism involving mineral surface hydrophobicity/wettability changes may provide an alternative or a complementary way for using biosurfactant activity. Mineral surface wettability can be changed by the presence of microorganisms in a reservoir via several different mechanisms:

- direct attachment of microbes to the surface (limited to pores that are in the range of 1 μm or more, that is, to the pores that are large enough to be accessible to bacteria);

- adsorption of bacterial metabolites to mineral surfaces: the products of microbial metabolism, such as surfactants, can adsorb to mineral surfaces, and have the potential to alter the wettability through increasing or decreasing surface hydrophobicity; this mechanism can operate at any pore diameter and

8 may provide an alternate, compared to IFT reduction, mode for biosurfactants to alter the mobility of oil;

- coating of the mineral surface with exopolysaccharides or other biopolymers (this mechanism is closely linked to cell attachment; although, the polymers can stay on mineral surfaces even after the release or death of the cells; Wanger et al., 2006; Gray et al., 2008).

The three-phase contact angle between the oil, the water, and the mineral surface is the important parameter for determining the local pressure gradient and interfacial tension required to mobilize the oil in a water flood. A certain unique angle of contact leads to capillary effects such as the rise of a liquid up a vertical tube, or the spontaneous imbibition of oil in porous structures. In the case of an oil reservoir, some “imperfections” modify such effects. These “imperfections” include: (a) surface roughness and heterogeneities on the solid substrate that can lead to hysteresis effects in contact angle measurement, and (b) the adsorption of colloidal structures at the fluid-fluid interface may lead to the development of an interfacial “skin” that would complicate surface rheological properties (Gray et al., 2008). Such an elastic “skin” formed by bacteria adsorbed to an oil-water interface can significantly alter the final equilibrium shape compared to the situation without a

“skin”, as additional energy is required for stretching the “skin” (Kang et al.,

2008a; Kang et al., 2008b). Gray et al. (2008) argue that the formation of the

“skin” at the oil-water interface, in fact, can make the “contact angle” irrelevant to “wettability”. However, this “skin” itself may play a role in oil recovery, as it was found to affect transport properties of the emulsified oil droplets, particularly

9 enhance their capability to plug narrow pores and channels (Kang et al., 2008a; section 1.2.3).

According to Rao et al. (1992), a mixed-wet state of rock, in which it is partly oil-wet and partly water-wet, is the optimum wettability state for water- flood-driven oil recovery. Thus, the shift of reservoir wettability from either strongly oil-wet or strongly water-wet states to a mixed wettability state can potentially enhance oil recovery. However, among all wettability states, the oil- wet state is the least favourable (Rao et al., 2006). It turns out that the wettability changes may depend on the initial wetting conditions: microbial growth in an initially oil-wet system can result in a more water-wet state (Mu et al., 2002), while an initially water-wet system can become less water-wet

(Polson et al., 2002). Results of some experiments, in which bacterial growth led to both a wettability change and IFT reduction, showed that a change in wettability from a strongly water-wet towards a less water-wet state was beneficial for oil recovery from cores during water floods (Torsvik et al., 1995;

Kowalewski et al., 2006).

Wettability changes were also found to be more significant for oil recovery from carbonate reservoirs than from sandstone formations (Gray et al.,

2008). In particular, the change from oil-wet to water-wet conditions was found to be beneficial for oil recovery from carbonate reservoirs. However, not much attention has been paid so far to the wettability of carbonate rocks in the presence of bacterial cells (Gray et al., 2008).

1.2.3. Plugging of high permeability zones

Bacteria can penetrate long distances along fractures and high- permeability zones, causing plugging of those zones and leading to improved sweep efficiency of the oilfield water-flood (Beeder et al., 1996; Fujiwara et al.,

2004; Gray et al., 2008). It is the mechanism of MEOR that has consistently received much attention in the literature (Udegbunam et al., 1991; Stepp et al.,

1996; Gullapalli et al., 2000; Kowalewski et al., 2006; Gray et al., 2008).

Its principle is as follows. Indigenous or introduced to an oil reservoir microorganisms grow in high permeability zones until permeability is reduced to such an extent that the injected water is diverted to alternative flow paths; thus, the displacement of initially bypassed oil may be improved. Reduction of permeability should not lead to any increased oil recovery in a homogeneous core because most of the pores can be flooded with water, and, thus, a diversion of the flow path is not likely to occur. Plugging would be of more importance in heterogeneous (especially fractured) cores (Kowalewski et al.,

2006; Gray et al., 2008).

In one of the successful experiments on high-permeability zone plugging,

Fujiwara et al. (2004) added cells of Enterobacter sp. CJF-002 (originally found in the Fuyu oil field in China), molasses and nitrate to reservoir modeling columns to plug such zones in a laboratory test. The bacterium was found to be effective in forming an insoluble cellulose polymer that was capable of adhering to the reservoir rock and was not easily degradable by other organisms.

According to Fujiwara et al. (2004), Enterobacter sp. CJF-002 was able not only to produce large amounts of the polymer, but also outcompete other organisms if this strain was initially present at high local concentrations. Based on their

11 own experiment, as well as on some of the previous studies of plugging

(Jenneman et al., 1982; Lappin-Scott et al., 1988), Fujiwara et al. (2004) set out a series of requirements for successful application of this method:

- the reservoir must have high permeability zones or fractures between injectors and producers resulting in poor sweep efficiency;

- the added organisms must be able to survive in the presence of the crude oil and tolerate the reservoir conditions (e.g. temperature and salinity);

- the biologically produced plugging substance(s) must be resistant to degradation by the indigenous microorganisms (otherwise the treatment would be required at frequent intervals);

- the formation of a long-lasting flow barrier within the formation is essential for the effective treatment;

- the activity of the added organisms does not have to lead to removal of the pre-existing organisms.

1.2.4. Microbial solvent production in situ

Solvents can dissolve in the crude oil in situ and reduce its viscosity. A microbially generated liquid solvent such as butanol would partition between the oil and water phases in the reservoir. The portion of the solvent dissolved in the oil may reduce oil viscosity; therefore, its mobility would increase leading to increased oil recovery. However, as Gray et al. (2008) estimated, achieving of just 5-6% incremental oil recovery may require up to 50% reduction in oil viscosity. The difference in viscosities of the particular oil and butanol or a similar solvent must be taken into account: if this difference is little, then

12 viscosity reduction cannot be significant. Generally, even a small potential gain in oil recovery may require very large volumes of a solvent (Gray et al., 2008).

Apart from the alcohol solvents, microorganisms have been shown to produce hydrocarbons (Park et al., 2005; Frias et al., 2009). Park et al. (2005) reported production of C15-C20 alkanes by Vibrio furnissii, however this result was found to be irreproducible (Wackett et al., 2007). Frias et al. (2009) showed that an Arthrobacter species was able to produce C29 alkanes. Certainly, the last result does not necessarily attest the possibility of the biological production of an efficient hydrocarbon solvent. However, the very possibility of the production of alkanes by microorganisms lets one think that strain(s) consuming high molecular weight hydrocarbons and converting them into hydrocarbons with lower molecular weight, which would dissolve hardly accessible oil in the porous matrix, may be found or created. Creating such strain(s) may also be an approach to producing “alternative” fuels.

1.3. Successful MEOR applications and tests

1.3.1. Conditions of success

A reservoir, a bacterial system, nutrients, and a protocol of well injection are the essential components of any complete MEOR system (Lazar et al.,

2007). Stimulation of indigenous microorganisms through nutrient injections, injection of exogenous microorganisms(s) and nutrients (in situations where there are no suitable indigenous populations), or injection of ex situ produced products (in cases where there are no suitable indigenous microorganisms and exogenously added microorganisms cannot survive the reservoir conditions)

13 have been used as general MEOR approaches (Youssef et al., 2009; Banat et al., 2000).

Stimulation of indigenous microorganisms through nutrient injections requires the presence of such microorganisms, which are able to produce biosurfactants, solvents, acids, gases etc. The presence of the appropriate microorganism or activity can be verified using microbiological and molecular techniques. Application of molecular techniques requires sufficient amount of biomass to extract DNA (Youssef et al., 2007) and detect the presence of specific genes with PCR or qPCR. When biosurfactant production is being studied, then it may be easy to define target genes (e.g. srfA for surfactin, rhlR for rhamnolipid or licA for lichenysin, production). However, in cases where acid, solvent, and gas production are being studied, the choice of the appropriate gene targets may be less clear (Youssef et al., 2009). The analysis of the 16S rRNA content using universal bacterial primers has been proven to be an effective tool for identifying particular members of a microbial community.

To ensure the production of the desired metabolite or activity, systematic amendments of C, N, P etc. sources have been recommended (Harvey and

Harms, 2001; Youssef et al., 2009). According to a number of authors (Youssef et al., 2009; Jenneman et al., 1985; Sharma and McInerney, 1994; Sharma et al., 1993), the addition of electron acceptors (e.g. nitrate), sources of carbon and energy, trace metals, vitamins etc. may stimulate both growth of microorganisms throughout a reservoir and production of surface-active substances. Injections of large amounts of easily biodegradable and cheap carbohydrates (e.g. molasses) could be applied for stimulating acid, solvent, and gas formation as a result of anaerobic fermentation (Youssef et al., 2009).

Interestingly, Guerra-Santos et al. (1986) showed that P. aeruginosa

DSM 2659 produced higher concentration of rhamnolipids under limiting conditions of Mg2+, Ca2+, K+, Na+, Fe2+ ions and trace elements. Indeed, in some cases production of surface-active metabolites increase when microorganisms grow under nutrient-limiting conditions (Cameotra and Makkar,

1998; Sheehy, 1992). This may be the case because surfactants may help pathogens lyse cells of host organisms to get the nutrients. As long as microorganisms become hydrophobic and cell wall components may act as surface-active agents, especially when starvation is induced, periodic cycles of nutrient-excess and nutrient-limitation have been suggested (Youssef et al.,

2009). Lappin-Scott et al. (1988), MacLeod et al. (1988) and Fontes et al.

(1991) found that starved cells, being smaller than non-starved cells, better penetrated porous material. However, conflicting recommendations exist on the use of starved or non-starved cells (e.g., Camper et al., 1993; Cunningham et al. 2007).

Nitrate as an electron acceptor deserves special attention because it is routinely injected in oil reservoirs for souring control (Davidova et al., 2001). The addition of nitrate may lead to enhancement of oil recovery through growth of biomass and stimulation of different microbial processes, such as production of biosurfactants and/or gases and/or biopolymers (Hitzman et al., 2004; Dennis and Hitzman, 2007). Biosurfactant production may be enhanced by the addition of nitrate theoretically if the available source of carbon and energy is water- insoluble. In fact, microorganisms may need to solubilize it (with the produced biosurfactants) to make it consumable. At the same time, some strains, being injected in an oil reservoir, may be capable of biosurfactant production when

15 water-soluble and, thus, easily biodegradable sources of carbon and energy, such as carbohydrates, are available (Youssef et al., 2007).

There is the evidence that, generally, biosurfactant production requires a fine balance between carbon and nitrogen (Youssef et al., 2009). Moreover, several authors showed that rather nitrogen-limiting conditions facilitated biosurfactant production: particularly, it was observed that the synthesis of rhamnose lipids in P. aeruginosa happened after the depletion of a nitrogen source shifting the cells into the stationary growth phase (Guerra-Santos et al.,

1984; Ramana and Karanth, 1989); Syldatk et al. (1985) found the addition of nitrogen sources to inhibit the production of rhamnose lipids by resting cells of

Pseudomonas sp. DSM 2874; Ochsner et al. (1995) found that the genes of P. aeruginosa for rhamnose lipid synthesis were expressed in Pseudomonas fluorescens and Pseudomonas putida only under nitrogen-limiting conditions;

Davis et al. (1999) found that limitation of ammonium nitrate as a source of nitrogen led to the enhancement of surfactin production by Bacillus subtilis; nitrogen limitation was found to cause overproduction of the biosurfactant by

Candida tropicalis, and a C/N ratio of >11 to maximize rhamnolipid production by Pseudomonas sp. (Cameotra and Makkar, 1998; Gautam and Tyagi, 2006).

However, the option with the nitrogen limitation does not seem to be viable under the anaerobic conditions of an oil reservoir. On the contrary, it is reasonable to try to achieve growth-associated biosurfactant production

(Hitzman et al., 2004; Dennis and Hitzman, 2007).

Inoculation of a reservoir with exogenous microorganisms (possibly, genetically engineered ones) may be needed when an indigenous microbial population lacks an appropriate capability (Youssef et al., 2009; Urgun-Demirtas

16 et al., 2006). The exogenous microorganism has to be able to grow under the existing reservoir conditions and survive competition with an indigenous population (Bryant, 1991). Alternatively, one may choose to design a specific nutrient package for preferential stimulation of growth and metabolism of exogenous microorganisms (Youssef et al., 2007). The general problem of this approach may consist in having most MEOR-effects occurred in a near- injection-wellbore region. Thus, the injected microorganisms should be deliverable to the oil reservoir zones where most trapped oil is located. It is noteworthy in this respect that Bae et al. (1996), Gullapalli et al. (2000), Jang et al. (1983) and McInerney et al. (2005b) showed feasibility of the use of spores for transporting microorganisms throughout a reservoir. In the experiment conducted by Youssef et al. (2007), Bacillus strain RS-1 and Bacillus subtilis subsp. spizizenii NRRL B-23049 were injected, together with glucose, sodium nitrate and trace metals, into the first group of oil production wells; only glucose, sodium nitrate, and trace metals – in the second group; and only formation water in the third group of wells. It was shown that, in the presence of Bacillus strain RS-1 and Bacillus subtilis subsp. spizizenii NRRL B-23049 (in the first group of wells), large amounts of glucose were consumed and biosurfactant was produced in a concentration sufficient for mobilization of trapped oil from sandstone cores (Youssef et al., 2007).

Partial loss of nutrients and target products (e.g. biosurfactants) due to adsorption to rock material is one of the important issues to consider when

MEOR approaches are applied. Using modified forms of nutrients (e.g. organophosphates instead of inorganic phosphates; Jenneman and Clark,

1994a; Jenneman and Clark, 1994b) can help in handling this issue.

It is, in principle, possible to transform the indigenous population with the use of gene delivery vehicles (Seow and Wood, 2009), e.g. viruses, to let microorganisms overproduce biosurfactants and/or acids and/or solvents etc.

Regulatory issues and negative public response related to the spreading of genetically modified organisms in the environment may prevent field applications of this method (Urgun-Demirtas et al., 2006). In addition, not much research has been conducted on MEOR potential of genetic modification of indigenous microbial populations.

An addition of substances produced by microorganisms ex situ (e.g. biosurfactants) may be considered if indigenous microorganisms are not suitable for the desired outcome and/or conditions in the reservoir are too harsh for survival of exogenous microorganisms.

1.3.2. Overview of some significant achievements in the history of MEOR

The idea of using microorganisms for release of oil from porous media was first suggested by Beckman (1926). In the 1940s, ZoBell (1947) and his research group conducted systematic laboratory experiments and described release of oil from sand packed columns by bacterial products such as gases, acids, solvents, surface-active agents, and cell biomass; the corresponding processes have been patented. According to ZoBell (1947), dissolution of inorganic carbonates by bacterial metabolites, production of bacterial gases, decreasing the viscosity of oil (and thereby promoting its flow), production of surface-active or wetting substances by some bacteria, and high affinity of bacteria to solids are the main factors of microorganism-mediated oil recovery.

According to Lazar et al. (2007), the first MEOR field test was carried out in the Lisbon field, Union County, AR, USA, in 1954. In 1957, facilitation of oil recovery through the use of microorganisms able to convert inexpensive substrates such as molasses into gases, acids, solvents, and biosurfactants was patented (Updegraff, 1957). In 1958, von Heinningen et al. suggested a new technology, selective plugging recovery, based on producing polysaccharide slime by the injected microorganisms in situ with the molasses added. Dostalek and Spurny (1958), Kuznetsov et al. (1963), Senyukov et al.

(1970), Karaskiewicz (1974), Lazar (1978), and Ivanov et al. (1983) conducted experiments leading to a number of field trials based on the injection of mixed anaerobic or facultative anaerobic bacteria (such as Clostridium, Bacillus,

Pseudomonas, Micrococcus, Mycobacterium etc.) able to generate significant amounts of gases, acids, solvents, polymers, surfactants, and biomass.

Hitzman (1988) and Lazar (1991) described the possibilities of using viscosifying biopolymers, such as xanthan and scleroglucan in EOR.

It was shown in a number of experiments that production of acids, solvents and gases in situ led to increases in oil recovery (Almeida et al., 2004;

Behlulgil and Mehmetoglu, 2002; Bryant, 1988; Bryant and Burchfield, 1989;

Bryant and Douglas, 1988; Chang, 1987; Desouky et al., 1996; Jinfeng et al.,

2005; Maudgalya et al., 2004; Rauf et al., 2003; Wagner, 1985; Wagner et al.,

1995). Heterochthonous microorganisms and molasses or some other types of easily-fermentable carbohydrates were used in most of these studies.

Brown (1992) profitably applied MEOR technology in a large-scale field trial on 146 wells. According to Zhang et al. (1999), Brown’s project was the first to apply MEOR technology profitably on a commercial scale.

Belyaev and Ivanov (1990) and Ivanov et al. (1993) reported successful application of the method based on stimulation of indigenous microbial communities by introducing oxygen and salts with water injection. According to

Wang (1991) application of biopolymers produced by several species, namely,

Leuconostoc mesenteroides, Pseudomonas aeruginosa, Brevibacterium viscogenes, Corynebacterium gumiform, and Xanthomonas campestris in

Chinese oil fields was successful as well. Wagner et al. (1993) enhanced oil production from a carbonate reservoir via injecting inocula of Clostridia species and molasses; Lazar et al. (2007) mentioned some MEOR applications in

Tatarstan oil fields in Russia, which relied on the experience of Wagner et al.

(1993). Lazar (1996) successfully applied single-well stimulation and microbial flooding recovery technologies involving injections of mixed enrichment cultures and molasses into several Romanian oil reservoirs. Jack (1993) and Lazar et al.

(2007) described plugging of oil reservoir high permeability zones with

Leuconostoc and some other bacteria.

Thus, with about a century of research and field trials, MEOR in situ has been shown to be an efficient technology in many cases. It has some advantages compared to other EOR methods. However, there are drawbacks as well. They will be overviewed in the next section.

1.4. Major problems of MEOR

According to Gray et al. (2008), adsorption to the mineral surfaces is “the

Achilles heel” of both chemical and biological surfactants in reservoir applications. For this reason, mobilization of residual crude oil with surfactants requires their presence in large amounts; thus the treatment may be either

20 worthless (if bacteria are unable to produce the required amounts of biosurfactant) or prohibitively expensive. Gray et al. (2008) estimated adsorption values for sandstones to be in the range of 0.1 to 1 mg of surfactant per gram of rock; if one assumes residual oil saturation to be 30% and the incremental oil recovery – 15%, then 23,000 m3 of incremental oil would be recovered per required 204,000 kg of surfactant (Gray et al., 2008).

Sustained injection of nutrients required for biosurfactant production by an introduced/exogenous strain may stimulate the growth of competing indigenous organisms, which can consume nutrients (without producing a biosurfactant), or even consume the biosurfactant (Gray et al., 2008; Voordouw,

2011).

Thus, conditions of microbial growth in an oil reservoir often do not allow production of target metabolites in the amounts needed for microbial enhancement of water-flood-driven oil recovery. Due to the disadvantages of

MEOR, its usefulness has been slowly recognized by the industry (Lazar et al.,

2007). To make MEOR really highly attractive for industrial applications, particularly in its combination with a water flood, it is needed that microorganisms, either indigenous to an oil reservoir or introduced to it, substantially grow in number and/or produce really large amounts of MEOR- active metabolites. It is also, in principle, possible to use microbial activity in a non-water-flood-driven oil recovery context, namely, in combination with solution gas drive, so that small amounts of microbial metabolites and/or cells may suffice for enhancement of oil recovery. The issues associated with the idea of combining MEOR-technology with solution gas drive will be discussed in the next section.

1.5. Microbial enhancement of solution-gas-driven oil recovery

1.5.1. The principles of solution-gas-driven oil recovery

Being the mechanism of primary oil recovery, solution gas drive is also often used at the stage of EOR (U.S. Department of Energy, 2011). To artificially create solution gas drive, a gas (for example, CO2, CH4 or N2) is injected into a reservoir at high pressure; the higher pressure, the more gas will dissolve in oil and water. Dissolved gas will promote swelling of oil and water; the volume of oil will expand to a greater extent than that of water due to higher solubility of gases in the former than in the latter. Gas dissolution will also reduce oil viscosity, thus leading to increase of its mobility (Holm and Josendal,

1974) and creating beneficial conditions for subsequent oil recovery.

When pressure is decreased, the gas will come out of solution, and liquids (oil and water) will be produced (Holm and Josendal, 1974). High rates of heavy oil recovery were observed during this process. However, many of its aspects are still poorly understood (Pooladi-Darvish and Firoozabadi, 1999;

Shahvali and Pooladi-Darvish, 2009).

According to Smith (1988), when solution gas is released, micro-bubbles form and move to a production well together with a continuous oil phase.

Increase of pressure drop across the core accompanies bubble formation (Tang and Firoozabadi, 1999).

Simultaneous flow of bubbles and oil with oil represented as droplets rather than a continuous phase may occur as well. Indeed, oil droplets trapped in a porous medium, upon being released due to swelling (as a result of gas dissolution in them) and/or due to pressure from gas bubbles in an aqueous phase, may spread over and be carried by the bubbles in the way described by

Moosai and Dawe (2003). Lubrication effects caused by nucleation of bubbles at the pore walls of a porous medium have been thought to enhance oil mobility and, thus, contribute to oil production as well (Shen and Batycky, 1996).

Contrary to the idea of simultaneous flow of micro-bubbles and oil suggested by Smith (1988), Firoozabadi and Aronson (1999) found that only oil

(and no gas) was produced before critical gas saturation was achieved, despite the fact that bubbles were formed. According to Alshmakhy and Maini (2009), critical gas saturation is the minimum gas saturation that allows gas to flow as a continuous phase; alternatively, critical gas saturation may be defined as the minimum gas saturation, at which gas flow occurs in the two-phase gas-oil flow regime (Firoozabadi, 2001). Before critical gas saturation is achieved, production of oil may occur just because of the expansion of gas in the form of immobile bubbles. Firoozabadi and Aronson (1999) found oil to be produced along with intermittent portions (“slugs”) of gas when critical gas saturation is achieved.

High efficiency of solution gas drive for heavy oils might be due to low gas mobility and relative permeability; high bubble density (number of bubbles per unit area or volume) might contribute as well. Low bubble density and high gas mobility could result in low efficiency of solution gas drive in very light oil systems (Firoozabadi and Aronson, 1999). Heavy oil was found to be produced much more efficiently than light oil also during special flow behavior called foamy oil flow observed by Maini et al. (1993) at certain operating conditions

(particularly at pressure gradients high enough to mobilize gas bubbles grown to a certain size). Low gas mobility is believed to be mainly due to high oil viscosity (Firoozabadi, 2001). According to Talabi et al. (2003), relative

23 permeability to gas is a function of both oil viscosity and critical gas saturation, the latter being dependent on pressure depletion rate. Simulation experiments conducted by Ostos and Maini (2004) showed that oil relative permeability increased and gas relative permeability decreased with the increase of capillary number.

Counter-diffusion-induced gas drive (Graves et al., 1972; Graves et al.,

1973), which can be considered a specific type of solution gas drive, can be used in the laboratory conditions where direct pressurization of cores or oil reservoir model columns is not possible or if simultaneous testing of numerous samples is required. Gas drive in a liquid medium can be induced by counter- diffusion when a gas having lower diffusivity in a liquid (Gas 1) is dissolved in it, while the liquid is surrounded by a gas with higher diffusivity in it (Gas 2). Due to difference in diffusivities, the rate of the influx of Gas 2 into a liquid phase will exceed the rate of the efflux of Gas 1 from a liquid phase to a gas phase, leading to temporary supersaturation of the liquid phase with gas (the amounts of dissolved gases will exceed those corresponding to equilibrium at the ambient pressure), under-pressurization of the gas phase, and formation of mobile bubbles in the liquid phase (Graves et al., 1972; Graves et al., 1973). It may be possible to construct a column containing oil and water, from which liquids would be produced under counter-diffusion-induced gas drive.

1.5.2. Enhancement of solution-gas-driven oil recovery with surfactants in high and low concentrations

Let’s consider possible consequences of adding surfactants in concentrations below and above critical micelle concentration (CMC) to a core,

oil

water

gas

Figure 1A. Meeting of a gas bubble with an oil droplet at a high surfactant concentration; the bubble and the droplet are tightly covered with surfactant molecules (hydrophilic parts of surfactant molecules are shown as red circles; hydrophobic “tails” can be seen in the bubble as well). The bubble and the droplet cannot get in touch with each other because of electrostatic repulsion between surfactant hydrophilic “heads”. (However, repulsive forces can be weakened through increasing of the ionic strength of a solution).

Figure 1B. Meeting of a gas bubble with an oil droplet when surfactant is present in a low concentration. Electrostatic repulsion between surfactant hydrophilic “heads” is not sufficient for preventing the bubble and the droplet getting in touch with each other; thus, the droplet can eventually spread over the bubble, while surfactant can still reduce an oil-water IFT.

25 from which trapped heavy oil is produced due to solution gas drive with mobile bubbles (Smith 1988) or intermittent gas flow (Firoozabadi and Aronson, 1999;

Firoozabadi, 2001). CMC can be defined as the concentration, above which most newly added surfactant molecules form micellar aggregates (Ruckenstein and Nagarajan, 1975); as a result, inceasing surfactant concentration above

CMC cannot lead to substantial IFT changes.

In the case of mobile bubbles, if oil droplets are capable of spreading over the bubbles (Moosai and Dawe, 2003), then after adding a surfactant, its hydrophobic parts will be directed into the gas phase of a bubble and into the oil phase of a droplet; at the same time, hydrophilic parts of a surfactant will be on the surfaces of a bubble and a droplet. If a surfactant concentration is high

(near and above CMC), these surfaces will be tightly covered with the hydrophilic parts of a surfactant. Thus, strong electrostatic repulsion will exist between the surfaces of a bubble and a droplet, and a droplet could not spread over a bubble (Figure 1A). Therefore, if solution gas drive leads to production of any liquids, then heavy oil would be unlikely to be produced and mostly water, having higher mobility than that of oil, will be produced. If a surfactant concentration is low (below CMC), bubble and droplet surfaces will not be tightly covered with a surfactant. Thus, despite weak electrostatic repulsion between a bubble and a droplet, the latter will still be able to cover the former, and a surfactant present on the bubble-droplet surface (Figure 1B) will decrease an oil-water IFT and, thus, facilitate movement of the bubble-droplet to a production well. Therefore, the presence of surfactants in low concentrations may be beneficial for solution-gas-driven oil recovery, while the presence of the same surfactants in high concentrations may be detrimental for this process.

If a surfactant is added in a high concentration, then one could also observe foam formation. Presence of foam would lead to decrease of permeability of a porous matrix to gas and, possibly, to improvement of gas-to- oil mobility ratio (Farajzadeh et al., 2008). Foams are also known to be capable of reducing gas mobility to zero (Bernard and Holm, 1964). If it equals zero, then oil and water may be displaced because of swelling and/or gas expansion, with water likely being preferentially produced due to having higher mobility than that of heavy oil. In the case with quazi-continuous, intermittent gas flow, IFT reduction and foam formation could be observed as well if a surfactant was added in a high concentration. If a surfactant is added in a low concentration, then oil-water IFT would be reduced (possibly to a lesser extent than in the case with a high surfactant concentration), but foam formation would be unlikely.

Without foam, oil production may be less efficient than that with foam in the case if original gas mobility is high: indeed, if foam is absent, then a process known as viscous fingering (Homsy, 1987) may be dominating. However, if original gas mobility is low, then foam formation may reduce gas mobility to zero, as it was mentioned above, and make oil production more difficult; in this case, adding a surfactant in a low concentration would be required.

Surfactant adsorption to a rock surface must also be taken into consideration. The charge a given rock carries is the main determinant of the interaction of its surface with ionic surfactants (Zhang and Somasundaran,

2006). For example, it is known that silica is negatively charged at reservoir conditions (Paria and Khilar, 2004; Hirasaki et al., 2011), and large amounts of cationic surfactants would adsorb onto its surface (Paria and Khilar, 2004).

Despite the fact that an anionic surfactant carries the same charge as the silica

27 surface, its adsorption may occur as well, although its adsorbed amount would be much smaller than that of a cationic surfactant. The adsorbed amount of a non-ionic surfactant would be comparable to that of an anionic surfactant (Paria and Khilar, 2004). Let’s consider a situation with a cationic surfactant and a negatively charged porous matrix. If it is added in a low concentration, then its adsorption would lead to the formation of a surfactant monolayer with hydrophobic moieties directed to an aqueous phase, thus rendering the matrix surface more hydrophobic/oil-wet (Rao et al., 2006). However, with the concentration increase, a bilayer may form, with hydrophobic moieties in the midst of the bilayer and hydrophilic moieties facing either rock/sand or a water phase. At this stage, the rock/sand surface would be rather hydrophilic due to an adsorbed bilayer, and the charge of the surface would become rather positive; with the further concentration increase, micellar aggregates would form on the rock/sand surface (Bi et al., 2004). Therefore, the addition of a surfactant charged oppositely to the charge of a porous matrix may lead either to more oil getting attached to sand (making its displacement from a column more difficult) or to no significant wettability changes. If adding a cationic surfactant leads to an unfavourable wettability change, then this change may overshadow the oil- water IFT reduction, so that the cumulative effect of its addition may be detrimental for solution-gas-driven oil recovery associated with either mobile bubble formation or quazi-continuous, intermittent gas flow. Anionic and non- ionic surfactants, being much less prone to adsorption to the silica surface, may not be capable of rendering it strongly oil-wet.

Chapter 2: Research objectives and hypotheses

2.1. Research objectives

1. To construct a column filled with silica sand to model an oil reservoir for conducting experiments on MEOR involving application of water flood and counter-diffusion-induced gas drive in a way that allows having a number of replicates.

2. To find out if stimulation of indigenous oil field microorganisms with nitrate leads to the enhancement of water-flood-driven and gas-driven recovery of heavy oil trapped in model columns.

3. To determine the compositions of microbial communities associated with the aqueous and oil phases of the produced water containing heavy oil.

4. To construct a recombinant E. coli strain carrying rhlAB genes responsible for production of rhamnolipids; find out if rhamnolipids are produced and test if an addition of the supernatant isolated from the culture of this strain to a model column leads to enhancement of water-flood-driven and gas-driven recovery of trapped heavy oil.

5. To compare the EOR-efficiency of the supernatant isolated from a culture of the recombinant E. coli strain to the EOR-efficiency of the supernatant isolated from a culture of a native rhamnolipid producer and to the EOR- efficiencies of the solutions of some chemical surfactants (ionic and non-ionic) under water flood and under counter-diffusion-induced gas drive.

2.2. Hypotheses

1. Stimulation of indigenous oil field microflora with nitrate will not lead to enhancement of water-flood-driven recovery of heavy oil trapped in model

29 columns because of the following: (a) numbers and sizes of microorganisms grown with nitrate will not be sufficient to plug high permeability zones; and (b) the influx of water will lead to diluting biosurfactants possibly produced by indigenous microflora during the time of incubation with nitrate, so that oil-water

IFT will not be sufficiently reduced.

2. Stimulation of indigenous oil field microflora with nitrate will lead to enhancement of counter-diffusion-induced gas-driven recovery of heavy oil trapped in model columns because, while gas flow will be sufficiently slow, microorganisms attached to oil will grow in number and likely produce biosurfactants in small amounts, which will still suffice for oil-water IFT reduction.

3. Primary hydrocarbon degraders will reside predominantly in the oil phase/oil-water interface of the oil-field produced water containing heavy oil, and, thus, they will most likely be responsible for MEOR-effects of nitrate injections (if any are found), while consumers of water-soluble substrates will be found mostly in the aqueous phase of the same oil-field produced water with heavy oil.

4. EOR-effects of supernatants from cultures of rhamnolipid-producers applied in combination with water flood will be similar to EOR-effects of solutions of chemical surfactants also applied in combination with water flood.

5. A supernatant from a culture of a native rhamnolipid-producer and solutions of chemical surfactants will not help enhance oil recovery under counter-diffusion-induced gas drive because (bio-)surfactants taken in relatively high concentrations will either (a) prevent spreading of oil droplets over gas

30 bubbles due to electrostatic repulsion between them or (b) stop gas flow through foam formation.

6. Cell-containing supernatants from cultures of strains capable of producing small amounts of biosurfactants will help enhance oil recovery under counter-diffusion-induced gas drive because biosurfactants taken in low concentrations, while not preventing spreading of oil droplets over gas bubbles and forming little or no foam, will sufficiently reduce oil-water IFT; presence of rhamnolipids may help render cell surfaces hydrophobic; being attached to an oil phase, the cells may contribute to oil-water IFT reduction.

Chapter 3: MEOR effects of nitrate injections

in experiments with miniature model columns

3.1. Introduction

Nitrate is often injected into oil reservoirs to control souring (Davidova et al., 2001; Gieg et al., 2011; Voordouw, 2011). Serving as an electron acceptor for heterotrophic nitrate reducing bacteria (hNRB), it is converted into nitrite, which prevents sulphide production through inhibition of the activity of sulphate reducing bacteria (SRB). In addition, sulfide-oxidizing NRB (soNRB) can directly oxidize sulphide present in an oil reservoir (Voordouw, 2011).

Results of some field applications and laboratory experiments provide evidence that nitrate injection may also help enhance oil recovery (Dennis and

Hitzman 2007; Hitzman and Sperl 1994). For example, a 200% increase in oil production with a 10% decrease in water cut was achieved in a Southern

Saskatchewan oil field, reportedly due to injection of 1–3 m3 of a solution of nitrate and nutrients, with a subsequent one-week cessation of water flood

(Town et al., 2009; Voordouw, 2011). Youssef et al. (2007) also achieved additional oil production when they applied nitrate injection; in their experiment, it was combined with the introduction of biosurfactant producing bacteria into the oil reservoir, together with carbohydrates (molasses) as carbon and energy sources for these bacteria. Similarly, Jackson et al. (2010) successfully tested some microbial strains and nutrient packages containing nitrate to improve oil production.

However, injection of microbial strains exogenous to oil reservoirs may not always be needed, as co-injected nutrient packages would stimulate growth

32 of indigenous microflora as well (Voordouw, 2011). Thus, instead of exogenous strains, native microbial populations may be relied on, provided organisms with necessary characteristics are present in situ (Sugihardjo et al., 1999; Voordouw,

2011). Some recent findings support the idea about the viability of this approach. In particular, Simpson et al. (2011) demonstrated the presence of biosurfactant-producing Bacilli in Oklahoma oil reservoirs. In Chapter 4 it is shown that some microbial genera native to an oil field in Western Canada are strongly attached to oil. If microbes producing biosurfactants and/or attached to oil are available in an oil reservoir, then it is, indeed, reasonable to expect that an addition of substances stimulating their activity would lead to increased growth of oil-attached biomass and production of biosurfactants required for enhancement of oil recovery (Kowalewski et al., 2006; Kang et al., 2008a).

Other possible MEOR-effects of the stimulated oil-attached bacteria may be due to localized (exactly at the oil-water interface) production of exopolysaccharides, alcohols and/or acids (Youssef et al., 2009; Voordouw, 2011), although the efficiency of industrial-scale applications relying on production of such substances has been questioned (Gray et al., 2008).

If one expects to achieve oil-water IFT reduction due to localized production of biosurfactants and adhesion of relatively large numbers of microbes to oil, then it is important that they use oil components as carbon and energy sources. Otherwise, it is possible that originally hydrophobic bacterial cell surfaces become hydrophilic (in a way similar to that described by Obuekwe et al., 2008) to make consumption of carbohydrates or other hydrophilic carbon and energy sources easier when they become available. However, feeding

33 indigenous microflora with carbohydrates may be beneficial if one tries to achieve plugging of high permeability zones with biomass (Voordouw 2011).

In this chapter, results of the monitoring of MEOR-effects caused by stimulation of indigenous microbial communities with nitrate using miniature model columns under counter-diffusion-induced gas drive and water flood will be discussed.

3.2. Materials and methods

3.2.1. Oil samples

Experiments were conducted with two samples of heavy oil. The first oil sample, EO, from the MHGC (Medicine Hat Glauconitic C) field (near Medicine

Hat, AB, Canada; also known as the Enermark field), had an API gravity of 16º and a viscosity of 3400 cp at 20ºC, 340 cp at 50ºC and 70 cp at 80ºC. The second oil sample, LO, from the Loma Alta Sur (LAS) field (Argentina), had an

API gravity of 21º and a viscosity of 30 cp at 45ºC (at reservoir conditions). Both samples of oil contained small amounts of water.

3.2.2. Enrichment cultures with EO or LO as sources of carbon and energy

Four 120-mL serum bottles with 47.5 mL of sterile anaerobic CSB-K medium (2.75 g/L NaCl, 0.045 g/L KH2PO4, 0.54 g/L MgCl2•6H2O, 0.21 g/L

CaCl2•2H2O, 0.2 g/L NH4Cl, 2.52 g/L NaHCO3, 3 drops of resazurin, 1 mL stock solution of trace elements, 1 mL tungstate-selenite stock solution, pH 7.5;

Lambo et al., 2008) containing 20 mM (1.7 g/L) NaNO3 were prepared to establish enrichment cultures with EO or LO as sources of carbon and energy to use in further experiments with syringe sandpack columns (section 3.2.7).

Two bottles received 2.5 mL of produced water from the MHGC oil field (as a source of microorganisms) and 1 mL of EO each; two others received 2.5 mL of produced water from the LAS field and 1 mL LO each. The bottles were closed with butyl rubber stoppers and the headspace was filled with 90% (v/v) N2 and

10% CO2. They were incubated at 30ºC. Nitrate and nitrite concentrations were periodically measured through HPLC-analysis as described elsewhere (Lambo et al., 2008). A high-pressure liquid chromatograph equipped with an IC-Pak A

HC anion exchange column (Waters Corp., Milford, MA) was used. Isocratic elution of a 100 μL sample at 2.2 mL/min was achieved with the sodium borate- gluconate buffer in accordance with instructions provided by the column manufacturer. Nitrate was completely reduced in all bottles within 30-40 days; upon its complete reduction, it was aseptically added to restore its initial concentration of 20 mM to maintain growth of enrichment cultures.

3.2.3. Column assembly

Barrels of 30 or 20 mL polypropylene syringes (Becton, Dickinson and

Co. or Henke Sass Wolf) were used for making the columns (Figure 2). Barrels of 30 mL syringes had a length of 9.6 cm and an inner diameter of 2 cm; barrels of 20 mL syringes had a length of 8.5 cm and an inner diameter of 1.8 cm.

Glass wool (0.06 – 0.08 g) was placed at the bottom of the barrel, which was filled with 53.5 – 55 g (for a 30 mL column) or 34.5 – 35.5 g (for a 20 mL column) of 50 – 70 mesh quartz sand (Sigma-Aldrich). During this process, syringe barrels were manually shaken to achieve compact settling of the sand.

Figure 2. A partially-assembled 20 mL syringe sandpack column.

A polymeric mesh (Scotch-Brite, 3M) was placed on the top of the sand and the column was closed tightly with a size 4 (for a 30 mL column) or a size 3 (for a

20 mL column) rubber stopper (VWR, catalog numbers 59580-160 and 59580-

149). The pore volumes were defined through weighing columns saturated with synthetic produced water (SPW) vs. dry columns. The average pore volume

(PV) of a 30 mL column was 13.38±0.28 mL; and the average PV of a 20 ml column was 9.12±0.41 mL. The porosity of both 30 mL and 20 mL columns was

40.5±0.5%. Difference between injection and atmospheric pressure ΔP

(pressure drop) was monitored with a WIKA pressure gauge, type 232.53 (0 –

30 psi); the absolute permeability of the both 30 mL and 20 mL columns was found to be 30 D.

3.2.4. Saturating columns with synthetic produced water (SPW)

SPW for the experiments with EO had pH 7.5 and contained 90 mM

NaCl; 1 mM KCl; 1 mM MgCl2; 2 mM CaCl2; 10 mM NaHCO3. SPW for the experiments with LO had pH 7.8 and contained 2 mM Na2SO4; 90 mM NaCl; 1 mM KCl; 1 mM MgCl2; 2 mM CaCl2; and 10 mM NaHCO3. SPW was injected in dry columns manually with a syringe in aerobic conditions.

3.2.5. Saturating columns with oil

EO and LO were injected into vertically placed water-saturated columns with a Mandel-Gilson Minipulse 3 peristaltic pump in aerobic conditions. Upward flow was always used. Residual water saturation (Swr) was measured through weighing collected eluted water (vs. the amount of water originally present in the column). Average Swr was found to be 1.08±0.40 mL (8.1% of PV) for a 30

37 mL column with either EO or LO and 0.73±0.06 mL (8% of PV) for a 20 mL column with EO (20 mL columns were not used in the experiments with LO).

Oil-saturated columns were aged at 20ºC for seven to ten days to create wettability conditions maximally similar to those in an oil reservoir.

3.2.6. Eluting oil with SPW

Elution was conducted with SPW using the Mandel-Gilson Minipulse 3 peristaltic pump (Figures 3A and 3B) in aerobic conditions at a rate of 6.5 mL per hour until essentially no more oil was produced. Curves reflecting SPW- flood-driven production of oil and total liquids (oil and water) from EO- and LO- saturated 30-mL columns are shown in Figures 5 and 6. (Profiles of oil production from 20 mL columns were essentially the same). Following a brief period of producing only oil, water breakthrough was observed and a water-oil mixture was produced (Figure 4). About 0.3 PV of EO and about 0.2 PV of LO still remaining in a column by the time when essentially no more oil was being produced with the water flood (Figures 5 and 6) was considered trapped oil.

Volumes of trapped oil for all columns are shown in Table 1.

3.2.7. Experiments with columns containing trapped oil

Four sets of columns (sets I–III and V) with trapped EO (sets I-III consisting of three columns and set V consisting of two columns) and one set of columns with trapped LO (set IV consisting of five columns) were used for the experiments.

Figure 3. (A) Water-flooding of a 30 mL syringe sandpack column containing EO using the Mandel-Gilson Minipulse 3 peristaltic pump. (B) A close-up of a water-flooded 30 mL syringe sandpack column with an attached 50 mL polypropylene tube for collecting the effluent.

0.18 4.5 0.16 4 0.14 3.5

0.12 3 Vw/PV

0.1 2.5

0.08 2

Vo/PV 0.06 1.5 Vo/PV 0.04 1 Vw/PV 0.02 0.5 0 0 0 100 200 300 400 Time, min

Figure 4. EO production from a 20 mL column flooded with SPW at a flow rate of 6.5 mL/hour during several hours in the beginning of the flood.

0.8 45

0.7 40

0.6 35 30 0.5

25 Vw/PV 0.4

Vo/PV 20 0.3 15 0.2 10 Vo/PV 0.1 Vw/PV 5 0 0 0 10 20 30 40 50 Vtotal/PV

Figure 5. EO production from 30 mL columns (3 replicates) when flooded with SPW; flow rate – 6.5 mL/hour. Cumulative pore volumes of produced oil (Vo/PV) and water (Vw/PV) are plotted against total pore volumes of produced liquids (Vtotal/PV). Bars indicate standard deviations.

0.9 45 0.8 40 0.7 35 0.6 30

0.5 25 Vw/PV

Vo/PV 0.4 20 0.3 15 0.2 10 Vo/PV 0.1 5 Vw/PV 0 0 0 10 20 30 40 50 Vtotal/PV

Figure 6. LO production from 30 mL columns (3 replicates) when flooded with SPW; flow rate – 6.5 mL/h. Bars indicate standard deviations.

Set V columns were used to analyze the composition of EO remaining in a column after an SPW flood. When essentially no more oil was produced with the SPW flood, set V columns were disassembled and oil was extracted from the sand and walls of each column through washing the walls with 10-20 mL dichloromethane (DCM) and washing the sand placed in 250 mL polypropylene bottles (VWR) with 30-40 mL DCM. Then oil composition was analyzed as described below (section 3.2.9).

The following experiments with columns from sets I–IV were performed in an anaerobic hood with an atmosphere of 90% N2 and 10% CO2. At the beginning of the first long-time incubation round, each of the columns I-1, II-1 and III-1 received 1 PV of a 15:85 (v/v) mixture of nitrate-reducing enrichment culture (originating from the same field as EO) and CSB-K medium (section

3.2.2; Lambo et al., 2008) containing 20 mM KNO3; each of the columns I-2, II-2 and III-2 received 1 PV of CSB-K medium containing 20 mM KNO3; and each of the columns I-3, II-3 and III-3 received 1 PV of CSB-K medium without nitrate

(Table A1; Figures A1 and A2). Columns from set IV received one of the following solutions at the beginning of the first long-time incubation round: column IV-1 received 1 PV of 20 mM KNO3 in CSB-K medium; column IV-2 – 1

PV of CSB-K medium without nitrate; column IV-3 – 1 PV of 15:85 (v/v) mixture of nitrate-reducing enrichment culture (from the same field as LO) and CSB-K medium containing 90 mM KNO3; column IV-4 – 1 PV of 0.5 M KNO3 in CSB-K medium; and column IV-5 – 1 PV of 15:85 (v/v) mixture of nitrate-reducing enrichment culture (originating from the same field as LO) and CSB-K medium containing 900 mM KNO3 (Table A1; Figure A3). Enrichment cultures were injected into columns I-1, II-1, III-1, IV-3 and IV-5 before the first incubation

42 round, but not before subsequent rounds. All injections were conducted manually using 20 mL or 30 mL syringes. After the injections, crimp-sealed 10 mL vials filled with helium or argon were attached to all columns for creating counter-diffusion induced gas drive and collecting produced liquids and gases.

Then columns were placed vertically, with the effluent port on the top, in an incubator at 30°C (the incubator was located in the anaerobic hood) to study the effects of long-time (100- to 400-day) incubations.

Vials were periodically removed for measuring amounts of produced oil, water and gas concentrations, and 1 PV of the corresponding media (see this section above; no enrichments were added at this stage) was re-injected into each column (Tables A1, A2-1, A2-2 and A2-3). During re-injections, the initial 1

– 6 mL of effluents from columns were collected in microcentrifuge tubes for measuring concentrations of nitrate and nitrite using HPLC as described in section 3.2.2. After media re-injections, new 10 mL vials filled with helium or argon were attached to columns. These procedures were performed without removing the columns from the anaerobic hood.

After several rounds of incubations, column sets I and II were removed from the anaerobic hood and subjected to water floods (5 PV of SPW were flooded in each case) at a flow rate of 6.5 mL/hour at 20ºC and 30ºC, and the amounts of produced oil and water were determined as described below

(section 3.2.8). In the case of the column set II, a N2/CO2 (90%/10%) gas flood

(with pressure differences from 20 to 70 mbar) was also applied in two modes – simultaneously with and separately from the SPW flood (6.5 mL/hour) at 30ºC to model the water alternating gas (WAG) process (Kulkarni, 2003).

Upon completion of experiments, set I – IV columns were disassembled and sand with trapped oil was removed from them. DCM in the amount of 50 mL per column was used to extract oil from sand and inner column walls to determine compositions and amounts of oil remaining in columns.

3.2.8. Determination of the amounts of produced oil and water

A method similar to that suggested by McGill and Rowell (1980) was used for determination of the volumes of produced oil. The volumes of produced and remaining oil were determined through diluting oil fractions/extracts with

DCM (Figure 7A) and measuring the absorbance of the oil-DCM solution at 600 nm (A600) using a Mandel Pharma-Spec UV-1700 spectrophotometer. The volumes of the oil-in-DCM solution and water were determined using a measuring cylinder. Standard curves were used for calculating the volumes of

EO or LO (the curve used for the determination of the volumes of EO is shown in Figure 7B).

3.2.9. Oil composition analysis

Aliquots of oil-in-DCM solutions obtained from columns after their disassembly were taken to analyze the oil composition using gas chromatography – mass spectrometry (GC-MS). Samples of original EO and LO were diluted 21-fold in DCM and analyzed as well. GC-MS analysis was conducted as described elsewhere (Gieg et al., 2008; Agrawal et al., 2012). An oil-in-DCM solution was injected in the amount of 1 μL by an autoinjector

(7683B series, Agilent Technologies) into a gas chromatograph (7890N series,

Agilent Technologies), which was connected to a mass-selective detector

Figure 7A. Addition of DCM to a fraction of the produced oil and water.

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 absorbance at 600 600 nm at absorbance 0.1 y = 0.4493x + 0.0088 R² = 0.9995 0 0 0.5 1 1.5 2 2.5 concentration of oil, mL/mL, x10-3

Figure 7B. Absorbance (A600) of EO in DCM: a standard curve.

(5975C inert XL MSD series, Agilent Technologies). The gas chromatograph was equipped with an HP-1 fused silica capillary column (length 50 m, inner diameter 0.32 mm, film thickness 0.52 μm; J&W Scientific). Helium was used as the carrier gas. Peaks of interest were identified using Wiley registry of mass- spectral data; standard mass-spectra for some compounds (ethylbenzene, n- heptane, toluene, m-xylene and o-xylene) were used as well.

3.2.10. Determination of concentrations of gases produced during long- time incubations

Concentrations of gases (N2, CO2, CH4 and H2) produced into 10 mL vials during long-time incubations were measured with a gas chromatograph

(Agilent Technologies 7890A) equipped with HP-Plot/Q and HP-Molesieve columns (Agilent Technologies) and a thermal conductivity detector. Helium was used as a carrier gas for detection of N2, CO2 and CH4, and nitrogen was used as a carrier gas for detection of H2. Peak areas were converted into concentrations (mmol/L in a headspace) using standard curves.

3.2.11. Enrichment cultures with heptane as a sole source of carbon and energy

A set of eight 120-mL serum bottles with 47.5 mL of sterile CSB-K medium (section 3.2.2) and 2.5 mL of produced water from the MHGC oil field

(as a source of microorganisms) in each bottle was prepared to establish microcosms to test if oil field NRB can consume heptane as a sole source of carbon and energy. The bottles were closed with butyl rubber stoppers and the headspace was filled with 90% (v/v) N2 and 10% CO2. Heptane was purchased

46 from Sigma-Aldrich. Microcosms #1 and #2 contained 20 mM nitrate (1 mmol per 50 mL) and 22 µL of heptane (0.15 mmol); microcosms #3 and #4 contained 20 mM nitrate and no heptane; microcosm #5 had 50 mM nitrate and

55 µL of heptane; microcosm #6 contained 50 mM nitrate and no heptane; microcosm #7 had 80 mM nitrate and 100 µL of heptane; and microcosm #8 had 80 mM nitrate and no heptane. The added amounts of heptane were in excess over the amounts of available nitrate (the oxidation of one molecule of heptane to CO2 requires nine molecules of nitrate to be reduced to N2). The bottles were incubated at 30ºC, and nitrate and nitrite concentrations were periodically measured as described above (section 3.2.2). Additionally, the ninth microcosm, #0, was established in a 250-mL bottle without the oil field produced water. This bottle was filled with 100 mL of non-sterile CSB-K medium

(section 3.2.2) containing 50 mM nitrate (the medium was used as the source of microorganisms); 100 µL of heptane was added into the bottle with microcosm

#0. Upon being tightly closed with a screw cap, it was incubated at 20ºC for 40 days.

3.2.12. DNA pyrosequencing

To identify NRB able to use heptane as a sole source of carbon and energy, DNA was isolated from aliquots (5 mL) of microcosms #0, #1, #2 and

#5 (section 3.2.11). The FastDNA Kit and the FastPrep-24 homogenizer (MP

Biomedicals, Irvine, CA, USA) were used for DNA isolation. A single-step PCR amplification of 16S rDNA-fragments (30 cycles) was performed with FLX titanium primers 454T_RL_X and 454T_FwB with the sequences 926Fw16S

(AAACTYAAAKGAATTGRCGG) and 1392R16S (ACGGGCGGTGTGTRC) as

47 their 3’-ends. Primer 454T_RL_X had a 26 nucleotide A-adaptor

(CCATCTCATCCCTGCGTGTCTCCGAC) and a 10-nucleotide multiplex identifier barcode sequence. Primer 454T_FwB had a 26 nucleotide B-adaptor sequence (CCTATCCCCTGTGTGCCTTGGCAGTC). The PCR product was checked by agarose gel electrophoresis, purified with a QIAquick PCR

Purification Kit (Qiagen). The concentration of the purified product was determined with a Qubit fluorometer (Invitrogen) using a Quant-iT™ dsDNA HS

Assay Kit (Invitrogen). PCR-amplified fragments (20 L of 20 ng/L solution) were pyrosequenced at the Genome Quebec and McGill University Innovation

Centre (Montreal, Quebec) using a Genome Sequencer FLX Instrument with a

GS FLX Titanium Series Kit XLR70 (Roche Diagnostics Corporation).

3.2.13. Analysis of pyrosequencing data

Phoenix 2, an in-house developed small subunit rRNA gene data analysis pipeline, was used to conduct the analysis. Raw pyrosequence reads were subjected to stringent systematic checks in order to remove low quality reads and minimize sequencing errors that could be introduced during the pyrosequencing process (Huse et al., 2007). Eliminated sequences included those, which: (i) did not perfectly match the adaptor and primer sequences, (ii) had ambiguous bases, (iii) had an average quality score below 27, (iv) contained homopolymer lengths greater than 8, and (v) were shorter than 200 bp after clipping off the primers. The remaining high-quality sequences were compared against the non-redundant SSURef data set of SILVA102 (Pruesse et al., 2007) using the Tera-Blast algorithm on a 16-board TimeLogic Decypher system (Active Motif, Inc.). Sequences that had the best alignment covering

48 less than 90% of the trimmed read length, with greater than 90% sequence identity to the best BLAST match within the matched region and no match to the ends of the sequence, were labeled as potential chimeras and excluded from further analysis. While this might remove some true high-quality BLAST matches, in most cases a low percentage (less than 19%) of the sequences were eliminated, indicating that most of the real BLAST hits were included in the final study.

Sequences which passed the quality control and chimerical sequence removal were clustered into OTUs (Operational Taxonomic Units) at 3% distance by using the average linkage algorithm (Schloss and Westcott, 2011).

After grouping sequences into OTUs, several alpha diversity indices (Table 7), as well as the total numbers of OTUs were calculated for each sample. A taxonomic consensus for each representative sequence from each OTU was derived by using the classifier.otu function implemented in the Mothur software package (Schloss et al., 2009).

3.3. Results

3.3.1. Oil production during long-time incubation of columns with trapped

EO and LO

In all three sets of columns with trapped EO (sets I – III), columns containing nitrate produced consistently more oil than columns without nitrate during long-time incubations. While there were no true replicates in sets I – III, it was still reasonable to calculate average amounts of produced oil as trend indicators for similarly treated columns. Indeed, although incubation durations and numbers of incubation rounds for sets I – III differed, columns from each

49 single set were always incubated during the same time (Tables 1, A2-1, A2-2 and A2-3); column pore volumes in different sets differed (set I consisted of 30 mL columns and sets II and III consisted of 20 mL columns), but the OOIP/PV ratios for 20 mL and 30 mL columns were essentially the same (Table 1). On average, during long-time incubations, 10.1±5.9% of original oil in place (OOIP) was recovered from a column with EO, nitrate and an enrichment culture (N-1, i.e. columns I-1, II-1 and III-1 in Table 1); 6.1±3.3% of OOIP was recovered from a column with EO and nitrate (N-2, i.e. columns I-2, II-2 and III-2 in Table

1); and 1.3±0.35% of OOIP was recovered from an EO-containing column without nitrate (N-3, i.e. columns I-3, II-3 and III-3 in Table 1). The incubations also led to the production of water/brine from some columns (Table 1; Figure

8A).

In the set of columns with trapped LO (set IV), columns containing nitrate

(IV-1, IV-3, IV-4, and IV-5) produced 1.3 to 2.4 times more oil during long-time incubations than column IV-2 without nitrate (Table 1). Column IV-3, which contained 90 mM nitrate and enrichment culture produced about 30% less oil than column IV-5, which, similarly, contained nitrate and enrichment culture

(Table 1). Column IV-4 with 0.5 M nitrate and no enrichment culture produced the largest amount of LO in this set (Table 1).

3.3.2. Gas production during long-time incubation of columns with trapped EO and LO

Concentrations of gases produced during long-time incubations varied

(Table 2). No correlation was found between N2 concentration in a vial and the

Figure 8A. Production of oil and water from columns from sets I and II during long-time incubations at 30ºC: in set I, by the end of an incubation round, columns I-1 and I-2 produced some oil, while column I-3 produced neither oil nor water; in set II, by the end of an incubation round, column II-1 produced some oil, column II-2 produced some water, and column II-3 produced neither oil nor water (during some other incubation rounds columns N-3 also produced some liquids). Differences between columns in each set are described in the text and in Table 1.

Figure 8B. Production of water from a 30-mL column with SPW (no oil) during long-time (45-day) incubation at 20ºC.

Table 1. Oil and water production from columns containing trapped oil with and without nitrate during incubations in an anaerobic hood at 30˚C. Set I – III columns contained EO and set IV columns contained LO. Media were replenished at the beginning of each round of incubations (Tables A2-1, A2-2, A2-3 and A2-4). No water flood was applied during incubations.

Conc. of Presence Total Total nitrate in Volume Water of volume of volume of Pore the Total of flood enrichm. oil water Column Type Column volume OOIP, OOIP/ beginning incub. Voi/ Voi/ Voi/ Vwi/ Voi/PV + trapped recovery, culture produced produced set of oil number (PV), mL PV of each time, PV OOIP Vot PV Vwi/PV oil (Vot), (OOIP- in the during during mL round of days mL Vot)/OOIP first incub. incub. incub., injection (Voi), mL (Vwi), mL mM I EO I-1 13.37 12.21 0.913 4.536 0.628 20* Yes 341 2.066 0.154 0.169 0.455 0.45 0.034 0.188 I EO I-2 13.92 11.83 0.850 3.156 0.733 20* No 341 1.002 0.072 0.085 0.318 0.5 0.036 0.108 I EO I-3 13.48 12.38 0.918 3.059 0.753 0 No 341 0.159 0.012 0.013 0.052 0.15 0.011 0.023 II EO II-1 9.22 8.49 0.921 2.445 0.712 20* Yes 401 0.551 0.060 0.065 0.225 1 0.108 0.168 II EO II-2 9.46 8.73 0.923 2.568 0.706 20* No 401 0.204 0.022 0.023 0.079 1.25 0.132 0.154 II EO II-3 9.66 8.87 0.918 3.046 0.657 0 No 401 0.150 0.016 0.017 0.049 0.9 0.093 0.109 III EO III-1 9.00 8.31 0.923 2.658 0.680 20* Yes 311 0.580 0.064 0.070 0.218 0 0 0.064 III EO III-2 8.65 7.85 0.908 2.756 0.649 20* No 311 0.590 0.068 0.075 0.214 0.9 0.104 0.172 III EO III-3 8.70 8.05 0.925 3.874 0.519 0 No 311 0.080 0.009 0.010 0.021 0 0 0.009 IV LO IV-1 13.64 12.74 0.934 1.897 0.851 20 No 112 0.364 0.027 0.029 0.192 0.14 0.010 0.037 IV LO IV-2 13.40 12.5 0.933 1.677 0.866 0 No 112 0.292 0.022 0.023 0.174 0.3 0.022 0.044 IV LO IV-3 13.01 12.19 0.937 1.953 0.840 90 Yes 112 0.454 0.035 0.037 0.233 0.38 0.029 0.064 IV LO IV-4 13.17 12.29 0.933 1.937 0.842 500 No 112 0.701 0.053 0.057 0.362 0.2 0.015 0.068 IV LO IV-5 13.18 12.24 0.929 2.369 0.806 900 Yes 112 0.629 0.048 0.051 0.265 0.4 0.030 0.078

*In one of the incubation rounds, 50 mM nitrate was injected instead of 20 mM nitrate (Tables A2-1, A2-2 and A2-3).

Table 2. N2 and CO2 production from columns containing trapped oil with and without nitrate during long-time incubations in an anaerobic hood at 30˚C.

Conc. of Presence Average nitrate in Average of conc. of N the Total 2 conc. of CO enrichm. in a vial 2 Column Type Column beginning incub. in a vial after culture after one set of oil number of each time, one round of in the round of round of days incub., first incub., incub., mmol/L injection mmol/L mM I EO I-1 20* Yes 341 21.7±10.3 1.2±0.6 I EO I-2 20* No 341 28.4±17.6 1.7±0.9 I EO I-3 0 No 341 15.9±12.2 1.0±0.3 II EO II-1 20* Yes 401 21.0±15.8 1.9±1.2 II EO II-2 20* No 401 20.0±14.0 1.4±0.4 II EO II-3 0 No 401 18.1±14.3 1.7±0.8 III EO III-1 20* Yes 311 18.5±10.0 1.4±0.8 III EO III-2 20* No 311 21.3±15.2 2.0±0.7 III EO III-3 0 No 311 18.7±14.6 1.2±0.6 IV LO IV-1 20 No 112 20.5±9.9 1.1±1.1 IV LO IV-2 0 No 112 14.8±0.3 1.7±0.2 IV LO IV-3 90 Yes 112 5.8±0.6 1.0±0.2 IV LO IV-4 500 No 112 32.4±28.4 1.2±0.7 IV LO IV-5 900 Yes 112 15.4±1.5 1.2±0.2

*In one of the incubation rounds, 50 mM nitrate was injected instead of 20 mM nitrate (Tables A2-1, A2-2 and A2-3).

Table 3. Concentrations of nitrate and nitrite at the beginnings and the ends of incubation rounds A and B conducted with set IV columns containing LO. Round A started on the day “0” (Table A2-4) and lasted for 37 days and Round B started on the 68th day after the beginning of the first incubation round and lasted for 44 days (Table A2-4).

Round A: Round A: Round B: Round B: Conc. of - - - - - average NO average NO NO conc. NO conc. NO in the 3 2 3 2 3 conc. in conc. in in effluent at in effluent at Column # beginning effluent at the effluent at the the end of the end of of each end of the end of the the round*, the round*, round, mM round*, mM round*, mM mM mM

IV-1 20 12.5±1.8 0.13±0.11 20.3 0.5 IV-2 0 0 0 0 0 IV-3 90 68.7±10.9 1.59±0.63 89.5 5.6 IV-4 500 455.3±13.7 1.91±1.07 498.7 4.5 IV-5 900 895.3±62.8 3.14±2.32 825.4 0 *Four 1.5 mL effluent fractions were collected from each column after Round A and one 1.5 mL effluent fraction was collected from each column after Round B. The first fractions collected after Round A had highest concentrations of nitrite and lowest concentrations of nitrate, while the following fractions had decreasing concentrations of nitrite and increasing concentrations of nitrate; average concentrations of these ions in effluents collected after Round A are shown in the Table.

54 presence of nitrate in a column during long-time incubation. Moreover, concentrations of N2 and CO2 in the vials attached to six reference columns, three of which were filled just with SPW (and no oil) and incubated at 30ºC for

40 days and three others, also filled just with SPW and incubated at 20ºC for 45 days (one of them is shown in Figure 8B), were similar to concentrations of these gases produced from experimental columns with oil. The reference columns also produced 0.5 to 1 mL of SPW during a long-time incubation round.

3.3.3. Nitrate and nitrite concentrations at the ends of long-time incubation rounds

- - No NO3 or NO2 were found in effluents collected from any column from sets I-III (containing EO) during media replenishments at the ends of single incubation rounds (HPLC-tests were conducted after two incubation rounds).

Nitrate and nitrite were found in all columns from set IV (containing LO) where nitrate was initially present (columns IV-1, IV-3, IV-4, and IV-5); concentrations

- - of NO3 and NO2 at the beginnings and the ends of these incubation rounds are shown in Table 3.

3.3.4. SPW-flood- and immiscible-gas-flood-driven oil production following long-time incubation rounds

Simultaneous and separate SPW and gas mix (90% N2 and 10% CO2) floods conducted at 30ºC after seven rounds of incubations with set II columns

(section 3.2.7) led to production of just traces of oil from column II-3. SPW flood

55 following five rounds of incubations conducted with set I columns at 20ºC did not lead to any oil production.

3.3.5. Compositions of EO and LO before and after long-time incubation of columns with trapped oil

GC-MS analysis showed that LO was generally poorer than EO in light hydrocarbons (e.g. LO contained no heptane and no toluene; Figure 10; Table

5). Full GC-MS chromatograms for EO and LO are shown in Figure 9. Pristane and phytane were present in EO, but not in LO; however, LO had some compounds, especially polyaromatic hydrocarbons (e.g. phenanthrene), which were absent in EO (Figures 12 and 15). To normalize the amounts of the compounds found in columns with EO, their peak areas were related to those of pristane found in the same columns. Peak areas for compounds found in columns with LO were related to those of phenanthrene. It was assumed that most pristane and phenanthrene would remain in the columns because of having sufficiently high molecular weight not to be easily eluted and, importantly, being recalcitrant to biodegradation (although their biodegradation cannot be totally ruled out; Atlas and Bartha, 1992; Bregnard et al., 1997;

Kosheleva et al., 2000). Peak areas of pristane in the original EO and incubated columns were similar (Figure 12), and the peak areas of phenanthrene in the original LO and incubated columns were similar as well (Figure 15). Thus, these reference compounds did not undergo or underwent negligible (bio)degradation in our experiments.

Extracts of EO obtained from incubated columns had much smaller peak areas for most light oil components (at least up to C10) than the original EO.

Figure 9. GC-MS chromatograms of the samples of original EO and LO.

Figure 10. Low molecular weight components in the samples of original EO and LO, as determined through GC-MS analysis (also see Tables 4 and 5).

Table 4. Relative concentrations of selected low molecular weight hydrocarbons in EO: ratios of areas for different peaks to peak areas for pristane used as an internal standard (retention time of pristane is 43.67; its peak areas are referred to as “p” in the Table). Compounds were tentatively identified with Wiley mass-spectral library; similarity to the library spectra was higher than 91% in all cases; standard mass-spectra for n-heptane, toluene, m-xylene and o-xylene were used for compound identification as well.

Samples # Retention time Compound EO:area/p I-1:area/p I-2:area/p I-3:area/p II-1:area/p II-2:area/p II-3:area/p III-1:area/p III-2:area/p III-3:area/p V-I:area/p V-II:area/p 1 5.412 benzene 0.099 0 0 0 0.026 0 0 0 0 0 0 0 2 5.837 2,3-dimethylpentane 0.178 0 0 0.038 0.021 0.019 0 0 0.011 0 0.159 0.156 3 6.088 3-methylhexane 0.206 0 0.017 0.029 0.016 0.013 0.024 0.019 0.014 0.026 0.171 0.170 4 6.463 1,2-dimethylcyclopentane 0.287 0.034 0.018 0.041 0 0.020 0.048 0.018 0.027 0.036 0.165 0.254 5 6.826 n-heptane 0.279 0 0 0.025 0 0 0.032 0.016 0.016 0.025 0.235 0.231 6 7.64 1,1,3-trimethylcyclopentane 0 0 0 0 0 0 0 0 0 0 0 0 7 7.958 ethylcyclopentane 0.251 0 0 0.031 0 0.025 0.030 0.017 0.021 0.019 0.218 0.215 8 8.27 1,2,4-trimethylcyclopentane 0.134 0.025 0.013 0.038 0 0 0.016 0 0 0.024 0.123 0.124 9 8.65 2,3,4-trimethylpentane 0 0 0 0 0 0 0 0 0 0 0 0 10 8.821 toluene 0.246 0 0 0 0.025 0.013 0.020 0.014 0.011 0.018 0.156 0.182 11 9.616 3-methylheptane 0.195 0.032 0.025 0 0.015 0.020 0.028 0.011 0 0.029 0.182 0.175 12 10.16 1-ethyl-3-methylcyclopentane 0.090 0 0 0 0 0 0 0 0 0.018 0.084 0.080 13 10.31 1-ethyl-2-methylcyclopentane 0.282 0.062 0.033 0.093 0.045 0.030 0.046 0.015 0.044 0.065 0.274 0.273 14 10.529 1,2-dimethylcyclohexane 0.141 0.046 0.026 0.064 0.039 0.025 0.036 0.020 0.031 0.046 0.137 0.139 15 10.729 n-octane 0.952 0 0 0 0 0 0 0 0 0 0.699 0.668 16 12.97 1,3,5-trimethylcyclohexane 0.153 0.038 0.028 0.056 0.020 0.029 0.057 0.029 0.039 0.043 0.092 0.092 17 13.225 m,p-xylene 0.474 0.053 0.045 0.077 0.037 0.049 0.078 0.043 0.051 0.069 0.419 0.423 18 13.957 3-methyloctane 0.291 0.111 0.060 0.108 0.042 0.065 0.088 0.035 0.065 0.080 0.419 0.417 19 14.226 o-xylene 0.292 0 0.051 0.100 0.049 0.045 0.072 0.039 0.060 0.069 0.242 0.249 20 15.208 n-nonane 0.754 0.190 0.141 0.221 0.093 0.120 0.175 0.059 0.144 0.165 0.727 0.725 21 17.335 1-ethyl-3-methylbenzene 0.244 0 0 0 0 0 0 0 0 0 0.231 0.203 22 17.422 1-ethyl-2-methylbenzene 0.299 0.085 0.053 0.153 0.054 0.066 0.083 0.035 0.054 0.111 0.283 0.285

Figure 11. Low molecular weight hydrocarbons present in the original EO and in columns with EO after incubations.

Figure 12. Higher molecular weight hydrocarbons present in the original EO and in columns with EO after incubations.

Figure 13. Low molecular weight compounds in the original EO and EO extracted from the water-flooded (containing trapped oil), but not subjected to a long-time incubation column V-I.

Table 5. Relative concentrations of selected low molecular weight hydrocarbons in LO: ratios of areas for different peaks to peak areas for phenanthrene used as an internal standard (retention time for phenanthrene is 44.84; its peak areas are referred to as “ph” in the Table). Compounds were tentatively identified with Wiley mass-spectral library; similarity to library spectra was higher than 91% in all cases; standard mass-spectra for n-heptane, toluene, m-xylene and o-xylene were used for compound identification as well.

Samples # Retention time Compound LO:area/ph IV-1:area/ph IV-2:area/ph IV-3:area/ph IV-4:area/ph IV-5:area/ph 1 5.412 benzene 0 0 0 0 0 0 2 5.84 2,3-dimethypentane 0.059 0.034 0.034 0.041 0.043 0.040 3 6.088 3-methylhexane 0 0 0 0 0 0 4 6.463 1,2-dimethylcyclopentane 0 0 0 0 0 0 5 6.826 n-heptane 0 0 0 0 0 0 6 7.64 1,1,3-trimethylcyclopentane 0.081 0.053 0.054 0.041 0.050 0.034 7 7.958 ethylcyclopentane 0 0 0 0 0 0 8 8.27 1,2,4-trimethylcyclopentane 0.089 0.078 0.059 0.028 0.062 0.051 9 8.65 2,3,4-trimethylpentane 0.090 0.048 0.056 0.034 0.055 0.041 10 8.821 toluene 0 0 0 0 0 0 11 9.616 3-methylheptane 0 0 0 0 0 0 12 10.16 1-ethyl-3-methylcyclopentane 0 0 0 0 0 0 13 10.31 1-ethyl-2-methylcyclopentane 0 0 0 0 0 0 14 10.529 1,2-dimethylcyclohexane 0 0 0 0 0 0 15 10.729 n-octane 0 0 0 0 0 0 16 12.97 1,3,5-trimethylcyclohexane 0.447 0.316 0.338 0.222 0.286 0.286 17 13.225 m,p-xylene 0 0 0 0 0 0 18 13.957 3-methyloctane 0 0 0 0 0 0 19 14.226 o-xylene 0 0 0 0 0 0 20 15.208 n-nonane 0 0 0 0 0 0 21 17.335 1-ethyl-3-methylbenzene 0 0 0 0 0 0 22 17.422 1-ethyl-2-methylbenzene 0.157 0.127 0.152 0.127 0.122 0.132

Figure 14. Contents of some low molecular weight hydrocarbons in LO after incubations (compared to their original contents in LO).

Figure 15. Contents of higher molecular weight hydrocarbons in LO after incubations (compared to their original contents in LO).

Otherwise, EO remaining in incubated columns and the original EO had similar oil composition profiles. General oil compositions and all peak areas were very similar in the original EO and EO extracted from columns, which were water- flooded (until essentially no more oil was produced), but not subjected to incubations (Table 4; Figure 13).

After several incubation rounds, heptane was completely depleted in columns I-1, I-2, II-1 and II-2, whereas about 10% of heptane originally present in EO was remaining in columns I-3 and II-3 of these sets (Table 4; Figure 11); columns III-1 and III-2 contained about 5.5% of remaining heptane and the column III-3 contained about 10% of remaining heptane (Table 4). Toluene was absent in set I columns after incubations (Figure 11; Table 4); however, it was still present in the amount of up to 5% of its original content in EO in all columns from sets II and III (Table 4).

After incubation of the columns with LO, no noteworthy differences between amounts of oil components remaining in columns with and without nitrate were found; e.g. such light hydrocarbons as 1-ethyl-2-methylbenzene,

2,3-dimethylpentane, 1,1,3-trimethylcyclopentane and some others (Table 5;

Figure 14), present in the original LO, were about equally represented in columns with and without nitrate after long-time incubations.

Contents of different hydrocarbons in water-flooded (containing trapped oil), but not subjected to a long-time incubation columns V-I and V-II only slightly differed from their contents in the original EO (Table 4; Figure 13).

Among the hydrocarbons listed in Table 4, benzene, toluene and n-octane were subject to the greatest content reduction during water flood. However, columns

V-I and V-II still had much greater amounts of these compounds than all other columns, which were incubated for long time.

3.3.6. Biodegradation of heptane by NRB in enrichment cultures

Establishment of enrichment cultures with heptane as a sole source of carbon and energy, conducted to determine if differences in its contents in columns containing EO with nitrate and those without nitrate (Figure 11 and this section above) could be explained by biodegradation of heptane, was successful in the cases of microcosms #1, #2 (both containing 20 mM nitrate and 22 µL of heptane; section 3.2.11) and #5 (with 50 mM nitrate and 55 µL of heptane). Microbial growth in these microcosms was indicated by reduction of nitrate to nitrite (Figures 17 and 19). The biomass was clearly visible in microcosms #1 (Figure 16) and #2. No reduction of nitrate to nitrite was observed in microcosms #3 (Figure 18), #4, #6 (Figure 20) and #8 (Figure 22), all without heptane; nitrate was also not reduced to nitrite in microcosm #7 containing 100 µL of heptane (Figure 21). Azoarcus was found to be a dominant genus in microcosms #1, #2 and #5; Delftia was a dominant genus in microcosm #0 (Table 6). Alpha-diversity indices (Table 7) showed that microcosms #1 and #2 were more diverse than microcosms #0 and #5 (e.g.

Shannon-Wiener indices were 3.1 and 2.6 for #1 and #2 respectively, whereas for #0 and #5 they were 1.33 and 0.15 respectively; Simpson indices were 0.12 and 0.19 for #1 and #2 respectively, whereas for #0 and #5 they were 0.51 and

0.97 respectively). The richness shown by rarefaction curves (Figure 27) was also higher in microcosms #1 and #2 than in microcosms #0 and #5.

3.4. Discussion

3.4.1. Mechanisms of oil production during long-time incubation of columns with trapped oil

According to the results of our study, long-time incubation of miniature model columns containing trapped oil and nitrate at 30ºC in anaerobic conditions leads to water-flood-independent production of more oil than incubation of such columns without nitrate. Such water-flood-independent oil production occurs because of counter-diffusion induced gas drive (Graves et al.,

1972; Graves et al., 1973). As it was described in section 1.5.1, gas drive can be induced due to the difference in the rates of diffusion of different gases in liquids. In our experiments, helium (or argon), being the only gas in the attached vials, diffused into columns (in accordance with Le Chatelier’s principle), while nitrogen and carbon dioxide, dissolved in water and oil, diffused into vials.

Because diffusivities of helium and argon in water are higher than diffusivities of nitrogen and carbon dioxide (Lever et al., 1971; Graves et al., 1973; Prinzhofer,

2012), the rate of helium (or argon) influx into a column had to exceed the rates of the efflux of nitrogen and carbon dioxide out of a column. This could lead to supersaturation of liquid phases with gases (i.e. the total amount of dissolved gases could exceed that corresponding to equilibrium at the ambient pressure), under-pressurization of a vial and formation of bubbles in liquid phases (Graves et al., 1972; Graves et al., 1973). Thus, gas drive essentially identical to solution gas drive (Holm, 1987; Holm and Josendal, 1974) would be created, and oil and/or water could be pushed out of a column into a vial. Some dissolved gases could be released from solutions as bubbles due to the drop of their solubility in oil and water after columns were transferred from a 20ºC into a 30ºC

65 environment (as long-time incubations were conducted at 30ºC, while all procedures preceding them were conducted at 20ºC; section 3.2.7).

Temperature change could facilitate oil release in two more ways: with the temperature increase, oil would expand and its mobility would increase due to decrease in viscosity.

Attachment of oil droplets to floating gas bubbles in the way similar to that observed during wastewater clean-up (Moosai and Dawe, 2003; section

1.5.1) could play an important role in the counter-diffusion-induced-gas-driven oil recovery. Some other processes (sections 1.5.1 and 1.5.2; Kumar et al.,

2000; Shahvali and Pooladi-Darvish, 2009) could contribute to oil and water production as well. In particular, liquid swelling might have noteworthy effects.

According to The Engineering Toolbox (2012), solubility of He in water at 20ºC is 0.0015 g/kg and at 30ºC it is 0.0014 g/kg (=1.4 mg/kg) and solubility of Ar in water at 20ºC is 0.06 g/kg and at 30ºC it is 0.05 g/kg (=50 mg/kg). According to

Burton (2004), solubilities of He and Ar in oil exceed those in water 1.7 and 5.3 times respectively; thus, solubility of He in oil at 20ºC is 0.0015x1.7=0.0026 g/kg and at 30ºC it is 0.0014x1.7=0.0024 g/kg (=2.4 mg/kg), and solubility of Ar in oil at 20ºC is 0.06x5.3=0.32 g/kg and at 30ºC it is 0.05x5.3=0.27 g/kg (=270 mg/kg). Our 30 ml columns having PV of 13.4 ml contained 30% of trapped EO and 70% of water (sections 3.2.3 and 3.2.6; Figure 5). Therefore, total amount of He, which could be dissolved in water, equalled 13.4*0.7*1.4/1000=0.013 mg, and in oil 13.4*0.3*2.4/1000=0.01 mg; similarly, total amount of Ar, which could be dissolved in water, equalled 0.47 mg, and in oil 1.09 mg. This indicates that production of noticeable amounts of oil and water might occur due to their swelling after concentrations of Ar in the gaseous and liquid phases reached

66 equilibrium when the concentrations of N2 and CO2 in water and oil still exceeded their equilibrium concentrations (because of their lower diffusivities than that of Ar). Swelling of liquid phases had to be rather insignificant when He was used instead of Ar.

The experiment with columns filled just with SPW (and no oil) and incubated at 30ºC or 20ºC (Figure 8B) confirmed that during long-time incubations liquids were driven out of the columns mainly due to a physical, not biological process. Indeed, liquid (water) amounts produced from these columns during long-time incubations and concentrations of N2 and CO2 found in the vials originally filled with helium at the ends of incubation rounds were comparable to those found in the vials attached to columns with oil.

Interestingly, the concentrations of N2 and CO2 found in the vials (Table

2) at the ends of incubation rounds exceeded those expected based on their solubilities in water and oil, and were comparable to their concentrations in the anaerobic hood. Solubility of N2 in water at 20ºC is 0.02 g/kg (The Engineering

Toolbox, 2012). This means that 0.7 mM of dissolved N2 was present in water.

Taking into account that its solubility in oil is 5.2 times higher than that in water

(Burton, 2004), its concentration in oil can be assumed to be 3.6 mM. We can also take into account that 70% of a column volume is occupied by water and

30% of it is occupied by oil and assume that the concentration of dissolved N2 equals 0.02 mmol per PV (13.38 ml). If 1000 mmol of a dissolved gas (treated as an ideal gas) leaves a solution, then it will occupy 22.4 L, (to reach a concentration of 44.6 mmol/L); if 0.02 mmol of it leaves a solution, then its concentration in a headspace would be 0.9 µmol/L. Table 2 shows that concentrations of N2 in vials (headspaces) at the ends of incubation rounds

67 were much higher than 0.9 µmol/L. This suggests that gas was able to penetrate a column and/or a vial from an anaerobic hood (where N2 was present in the concentration of 90%, which was close to 44.6 mmol/L). Control vials filled with helium and not attached to a column had only traces of N2 after

40-day incubations, indicating that it was the columns, not the vials to be permeable to gas. Columns, indeed, contained gas-permeable materials: stoppers were made of rubber, and 20 mL and 30 mL syringe barrels were made of polypropylene (section 3.2.3; Figures 2 and 3). In addition, the area of contact between the stopper and the syringe barrel (Figure 3B), not being permeable to liquids, was likely permeable to gas. This means that not only gases (N2 and CO2) originally dissolved in liquids and those present in vials (He or Ar), but also those permeating the columns from the anaerobic hood (N2 and

CO2) were involved in the counter-diffusion process, possibly contributing to gas bubble formation.

Theoretically, a noticeable amount of N2 could be produced microbially if

20 mM of nitrate present in a column is totally reduced to nitrogen gas.

Stoichiometrically, 10 mM of N2 would be formed out of 20 mM of nitrate. Let’s assume that the pore volume of a column containing trapped EO is 13.38 mL

(section 3.2.3) and 70% of PV (=13.38*0.7=9.37 mL) is occupied by SPW

(section 3.2.6), where all nitrate is dissolved; a concentration of 20 mM nitrate re-calculated per 9.37 mL equals CNO3= 20*9.37/1000=0.19 mmol per 9.37 mL;

- if 10 mM of N2 is formed out of 20 mM NO3 , then concentration of the formed

N2 per 9.37 mL would equal CN2=0.19/2=0.1 mmol per 9.37 mL. At atmospheric pressure, 1000 mmol of N2 would occupy 22400 mL, and 0.1 mM would occupy

VN2=0.1*22400/1000=2.24 mL, which may be sufficient for displacing noticeable

Figure 16. Enrichment culture of NRB capable of consuming heptane as a sole source of carbon and energy (microcosm #1 containing 20 mM nitrate and 22 µL of heptane per 50 mL of the medium; section 3.2.11). Microbial growth is seen near the bottom of the bottle.

, mM , - 2 20 nitrite

or NOor -

3 nitrate 10

conc. of NO of conc. 0 0 10 20 30 40 50 time, days

Figure 17. Concentrations of nitrate and nitrite in microcosm #1 initially containing 20 mM nitrate and 22 µL of heptane per 50 mL of the medium (section 3.2.11) as functions of time.

25 , mM ,

- 20 2

or NOor

- 3 10 nitrite 5

nitrate conc. of NO of conc. 0 0 10 20 30 40 50 time, days

Figure 18. Concentrations of nitrate and nitrite in microcosm #3 initially containing 20 mM nitrate and no heptane (section 3.2.11) as functions of time.

, mM , - 2 50 nitrite 40

or NOor nitrate -

3 30 20 10

0 conc. of NO of conc. 0 10 20 30 40 50 time, days Figure 19. Concentrations of nitrate and nitrite in microcosm #5 initially containing 50 mM nitrate and 55 µL of heptane per 50 mL of the medium (section 3.2.11) as functions of time.

, mM , 50

- 2

or NOor -

3 30 nitrite 20 nitrate

10 conc. of NO of conc. 0 0 10 20 30 40 50 time, days

Figure 20. Concentrations of nitrate and nitrite in microcosm #6 initially containing 50 mM nitrate and no heptane (section 3.2.11) as functions of time.

, mM , 80 - 2 70

60 or NOor

- 50 3 40 30 nitrite 20 nitrate 10 conc. of NO of conc. 0 0 10 20 30 40 50 time, days

Figure 21. Concentrations of nitrate and nitrite in microcosm #7 initially containing 80 mM nitrate and 100 µL of heptane per 50 mL of the medium (section 3.2.11) as functions of time.

, mM , - 2 70

60 or NOor

- 50 3 40 nitrite 30 nitrate 20

conc. of NO of conc. 10 0 0 10 20 30 40 50 time, days

Figure 22. Concentrations of nitrate and nitrite in microcosm #8 initially containing 80 mM nitrate and no heptane (section 3.2.11) as functions of time.

Table 6. Numbers and fractions of reads for genera found in enrichment cultures (microcosms) with heptane as a sole source of carbon and energy. The genera are arranged in a descending order starting with an entry having the largest fraction of total reads in a single microcosm, followed by an entry with the next largest fraction of total reads in a single microcosm etc. (the largest fraction of total reads in each row is shown in bold). Only those having representation of more than 1% of total reads in at least one of the microcosms are shown.

Microcosm #0 Microcosm #1 Microcosm #2 Microcosm #5 fraction fraction fraction fraction Entry number of total number of total number of total number of total Phylum; Class; Order; Family; Genus # of reads reads of reads reads of reads reads of reads reads (%) (%) (%) (%) 1 Proteobacteria;Betaproteobacteria;Rhodocyclales;Rhodocyclaceae;Azoarcus 0 0 1938 34.283 2673 41.378 4541 98.375 2 Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Delftia 3580 71.372 0 0 0 0 0 0 3 Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;? 452 9.011 42 0.743 782 12.105 1 0.022 4 Chloroflexi;Anaerolineae;Anaerolineales;Anaerolineaceae;? 0 0 525 9.287 561 8.684 8 0.173 5 Bacteroidetes;Flavobacteria;Flavobacteriales;Flavobacteriaceae;? 0 0 365 6.457 507 7.848 1 0.022 6 Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Variovorax 378 7.536 0 0 0 0 0 0 7 Bacteroidetes;Flavobacteria;Flavobacteriales;Flavobacteriaceae;Vitellibacter 0 0 395 6.987 250 3.87 3 0.065 8 Bacteroidetes;Flavobacteria;Flavobacteriales;Flavobacteriaceae;Kaistella 0 0 143 2.53 406 6.285 0 0 9 Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Simplicispira 0 0 341 6.032 160 2.477 1 0.022 10 Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Thermomonas 0 0 298 5.272 4 0.062 0 0 11 Bacteroidetes;B._class_incertae_sedis;B._order_incertae_sedis;B._incertae_sedis 239 4.765 29 0.513 1 0.015 3 0.065 12 Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;? 9 0.179 43 0.761 185 2.864 21 0.455 13 Proteobacteria;Betaproteobacteria;Rhodocyclales;Rhodocyclaceae;Thauera 0 0 156 2.76 0 0 3 0.065 14 Proteobacteria;Betaproteobacteria;Rhodocyclales;Rhodocyclaceae;Azospira 0 0 132 2.335 29 0.449 2 0.043 15 Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodobacteraceae 0 0 101 1.787 96 1.486 2 0.043 16 Proteobacteria;Alphaproteobacteria;Caulobacterales;Caulobacteraceae;Brevundimonas 84 1.675 5 0.088 5 0.077 0 0 17 Proteobacteria;Gammaproteobacteria;Pseudomonadales;Pseudomonadaceae;Pseudomonas 78 1.555 22 0.389 13 0.201 0 0 18 Bacteria;Proteobacteria;Betaproteobacteria;Rhodocyclales;Rhodocyclaceae;? 1 0.02 83 1.468 25 0.387 1 0.022 19 Bacteroidetes;Flavobacteria;Flavobacteriales;Flavobacteriaceae;Flavobacterium 0 0 72 1.274 20 0.31 0 0 20 Others 195 3.887 963 17.034 743 11.502 29 0.628 21 All entries 5016 100 5653 100 6460 100 4616 100

300

250

200

150 #0

100 #1 #2 50

#5 number of new OTUs found OTUs new of number 0 0 1000 2000 3000 4000 5000 6000 7000 number of sequences sampled

Figure 23. Rarefaction curves for enrichment cultures (microcosms) with heptane as a sole source of carbon and energy.

Table 7. Alpha-diversity indices calculated for enrichment cultures (microcosms) with heptane as a sole source of carbon and energy.

#0 #1 #2 #5 sobs 66 244 185 47 chao 160.5 401.76 300.5 87.6 ace 135.3 533.45 363.28 143.44 jackknife 160.4 462.36 277.59 91.6 shannon 1.33 3.1 2.6 0.15 simpson 0.51 0.12 0.19 0.97

74 amounts of liquids from the column. Indeed, some strains of NRB were found to be capable of producing such amounts of N2, which were sufficient for increasing pressure inside a model column/oil reservoir, leading to production of oil and/or water (Nuryadi et al., 2011).

However, the fact that concentrations of N2 (and CO2) produced from columns without nitrate (and even from ones with SPW and without oil) were essentially not different from those produced from the columns with nitrate

(Table 2) indicates that microbial gas production could play only a minor role in creating gas drive in our case.

3.4.2. The mechanism of nitrate-mediated enhancement of gas-driven oil recovery

Although gas drive was induced mainly by physical, not biological factors, addition of nitrate to columns with trapped oil (especially with EO; Table

1) still made a difference during their long-time incubations. Nitrate reduction in the columns with EO by the ends of incubation rounds (section 3.3) suggested that microbial growth was taking place in them. This was less clear in the case of columns with LO, although some nitrate was reduced in them as well (Table

3). In similar experiments, Hitzman (1994) showed that addition of nitrate stimulated growth of NRB in sandstone cores when VFA (acetate and propionate) were used as sources of carbon and energy for microorganisms.

Chakraborty and Coates (2004) demonstrated anaerobic biodegradability of several monoaromatic hydrocarbons, which can be present in different types of oil. According to Lambo et al. (2008) and Agrawal et al. (2012), a particular oil component, toluene, is a preferred substrate/electron donor for microbial nitrate

75 reduction in the field of origin of EO. In this study, heptane was shown to be readily consumed by NRB from this field as well (section 3.3 and this section below).

According to Dennis and Hitzman (2007) and Mulligan (2005), growth of

NRB is accompanied by the production of biosurfactants. Indeed, microbes need biosurfactants for enabling themselves to uptake hydrocarbons through either getting attached to them (Zhang and Miller, 1994) or through their solubilization (Ron and Rosenberg, 2002). As it was already mentioned above

(section 1.2.1), cooperative action of surfactants and bacterial cells can be powerful in reducing oil-water IFT (Bryant, 1989; Kowalewski et al., 2006).

Thus, enhancement of oil production from columns with nitrate could be enabled by the presence of more cells and biosurfactants on the surfaces of oil- droplets-bubbles than in the case of columns not stimulated with nitrate (Figure

1B). Increased numbers/concentrations of cells and biosurfactants (but not as many and not as high concentrations to fully cover bubble and oil droplet surfaces as it is shown in Figure 1A) would facilitate the movement of oil- droplets-bubbles in a water medium because of the reduced oil-water IFT. This conclusion is supported by the finding that the presence of cell-containing supernatants from cultures of some microorganisms, which can produce small amounts of biosurfactants, also enhances gas-driven oil production from miniature model columns during long-time incubations (Chapter 5).

Injecting enrichment cultures along with a nitrate-containing medium was not always associated with increased oil recovery compared to injecting a nitrate-containing medium alone (Table 1), indicating that sufficient numbers of microorganisms were originally present in EO and, possibly, in LO. This was

76 indeed possible because both EO and LO samples contained some water

(section 3.2.1).

3.4.3. Possible substrates for microbial growth during long-time incubation of columns with trapped oil

The likely preferred electron donor for microorganisms in the columns with EO (Lambo et al., 2008; Agrawal et al., 2012), toluene, was found to be totally absent in set I columns by the end of experiments. However, it was still present in small amounts in set II and III columns (Table 4). Its high volatility might play a role in the reduction of its concentrations in columns during long- time incubations. High volatility of other low molecular weight hydrocarbons could play a role in the reduction of their concentrations as well. The absence of consistent differences in residual toluene contents in columns with and without nitrate did not allow concluding that it was an important source of carbon and energy. Aerobic biodegradation of some toluene (and, possibly, of some other hydrocarbons) present in EO cannot be ruled out, because all injected and re- injected media contained some dissolved oxygen, which might be used by facultative anaerobes during several initial days of each incubation round, until all dissolved oxygen was depleted. In the case of set IV columns (containing

LO), toluene was totally absent in the original oil (Table 5).

Another light hydrocarbon, n-heptane, was completely absent in columns

I-1, I-2, II-1 and II-2 (with nitrate) by the end of experiments, whereas about

10% of heptane originally present in EO was still remaining in columns I-3 and

II-3 (without nitrate); similarly, columns III-1 and III-2 contained less heptane than the columns III-3 by the end of experiments (Table 4). The presence of

77 some heptane by the end of experiments in columns III-1 and III-2, but not in columns I-1, I-2, II-1 and II-2, could be explained by the fact that set III columns were incubated for shorter time than columns from sets I and II (Table 1).

Similarly to the situation with toluene, heptane originally present in a model column could be volatilized into vials during incubations and/or anaerobically and/or aerobically biodegraded. Persistent differences between the columns with EO, particularly between columns containing nitrate and those without it, indicate that during incubations, nitrate reduction indeed might be coupled to biodegradation of heptane. Other low molecular weight hydrocarbons, particularly ones mentioned in Table 4, could also serve as electron donors and/or be volatilized during long-time incubations.

3.4.4. NRB capable of heptane biodegradation

Successful establishment of enrichment cultures with heptane as a sole source of carbon and energy, nitrate (as an electron acceptor) and produced water (as a source of microorganisms) from the field of origin of EO (section

3.3; Figures 17 and 19) indicated that this hydrocarbon could be one of the preferred electron donors for oil field NRB. Table 6 shows that Azoarcus is clearly a dominant genus in microcosms #1, #2 and #5. This agrees with the information that Azoarcus and Thauera constituted the majority of the enriched denitrifiers grown with crude oil as the only substrate and capable of consuming

C5-C12 n-alkanes (Rabus et al., 1999). This is also in agreement with the reports by Rabus et al. (2001), Wilkes et al. (2002) and Grundmann et al. (2008), according to which short-chain alkanes can be readily consumed by NRB.

Rabus et al. (1999) indicated that during alkane consumption, bacteria adhered

78 to a crude oil layer and emulsified oil, suggesting that bacterial metabolites and/or cells had high surface activity. Hence, it may be worthwhile that species belonging to this genus (perhaps, together with some other representatives of the family Rhodocyclaceae and the representatives of families

Xanthomonadaceae, Anaerolineaceae, Flavobacteriaceae and

Comamonadaceae; Table 6) are paid particular attention to during the development of MEOR-technologies relying on indigenous microflora in the field of origin of EO. It will be interesting to see if Azoarcus tends to adhere to oil in the field of origin of EO. Chapter 4 will be dedicated to determining which genera are adhering to the oil phase and which ones are residing predominantly in the aqueous phase of the mixture of produced water and oil from the field of origin of EO.

Interestingly, Delftia was found to be a dominant genus in microcosm #0

(containing 100 µL/100 mL of heptane as the only source of carbon and energy), in which no oil field produced water was present and the non-sterilized medium was used as the source of microorganisms (section 3.2.11). This genus, having not been found in any of the enrichments containing oil field produced water (i.e. in microcosms #1, #2 and #5, Table 6), belongs to

Comamonadaceae, one of the families well represented in microcosms #1, #2 and #5. Species belonging to the genus Delftia have been reported to degrade m-xylene, anilines, terephthalate, protocatechuate, (S)-2,2- dimethylcyclopropanecarboxamide, mixed phenols and phenanthrene

(Khomenkov et al., 2005; Chen and Hickey, 2011; Jimenez et al., 2012;

Reinhold-Hurek and Hurek, 2006). However, no reports regarding degradation of heptane by Delftia were found in the literature.

The fact that microcosms #1 and #2 had higher species diversity than microcosm #5 (Table 7; Figure 27; section 3.3) indicated that heptane present in larger amount (55 µL/50 ml in microcosm #5 versus 22 µL/50 ml in microcosms #1 and #2) was toxic to most of bacteria; in fact, Azoarcus was the only one who was still thriving in this situation. In microcosm #0, the situation was similar to that in microcosm #5, although Delftia, not Azoarcus was the dominant bacterium in the former microcosm (Tables 6 and 7; Figure 27).

Failure to establish an enrichment culture in microcosm #7 containing

100 µL of heptane (section 3.3; Figure 21) indicates that this amount of it (per

50 mL of a nitrate-containing aqueous solution) is likely toxic to oil-field microflora, including Azoarcus. Indeed, there is the evidence that the presence of low molecular weight hydrocarbons in large amounts may render them toxic to their consumers (Nahar et al., 2000).

Nitrate-mediated enhancement of the recovery of EO, much richer than

LO in low molecular weight hydrocarbons (e.g. n-heptane and toluene; Figure

10; Tables 4 and 5), was greater than that of LO. Microbial activity expressed as the rate of nitrate reduction was also lower in the columns with LO than in the columns with EO (section 3.3 and Table 3). Therefore, the efficiency of nitrate injections for MEOR-purposes may ultimately depend on the composition of the recovered/trapped oil.

3.4.5. Long-time incubation of columns with trapped oil was not accompanied by plugging of high permeability zones

Neither combined SPW and gas mix floods (WAG) at 30ºC nor SPW floods at 20ºC conducted after several rounds of incubations with two column

80 sets (section 3.2.7) led to the release of trapped oil (except for oil traces from the set II column without nitrate during WAG process at 30ºC). This indicates that neither water nor gas could get into low permeability zones to displace trapped oil during these processes. Thus, the MEOR-effect due to plugging of high permeability zones (Gray et al., 2008; Voordouw, 2011) did not occur in our experiments. This is understandable because sand in our columns was packed homogeneously (section 3.2.3), so that large void spaces could not be formed. Therefore, high permeability zones in the columns were represented by some channels of the least resistance to water flow. An addition of carbohydrates could have led to much greater biomass increase than in the case, in which microorganisms could consume only oil components. Thus, one can hypothesize that channels of the least resistance to water flow still might have been plugged if microorganisms were provided with carbohydrates, in addition to nitrate.

3.4.6. Increased nitrate concentration did not help further enhance gas- driven oil recovery

In one of the incubation rounds, 50 mM nitrate was injected into some columns from sets I – III instead of 20 mM nitrate (Tables 1, A2-1, A2-2 and A2-

3), but no more oil was produced during this round than the average amount of oil produced during an incubation round with 20 mM nitrate. Hence, 20 mM nitrate (or even its lower concentration) was likely its maximum concentration that might help recover trapped EO under specified conditions. This might be explained by the limited presence of toluene and heptane as sources of carbon and energy in oil. However, the issue of the toxicity of low molecular weight

81 hydrocarbons present in relatively large amounts to NRB (which was discussed in section 3.4.4) should be taken into account as well.

3.5. Conclusions and implications

Based on the fact that counter-diffusion-induced-gas-driven oil recovery was improved when it was combined with stimulation of indigenous microflora with nitrate, one can conclude that in industrial settings combination of solution- gas-driven oil recovery with stimulation of indigenous microflora with nitrate can be beneficial. This combination may be especially successful if oil contains such low molecular weight hydrocarbons as n-heptane and toluene, which can be used as sources of carbon and energy. At the same time, one must keep in mind that at pressure, at which a gas becomes a supercritical fluid, microorganisms may be inactivated; e.g. their inactivation occurred when carbon dioxide in the form of a supercritical fluid was used (Dillow et al., 1999).

Findings described in this chapter lead to the question about the identities of microorganisms, which may be responsible for improved counter- diffusion-induced-gas-driven oil recovery upon a period of incubation with nitrate. Azoarcus is likely one of them. Chapter 4 is dedicated to making possible a broader answer to the posed question. As microorganisms attached to an oil-water interface are of a particular interest (Obuekwe et al., 2008; section 1.2.1), an attempt will be made to represent and analyze them apart from a microbial community residing in an aqueous phase of oil-containing produced water.

The other questions arising from the findings of this chapter are whether some well-known biosurfactants (e.g. rhamnolipids), in presence of microbial

82 cells, would improve counter-diffusion-induced-gas-driven oil recovery and whether high biosurfactant concentrations would be beneficial for gas-driven and water-flood-driven oil recovery. These questions will be addressed in

Chapter 5.

Chapter 4: Microbial communities associated with oil and water in

a mesothermic oil field

4.1. Introduction

Microorganisms residing at an oil-water interface (or, in other words, attached/adhering/associated to/with oil) may be present in core matrix pores, where oil is trapped in micro-droplets. It has been shown that production of biosurfactants and/or alcohols may be needed for microbial adhesion to hydrocarbons (Zhang and Miller, 1994; van Hamme et al., 2003; Abbasnezhad et al., 2008; Bødtker et al., 2009). On the other hand, surfactants, especially when combined with alcohols, are known to reduce IFT between oil and aqueous phases, and may in this way enhance oil recovery (Bryant and

Burchfield, 1989; Banat, 1995; Li et al., 2002; McInerney et al., 2005). Cells attached to an oil-water interface are also known to assist in reducing oil-water

IFT (Kowalewski et al., 2006) and to modify rheological properties of an oil- water interface in a way that may lead to plugging of high permeability zones

(Kang et al., 2008a). In some MEOR applications, transportability of microbes throughout an oil reservoir may be an important parameter (Sarkar et al., 1994), and in such cases reliance on the microbes associated with an aqueous phase may be required. Therefore, establishing identities of microorganisms isolated separately from oil and water phases of a produced oil-water mix may be useful for developing new or improving existing MEOR-technologies.

A number of methods have been used to study hydrophobic interactions of cells, for example those based on defining the degree of adhesion of cells to liquid hydrocarbons following a brief period of mixing, contact angle

84 measurements of dried cell layers, and partitioning of bacteria in aqueous polymer two-phase systems (Rosenberg et al., 1980). A method known as

MATH (microbial adhesion to hydrocarbons) has gained popularity due to its simplicity and acceptable level of accuracy, despite some disadvantages

(Rosenberg, 2006; van der Mei et al., 1995; Zoueki et al., 2010). It is based on defining the degree of adhesion of cells to liquid hydrocarbons following mixing

(Rosenberg et al., 1980; Rosenberg, 2006). In this study, an idea similar to one of the MATH method was employed in order to determine the compositions of the OC (microbial community associated with an oil phase) and the AC

(microbial community associated with an aqueous phase). In this method, oil- attached cells firstly rise to the surface with the oil at the gravitational acceleration (1 x g), and secondly they are dislodged from the oil and pelleted at about 104 x g.

Results presented in this chapter have been published in the article by

Kryachko et al. (2012a). An analysis of pyrosequencing data was conducted in collaboration with Dr. Christoph Sensen, Dr. Gerrit Voordouw and Xiaoli Dong.

4.2. Materials and methods

4.2.1. Isolation of biomass

Samples of oil-containing produced water were collected in 1L Nalgene bottles from production wells in the MHGC field (near Medicine Hat, Alberta,

Canada; it is the field of origin of EO) on 29/11/2009, 12/01/2010 and

02/02/2010. The bottles were filled to the top to prevent oxygenation (Voordouw et al., 2009), delivered to the laboratory within five hours after filling, and stored in an anaerobic hood (90% N2 and 10% CO2) at 20°C. The samples were

85 separated into oil and aqueous phases in the anaerobic hood in a separatory funnel (Figure 24). Following standing overnight, the aqueous phase was drained off and biomass associated with it was collected through centrifugation at 25,000 x g for 20 min in 250 mL centrifuge bottles. The pellets were resuspended in 1 ml of sterile synthetic brine (90 mM NaCl; 1 mM KCl; 1 mM

MgCl2; 2 mM CaCl2; 10 mM NaHCO3; pH 7.5) by agitating for 50-60 seconds with a Vortex mixer, transferred to 1.5 mL microcentrifuge tubes and centrifuged at 17,000 x g for 10 min. The supernatant was removed and collected biomass was subjected to DNA extraction. The separatory funnel containing the remaining oil phase was replenished with 1 – 1.5 L of sterile synthetic brine, manually shaken for 2 minutes, and allowed to stand for at least 24 hours.

Following this, the synthetic brine was drained off and the process was repeated at least twice to remove microorganisms associated with the aqueous phase. Oil-associated biomass was dislodged from the oil by centrifuging at

25,000 x g for 20 min in 250 mL centrifuge bottles containing 50 – 100 mL of sterile synthetic brine (depending on the amount of oil). The oil-containing supernatant was removed, 25 – 30 mL of sterile synthetic brine was added to the pellet and the tube was agitated for 60 seconds with a Vortex mixer to suspend the cells. The cell suspension was transferred to 50 mL centrifuge tubes and re-centrifuged. Sterile synthetic brine (1 mL) was added and the samples were processed by microcentrifugation as indicated above. The entire process was replicated seven times. Four samples of biomass isolated from oil- containing produced water collected on 29/11/2009 (two samples of oil- associated biomass and two samples of water-associated biomass) were used

Figure 24. Separation of produced water and oil (EO) in an anaerobic hood filled with 90% N2 and 10% CO2.

87 for an initial screening of the differences between compositions of two ACs (AC- dgge-1 and AC-dgge-2) and two OCs (OC-dgge-1 and OC-dgge-2), which was based on DGGE analysis; ten samples of biomass isolated from oil-containing produced water collected on 12/01/2010 and 02/02/2010 (five samples of oil- associated biomass and five samples of water-associated biomass) were used for conducting the analysis of compositions of five OCs (OC-1 – OC-5) and five

ACs (AC-1 – AC-5), which was based on the results of 16S-rRNA gene pyrosequencing (Table 8).

4.2.2. DNA isolation

DNA was isolated according to Marmur (1961), with a bead beating step modification. Oil- and water-associated biomass was re-suspended in 280 µL

0.15 M NaCl and 0.1 M EDTA (pH 8) and subjected to digestion with lysozyme

(20 µL of 5 mg/mL, at 37ºC for 10 min or longer). Samples were then transferred into 2 mL bead-containing lysing matrix tubes (MP Biomedicals) and subjected to bead-beating in a FastPrep-24 homogenizer (MP Biomedicals, Irvine, CA,

USA) for 1 min; subsequently, they were centrifuged at 17000 x g for 1 min and transferred to microcentrifuge tubes; 72 µL 5 M NaClO4 and 420 µL chloroform- isoamyl alcohol (24:1) were added, and the mix was placed on a rotating wheel for 1 hour or more. The mix was centrifuged for 3.5 min at 17000 x g in the same tube at 20°C; 350 µL of the top layer was transferred to another microcentrifuge tube and 700 µL of 95% ethanol was added. Following centrifugation (15 min, 17000 x g, 4ºC), the pellet was dissolved in 200 µL of TE

(10 mM Tris, 0.1 mM EDTA pH 8) and incubated at 20°C with DNAse-free

RNAse (25 µL, 50 µg/mL) and proteinase K (10 µL, 20 mg/mL) sequentially,

88 with each for 1 h. Following phenol extraction and ethanol precipitation the air- dried DNA pellets were dissolved in 30-50 µL of TE-buffer, depending on the size of the pellet.

4.2.3. DGGE

DGGE was used as an initial screening tool to determine the differences between compositions of the AC and the OC. Four samples of DNA (isolated from AC-dgge-1, AC-dgge-2, OC-dgge-1 and OC-dgge-2) were analyzed using this method. DNA was amplified through PCR with the primers B27F and

B534R (Cornish-Shartau et al., 2010). DGGE was performed as described by

Grigoryan et al. (2008) using DCode (Bio-Rad) system with Tris-acetate-EDTA

(TAE) buffer (pH 8) at 60°C. Electrophoresis, staining, band excision, PCR amplification, PCR product purification, and sequencing were conducted according to the procedures described elsewhere (Cornish-Shartau et al., 2010;

Grigoryan et al., 2008). DNA fragments were sequenced at the UCDNA

Services at the Faculty of Medicine of the University of Calgary. Sequences are available in GenBank under accession numbers HQ616515 to HQ616525.

4.2.4. DNA pyrosequencing

Ten samples of DNA isolated from OC-1 – OC-5 and AC-1 – AC-5 were amplified through a two-step PCR amplification. Namely, the first PCR (25 cycles) was performed with bacterial 16S rDNA primers 926Fw16S

(AAACTYAAAKGAATTGRCGG) and 1392R16S (ACGGGCGGTGTGTRC).

Once the presence of a 500 bp PCR product was confirmed by agarose gel electrophoresis, a second PCR (10 cycles) was performed using FLX titanium

89 primers 454T_RA_X and 454T_FwB with the sequences for 926Fw16S and

1392R16S as their 3’-ends. Primer 454T_RA_X had a 25 nucleotide A-adaptor

(CGTATCGCCTCCCTCGCGCCATCAG) and a 10-nucleotide multiplex identifier barcode sequence. Primer 454T_FwB had a 25 nucleotide B-adaptor sequence (CTATGCGCCTTGCCAGCCCGCTCAG). The PCR product was checked by agarose gel electrophoresis, purified with QIAquick PCR

Purification Kit (Qiagen), and the concentration of the purified product was determined with Qubit fluorometer (Invitrogen) using Quant-iT™ dsDNA HS

Assay Kit (Invitrogen). PCR products (25 µL of 5 ng/µL) were pyrosequenced at the Genome Quebec and McGill University Innovation Centre (Montreal,

Quebec) using a Genome Sequencer FLX Instrument with a GS FLX Titanium

Series Kit XLR70 (Roche Diagnostics Corporation).

4.2.5. Analysis of pyrosequencing data

The analysis was conducted using Phoenix 2, an in-house developed small subunit rRNA gene data analysis pipeline, as described in section 3.2.13.

The entire set of the raw reads is available at the NCBI Sequence Read Archive

(SRA) under accession numbers SRR088842, SRR088844, SRR088845,

SRR088846, SRR088847, SRR088848, SRR088850, SRR088852,

SRR088854 and SRR088855. Table 8 provides the information on the numbers of reads identified at the phylum, class and genus levels for ten samples which were analyzed.

To explore potential relationships between microbial communities from different environments, samples were clustered into Newick-formatted trees using the UPGMA algorithm with the distance between communities calculated

90 with the thetaYC coefficient as a measurement of similarity between community structures (Yue and Clayton, 2005) and the Jaccard index as a measurement of similarity between community memberships in the Mothur software package.

The trees were visualized using Dendroscope (Huson et al., 2007). In addition, parsimony testing (Fitch, 1971), unfrac.unweighted and unifrac.weighted

(Hamady et al., 2010; Lozupone and Knight, 2005) testing were used to determine the statistical significance of clustering within the trees.

4.3. Results

4.3.1. DGGE community analysis Generally, compositions of OC-dgge-1 and OC-dgge-2, on one hand, and AC-dgge-1 and AC-dgge-2, on the other hand, differed; the AC appeared to be more diverse than the OC; representation of Deltaproteobacteria was richer in the AC. Identifications of some bands present in similar positions in the AC and OC lanes were different (in particular, Bacteroidetes in the OC vs.

Deltaproteobacteria in the AC and Epsilonproteobacteria in the OC vs.

Gammaproteobacteria in the AC; Figure 25).

4.3.2. Alpha- and beta-diversity

The analysis was conducted with five OCs (OC-1 – OC-5) and five ACs

(AC-1 – AC-5). Two distinct OC and AC branch clusters were formed in the J- tree (Figure 26A). The clustering pattern within the YC-tree (Figure 26B) was very similar to that in the J-tree. The robustness of the inferred sample relation trees was confirmed using parsimony, weighted unifrac, unweighted unifrac and

AMOVA (Analysis of Molecular Variance) hypothesis testing (Table 9). This

91 indicated that the differences between OCs and ACs were statistically significant.

To eliminate possible differences introduced by sampling bias, data for five OC amplicon libraries were pooled together to form the OCP (pooled microbial community associated with the oil phase) and the data for five AC amplicon libraries were pooled together to form the ACP (pooled microbial community associated with the aqueous phase). The alpha-diversity indices

(Table 10) showed that the ACP was more diverse than the OCP (e.g. the

Shannon-Wiener indices were 4.38 and 3.89, and the Simpson indices were

0.039 and 0.054, respectively). The richness shown by rarefaction curves

(Figure 27) was also higher in the ACP than in the OCP. The non-parametric

Kendall tau rank correlation coefficient (Kendall, 1938) was calculated by using the R statistical software package (Gentleman, 2008). The p-value (0.0022) and tau (0.0551) suggested that the OCP and the ACP were highly independent from each other.

4.3.3. Characterization of the OCP and the ACP

4.3.3.1. Phylum and class levels

Proteobacteria and Euryarchaeota were found to be the most abundant phyla in both the OCP and the ACP (Table 11; Figures 28 and 29). The phyla

Actinobacteria, Firmicutes and Tenericutes (Table 11: #2, #3 and #1) were more numerous in the OCP, whereas the phyla Spirochaetes and Synergistetes had higher representation in the ACP (Table 11: #10 and #11). At the class level, the Alphaproteobacteria and Methanomicrobia, being the most numerous in both phases, had almost equal representation in the OCP and the ACP

Figure 25. 16S rRNA gene fragments of microorganisms belonging to the communities associated with the oil phase (OC-dgge-1 and OC-dgge-2) and the aqueous phase (AC-dgge-1 and AC-dgge-2), analyzed by DGGE. Bacterial 16S rRNA gene primers B27F and B534R (Cornish-Shartau et al., 2010) were used.

Figure 26. Clustering analysis of pyrosequencing data. (A) Jaccard sample relation tree built using the UPGMA clustering algorithm based on the classical Jaccard values. (B) ThetaYC tree built using the UPGMA algorithm based on Yue and Clayton (2005) theta values. Both trees were calculated in Mothur and visualized with Dendroscope software.

Table 8. Numbers of reads for AC-1 – AC-5 and OC-1 – OC-5 identified at the phylum (NP), class (NC) and genus (NG) levels.

Sample NP NC NG AC-1 7655 6306 4946 OC-1 10749 9142 6464 AC-2 6302 5652 4687 OC-2 8396 7166 6273 AC-3 7551 7159 5977 OC-3 7615 6655 5369 AC-4 10328 9473 7827 OC-4 9387 8198 6904 AC-5 9975 8813 7361 OC-5 10132 8892 6862 Total reads 88090 77456 62670

Table 9. Parsimony, weighted unifrac, unweighted unifrac, and AMOVA hypothesis testing results. The testing shows that there is a structure within the trees that significantly separates the OC and the AC branches.

Hypothesis test name P-values for sample P-values for sample tree based on Jaccard tree based on thetaYC values values Parsimony (ParsSig) 0.003* 0.005* Unweighted Unifrac 0.00* 0.00* (Wsig) Weighted Unifrac 0.00* 0.00* (Wsig) AMOVA (p-value) 0.004* 0.007*

*Values less than 0.05 were considered significant.

Table 10. Alpha-diversity indices calculated for the OCP and the ACP.

OCP ACP sobs 1090 1521 chao 1750 2654 ace 2244 3627 jackknife 2111 3221 shannon 3.89 4.38 simpson 0.054 0.039

Figure 27. Rarefaction curves for the OCP and the ACP.

Table 11. Numbers and fractions of reads for phyla found in the OCP and the ACP. The list is ranked according to the ratio of fraction of total reads in the OCP to fraction of total reads in the ACP (OCP/ACP). Phyla overrepresented in the OCP (OCP/ACP>2) or in the ACP (OCP/ACP<0.5) are in bold and in italics, respectively. OCP+ACP shows the sum of fractions of total reads for the OCP and the ACP. Phyla with OCP+ACP equalling less than 0.1% are not shown.

OCP: OCP: ACP: ACP: fraction of fraction of # Phylum number of number of OCP+ACP OCP/ACP total reads total reads reads (%) reads (%) 1 Tenericutes 306 0.66 57 0.14 0.80 4.85 2 Actinobacteria 4214 9.12 1080 2.59 11.71 3.53 3 Firmicutes 1738 3.76 462 1.11 4.87 3.40 4 Euryarchaeota 18499 40.04 14192 33.98 74.01 1.18 5 Thermotogae 302 0.65 326 0.78 1.43 0.84 6 Proteobacteria 18113 39.20 20734 49.64 88.84 0.79 7 Deferribacteres 892 1.93 1140 2.73 4.66 0.71 8 Chloroflexi 754 1.63 1102 2.64 4.27 0.62 9 Bacteroidetes 1094 2.37 1627 3.90 6.26 0.61 10 Spirochaetes 232 0.50 778 1.86 2.36 0.27 11 Synergistetes 63 0.14 272 0.65 0.79 0.21 All entries 46207 100 41770 100

Table 12. Numbers and fractions of reads for classes found in the OCP and the ACP. The list is ranked according to the ratio of fraction of total reads in the OCP to fraction of total reads in the ACP (OCP/ACP). Classes overrepresented in the OCP (OCP/ACP>2) or in the ACP (OCP/ACP<0.5) are in bold and in italics, respectively. OCP+ACP shows the sum of fractions of total reads for the OCP and the ACP. Classes with OCP+ACP equalling less than 0.1% are not shown.

OCP: OCP: ACP: ACP: fraction of fraction of # Phylum; Class number of number of OCP+ACP OCP/ACP total reads total reads reads (%) reads (%) 1 Bacteroidetes;Sphingobacteria 73 0.17 13 0.03 0.20 5.21 2 Tenericutes;Mollicutes 306 0.69 57 0.14 0.83 4.98 3 Firmicutes;Clostridia 1696 3.85 412 1.01 4.85 3.82 4 Actinobacteria; Actinobacteria 2089 4.74 538 1.32 6.05 3.60 5 Proteobacteria;Gammaproteobacteria 2315 5.25 637 1.56 6.81 3.37 6 Euryarchaeota;Methanobacteria 6107 13.85 2260 5.53 19.37 2.51 7 Proteobacteria;Alphaproteobacteria 14275 32.37 13525 33.07 65.44 0.98 8 Euryarchaeota;Methanomicrobia 12352 28.01 11866 29.01 57.02 0.97 9 Proteobacteria;Betaproteobacteria 934 2.12 1177 2.88 5.00 0.74 10 Bacteroidetes;Bacteroidia 525 1.19 708 1.73 2.92 0.69 11 Chloroflexi;Anaerolineae 702 1.59 1051 2.57 4.16 0.62 12 Bacteroidetes;Flavobacteria 27 0.06 42 0.10 0.16 0.60 13 Synergistetes;Synergistia 63 0.14 272 0.67 0.81 0.21 14 Proteobacteria;Epsilonproteobacteria 36 0.08 157 0.38 0.47 0.21 15 Proteobacteria;Deltaproteobacteria 513 1.16 5154 12.60 13.77 0.09 Others 2089 4.74 3029 7.41 All entries 44102 100 40898 100

(Table 12: #7 and #8). Deltaproteobacteria, Epsilonproteobacteria and

Synergistia had consistently higher representation in the ACP, whereas

Methanobacteria, Gammaproteobacteria and Clostridia were more numerous in the OCP (Table 12; Figures 30 and 31).

4.3.3.2. Order and genus levels

The most abundant genera of the OCP and the ACP are presented in

Figures 32 and 33. Three genera with the highest combined representation in the OCP and the ACP were all methanogens (Table 13: #27 Methanosaeta, #12

Methanobacterium and #8 Methanolobus). Of these, Methanosaeta was equally represented in two communities, whereas Methanobacterium and

Methanolobus had higher representation in the OCP. Two other methanogens of the order Methanosarcinales, #1 Methanosarcina and #6

Methanomethylovorans were more abundant in the OCP by 77- and 11-fold respectively. In contrast, methanogens of the order Methanomicrobiales, namely genera Methanofollis, Methanocalculus and Methanoculleus, had higher representation in the ACP (Table 13: #30, #44 and #45). Methanospirillum belonging to the order Methanomicrobiales was equally represented in the OCP and the ACP (Table 13: #23). Hence, there appeared to be a distinct distribution of methanogens with the order Methanosarcinales being predominantly oil-attached, except the acetotrophic methanogen Methanosaeta, and the order Methanomicrobiales being more abundant in the aqueous phase.

Although the class Alphaproteobacteria was equally represented in the

OCP and ACP (Table 12), large differences were seen in distributions at the genus level. Four genera were more abundant in the OCP (Table 13: #2

Table 13. Numbers and fractions of reads for genera found in the OCP and the ACP. The list is ranked according to the ratio of fraction of total reads in the OCP to fraction of total reads in the ACP (OCP/ACP). Genera overrepresented in the OCP (OCP/ACP>2) or in the ACP (OCP/ACP<0.5) are in bold and in italics, respectively. OCP+ACP shows the sum of fractions of total reads for the OCP and the ACP. Genera with OCP+ACP equalling less than 0.1% are not shown.

OCP: ACP: OCP: ACP: fraction of fraction of # Phylum;Class;Order;Genus number of number OCP+ACP OCP/ACP total reads total reads of reads (%) reads (%)

1 Euryarchaeota;Methanomicrobia;Methanosarcinales;Methanosarcina 68 0.15 1 0.00 0.16 62.72 2 Proteobacteria;Alphaproteobacteria;Sphingomonadales;Sphingomonas 80 0.18 2 0.00 0.19 36.90 3 Proteobacteria;Gammaproteobacteria;Oceanospirillales;Thalassolituus 1863 4.22 67 0.16 4.39 25.65 4 Proteobacteria;Betaproteobacteria;Rhodocyclales;Azovibrio 203 0.46 14 0.03 0.49 13.37 5 Actinobacteria;Actinobacteria;Actinomycetales;Propionicicella 1151 2.61 91 0.22 2.83 11.67 6 Euryarchaeota;Methanomicrobia;Methanosarcinales;Methanomethylovorans 252 0.57 22 0.05 0.63 10.57 7 Firmicutes;Clostridia;Clostridiales;Acetobacterium 1458 3.31 173 0.43 3.73 7.77 8 Euryarchaeota;Methanomicrobia;Methanosarcinales;Methanolobus 5878 13.33 1068 2.63 15.95 5.08 9 Tenericutes;Mollicutes;Acholeplasmatales;Acholeplasma 305 0.69 56 0.14 0.83 5.02 10 Actinobacteria;Actinobacteria;Actinomycetales;Rhodococcus 63 0.14 13 0.03 0.17 4.47 11 Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodovulum 2990 6.78 687 1.69 8.47 4.01 12 Euryarchaeota;Methanobacteria;Methanobacteriales;Methanobacterium 6106 13.84 2259 5.55 19.40 2.49 13 Deferribacteres;Deferribacterales;Deferribacteraceae;Calditerrivibrio 92 0.21 37 0.09 0.30 2.29 14 Proteobacteria;Alphaproteobacteria;Caulobacterales;Brevundimonas 532 1.21 214 0.53 1.73 2.29 15 Proteobacteria;Alphaproteobacteria;Rhodobacterales;Stappia 2265 5.14 988 2.43 7.56 2.11

(continued on the next two pages)

100

Table 13. Numbers and fractions of reads for genera found in the OCP and the ACP (continued).

16 Proteobacteria;Alphaproteobacteria;Rhizobiales;Kaistia 239 0.54 129 0.32 0.86 1.71 17 Proteobacteria;Gammaproteobacteria;Xanthomonadales;Pseudoxanthomonas 137 0.31 91 0.22 0.53 1.39 18 Bacteroidetes;Marinifilum 91 0.21 62 0.15 0.36 1.35 19 Proteobacteria;Alphaproteobacteria;Rhizobiales;Hyphomicrobium 333 0.76 234 0.58 1.33 1.31 20 Chloroflexi;Anaerolineae;Anaerolineales;Leptolinea 180 0.41 165 0.41 0.81 1.01 21 Bacteroidetes;Bacteroidia;Bacteroidales;Proteiniphilum 295 0.67 285 0.70 1.37 0.95 22 Proteobacteria;Alphaproteobacteria;Rhizobiales;Rhizobium 1964 4.45 1972 4.85 9.30 0.92 23 Euryarchaeota;Methanomicrobia;Methanomicrobiales;Methanospirillum 86 0.19 87 0.21 0.41 0.91 24 Thermotogae;Thermotogae;Thermotogales;Kosmotoga 151 0.34 160 0.39 0.74 0.87 25 Proteobacteria;Gammaproteobacteria;Pseudomonadales;Azomonas 216 0.49 262 0.64 1.13 0.76 26 Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodobacter 2633 5.97 3588 8.82 14.79 0.68 27 Euryarchaeota;Methanomicrobia;Methanosarcinales;Methanosaeta 5158 11.70 7042 17.31 29.01 0.68 28 Deferribacteres;Deferribacteres;Deferribacterales;Geovibrio 348 0.79 532 1.31 2.10 0.60 29 Actinobacteria;Actinobacteria;Actinomycetales;Georgenia 33 0.07 57 0.14 0.21 0.53 30 Euryarchaeota;Methanomicrobia;Methanomicrobiales;Methanofollis 70 0.16 127 0.31 0.47 0.51 31 Bacteroidetes;Bacteroidia;Bacteroidales;Petrimonas 152 0.34 294 0.72 1.07 0.48 32 Proteobacteria;Alphaproteobacteria;Rhizobiales;Ensifer 29 0.07 66 0.16 0.23 0.41 33 Proteobacteria;Alphaproteobacteria;Caulobacterales;Hyphomonas 380 0.86 948 2.33 3.19 0.37 34 Proteobacteria;Deltaproteobacteria;Desulfovibrionales;Desulfovibrio 61 0.14 157 0.39 0.52 0.36 35 Proteobacteria;Deltaproteobacteria;Desulfuromonadales;Pelobacter 30 0.07 79 0.19 0.26 0.35

(continued on the next page)

101

Table 13. Numbers and fractions of reads for genera found in the OCP and the ACP (continued).

36 Proteobacteria;Deltaproteobacteria;Desulfuromonadales;Desulfuromonas 78 0.18 217 0.53 0.71 0.33 37 Proteobacteria;Alphaproteobacteria;Rhizobiales;Mesorhizobium 18 0.04 69 0.17 0.21 0.24 38 Proteobacteria;Deltaproteobacteria;Syntrophobacterales;Syntrophus 75 0.17 291 0.72 0.89 0.24 39 Proteobacteria;Epsilonproteobacteria;Campylobacterales;Arcobacter 36 0.08 148 0.36 0.45 0.22 40 Synergistetes;Synergistia;Synergistales;Thermovirga 55 0.12 243 0.60 0.72 0.21 41 Proteobacteria;Betaproteobacteria;Rhodocyclales;Azonexus 22 0.05 113 0.28 0.33 0.18 42 Proteobacteria;Alphaproteobacteria;Rhodospirillales;Oceanibaculum 50 0.11 261 0.64 0.75 0.18 43 Proteobacteria;Alphaproteobacteria;Rhodobacterales;Donghicola 93 0.21 577 1.42 1.63 0.15 44 Euryarchaeota;Methanomicrobia;Methanomicrobiales;Methanocalculus 43 0.10 267 0.66 0.75 0.15 45 Euryarchaeota;Methanomicrobia;Methanomicrobiales;Methanoculleus 165 0.37 1398 3.44 3.81 0.11 46 Proteobacteria;Alphaproteobacteria;Rhodospirillales;Tistrella 23 0.05 270 0.66 0.72 0.08 47 Proteobacteria;Deltaproteobacteria;Desulfobacterales;Desulfobulbus 74 0.17 873 2.15 2.31 0.08 48 Proteobacteria;Deltaproteobacteria;Desulfovibrionales;Desulfomicrobium 18 0.04 216 0.53 0.57 0.08 49 Proteobacteria;Deltaproteobacteria;Desulfobacterales;Desulfocapsa 5 0.01 76 0.19 0.20 0.06 50 Proteobacteria;Alphaproteobacteria;Sphingomonadales;Sphingobium 11 0.02 175 0.43 0.46 0.06 51 Proteobacteria;Deltaproteobacteria;Desulfuromonadales;Geobacter 38 0.09 2900 7.13 7.21 0.01 Others 7407 16.79 10557 25.95 All entries 44103 100 40680 100

102

Sphingomonas, #11 Rhodovulum, #14 Brevundimonas and #15 Stappia), seven genera had higher representation in the ACP (Table 13: #50

Sphingobium; #46 Tistrella; #43 Donghicola; #42 Oceanibaculum; #37

Mesorhizobium; #33 Hyphomonas; and #32 Ensifer), and four genera were equally represented in the OCP and the ACP (Table 13: #16 Kaistia, #19

Hyphomicrobium, #22 Rhizobium, and #26 Rhodobacter). Likewise,

Betaproteobacteria were either oil-associated (Table 13: #4 Azovibrio) or mostly associated with the aqueous phase (Table 13: #41 Azonexus). Thalassolituus

(Table 13: #3) was the most prominent oil-associated representative of the class Gammaproteobacteria. The genera Pseudoxanthomonas and Azomonas

(Table 13: #17 and #25) were equally represented in the two communities.

Remarkably, all of eight genera identified within the Deltaproteobacteria preferred the aqueous phase (Table 13: #51 Geobacter; #49 Desulfocapsa; #48

Desulfomicrobium; #47 Desulfobulbus; #38 Syntrophus; #36 Desulfuromonas;

#35 Pelobacter; and #34 Desulfovibrio). Of these, Geobacter had the least affinity to oil, being 80-fold more abundant in the ACP than in the OCP. The epsilonproteobacterium Arcobacter was also predominantly associated with the aqueous phase (Table 13: #39).

The oil-association of the phyla Firmicutes and Tenericutes (Table 11) was largely because of members of the genus Acetobacterium and

Acholeplasma (Table 13: #7 and #9, respectively). Likewise, the preference of the phylum Synergistetes for the aqueous phase was mostly due to members of the genus Thermovirga (Table 13: #40). Among Actinobacteria, Propionicicella was one of the most highly represented genera in the OCP (Table 13: #5);

Rhodococcus (Table 13: #10) was also oil-associated; other Actinobacteria

103

20000 18000 16000

14000 12000 10000 8000 6000

number of reads of number 4000 2000 0

Figure 28. Phyla found in the OCP.

25000

20000

15000

10000 number of reads of number

5000

Figure 29. Phyla found in the ACP.

104

16000

14000

12000 10000 8000 6000

number of reads of number 4000 2000 0

Figure 30. The most abundant classes found in the OCP.

16000

14000 12000 10000 8000

6000 number of reads of number 4000 2000 0

Figure 31. The most abundant classes found in the ACP.

105

7000

6000 5000 4000 3000

2000 number of reads of number 1000

Kaistia

Stappia

Azovibrio

Geovibrio

Azomonas

Leptolinea

Rhizobium

Petrimonas

Rhodovulum

Rhodobacter

Hyphomonas

Acholeplasma

Thalassolituus

Methanosaeta

Propionicicella

Methanolobus

Proteiniphilum

Brevundimonas

Acetobacterium

Methanoculleus

Hyphomicrobium

Methanobacterium Methanomethylovorans

Figure 32. The most abundant genera found in the OCP.

8000

7000

6000

5000

4000

number of reads of number 3000

2000

1000

Stappia

Tistrella

Geovibrio

Azomonas

Geobacter

Rhizobium

Donghicola

Syntrophus

Petrimonas

Thermovirga

Rhodovulum

Rhodobacter

Hyphomonas

Desulfobulbus

Methanosaeta

Methanolobus

Proteiniphilum

Oceanibaculum

Methanoculleus

Desulfuromonas

Methanocalculus Hyphomicrobium Methanobacterium

Figure 33. The most abundant genera found in the ACP.

106 identified at the genus level (e.g. #29 Georgenia, Table 13) had about equal representation in the two communities.

4.4. Discussion

4.4.1. Microorganisms associated with the oil phase

It may be expected that microorganisms, which need to reside at an oil- water interface, would include, first of all, primary oil degraders, as their substrates are insoluble or poorly soluble in water. Secondly, hydrogenotrophs may need to stay adhered to an oil phase as well since H2 has higher solubility in oil than in water.

4.4.1.1. Gammaproteobacteria

Thalassolituus, the major gammaproteobacterium in the MHGC field, was found to be clearly oil-associated. Thalassolituus oleivorans is an obligate oil-degrading bacterium capable of utilizing water-insoluble alkanes (Yakimov et al., 2010). This species was previously found in crude oil-containing marine environments and characterized as an aerobic chemoorganoheterotrophic organism unable to grow by fermentation or nitrate reduction (Yakimov et al.,

2010). Finding of this bacterium in an oil reservoir may be an indication of the availability of oxygen in it (e.g. oxygen dissolved in the injected water). Another possibility is that some species of this genus might metabolize oil anaerobically.

4.4.1.2. Alphaproteobacteria

The four oil-associated genera Sphingomonas, Rhodovulum

Brevundimonas and Stappia have all been shown to degrade hydrocarbons.

107

Sphingomonas is a well-known degrader of polyaromatic hydrocarbons.

Although generally described as aerobic, it has been previously identified in anaerobic oil-degrading consortia (Zhang et al., 2011). Oil-degrading species of

Rhodovulum were isolated recently by Teramoto et al. (2010), whereas species of Brevundimonas and Stappia have also been characterized as oil-degraders

(Al-Awadhi et al., 2007; Chaîneau et al., 1999).

4.4.1.3. Betaproteobacteria

Azovibrio, found in this study to be associated with the oil phase, is closely related to Azoarcus (and also related to Thauera). Moreover, Azovibrio was formerly included in the genus Azoarcus sensu lato (Reinhold-Hurek and

Hurek, 2006). As it was discussed in section 3.4, Azoarcus is capable of heptane degradation. Some of its species are also known to degrade aromatic hydrocarbons under denitrifying conditions (Khomenkov et al., 2005; Chen and

Hickey, 2011; Jimenez et al., 2012; Reinhold-Hurek and Hurek, 2006), suggesting that Azovibrio may possibly also be able to degrade oil. It cannot be ruled out that Azovibrio and Azoarcus from the MHGC oil field found in the experiments described here and in Chapter 3 respectively are, in fact, the same bacterium, which was assigned different “best match” identities by Phoenix 2

(sections 3.2.13 and 4.2.5) at different time points (identification of Azoarcus and other heptane consumers was conducted in about two years after Azovibrio and other oil-associated bacteria were identified). Previously reported association of Azoarcus with the oil phase (Rabus et al., 1999), along with the fact that Azovibrio and Azoarcus are so closely related to each other, suggest that Azoarcus (section 3.4; Table 6) must also be adhering to EO.

108

4.4.1.4. Actinobacteria

Some of the representatives of this phylum/class (for example, some species of Rhodococcus) have been characterized as degraders of organic compounds that are poorly-soluble in water (Andreoni et al., 2000), and the finding that most Actinobacteria present in the MHGC field are oil-associated corresponds well with this information. Propionicicella was the most numerous genus of Actinobacteria in the OCP, but it was not abundant in the ACP.

Interestingly, Propionicimonas, closely related to Propionicicella, was found in the production water from another Canadian oil reservoir (Grabowski et al.,

2005a). Propionicimonas paludicola has been described as a fermenting bacterium (Akasaka et al., 2003) whose presence in the oil reservoir production water/oil mixture was not easy to explain (Grabowski et al., 2005a).

Propionicicella superfundia was found to be a chlorosolvent-tolerant, propionate-forming, facultative anaerobic bacterium (Bae et al., 2006). Unable to directly biotransform chlorinated solvents, it nevertheless has been thought to play a role in their biodegradation (Bae et al., 2006). It is not likely that chlorosolvents have been present in the produced water-oil mixture. However the apparent hydrophobicity of the cells of this bacterium may assist it in consuming hydrocarbons insoluble in water.

4.4.1.5. Mollicutes and Clostridia

Acholeplasma was one of the ten most abundant genera in the OCP. It is known to be a carbohydrate fermenter (Green et al., 2008). Somewhat unexpected in an oil field, it was previously found in a similar environment, a coalbed methane well (Green et al., 2008). Its association with the oil phase

109 indicates that its species inhabiting the oil field must be able to use hydrocarbons (or hydrogen, which seems to be less likely) as electron donors.

The hydrogenotrophic acetogen Acetobacterium was also strongly oil- associated (Table 13).

4.4.2. Positioning of methanogens

Oil in the MHGC field, from which the samples used in this study were taken, is produced by water injection. The injected water contains a low concentration of sulfate (1 mM) and nitrate (2 mM). The latter is added to prevent sulfide production by sulfate reducing bacteria or oxidize already present sulfide. As indicated elsewhere (Voordouw et al., 2009), both injected electron acceptors are reduced completely along the flow path. As a result, produced waters contain no nitrate or nitrite, no sulfate and on average 0.5 mM sulfide (Voordouw et al., 2009). The predominance of methanogens in the microbial community of produced waters (Table 13) is in agreement with the previously noted depletion of electron acceptors and indicates that oil degradation coupled to methanogenesis may be a primary degradative mechanism in the vicinity of producing wells.

Methanogenic oil degradation depends on the activities of: (i) syntrophic bacteria hydrolyzing hydrocarbons to H2, CO2 and acetate, (ii) hydrogenotrophic methanogens converting H2 and CO2 to methane and (iii) acetotrophic methanogens converting acetate to methane and CO2 (Zengler et al., 1999;

Head et al., 2003; Gieg et al., 2008). Syntrophic hydrocarbon degradation requires effective removal of hydrogen and acetate to drive this thermodynamically uphill reaction (Zengler et al., 1999). The

110 deltaproteobacterium Syntrophus, identified by Zengler et al. (1999) as a primary degrader of a methanogenic hexadecane-degrading consortium, was found in this study. Representatives of the phyla Deltaproteobacteria,

Firmicutes, Actinobacteria, Chloroflexi and Bacteroidetes, reported by Gieg et al. (2008) to belong to an oil-consuming methanogenic consortium, were found as well (Tables 11 and 13).

H2 generated during anaerobic, methanogenic oil degradation or other anaerobic processes will distribute over both oil and aqueous phases. However, it is more soluble in oil than in water at 30oC (Cai et al., 2001). Association of obligate hydrogenotrophs, such as hydrogenotrophic methanogens with the oil phase may, therefore, be expected. Indeed, two of the three most abundant genera were oil-associated hydrogenotrophic methanogens (Table 13: #8

Methanolobus and #12 Methanobacterium). The acetotrophic methanogen

Methanosaeta, one of the most abundant genera, does not need to be oil- associated, because acetate is more soluble in water than in oil.

Methanosarcina species consume both acetate and H2 (Beckman et al., 2011).

However, strong association of this species with oil (Table 13) indicates that hydrogen may be its preferred substrate in the MHGC field. Species of

Methanomethylovorans produce methane from dimethylsulfide, methanethiol and methylamine (Lomans et al., 1999), compounds that dissolve in oil rather than water, explaining the association of this genus with oil (Table 13). The hydrogenotrophic methanogens Methanocalculus and Methanoculleus require acetate as a carbon source (Lai et al., 2002; Lai et al., 2004; Mikucki et al.,

2003); this may explain their preferential association with the aqueous phase.

111

4.4.3. Other microorganisms associated with the aqueous phase

4.4.3.1. Deltaproteobacteria

DNA pyrosequencing analysis showed that all eight identified genera of the Deltaproteobacteria were associated with the aqueous phase (Table 13), in agreement with the fact that most of them (e.g. Geobacter, Coates et al., 2001;

Desulfovibrio, Park et al., 2008) use water-soluble organic acids like acetate as electron donors for sulfate or Fe(III) reduction. One might expect the genus

Syntrophus to be rather oil-associated, as it is known to be a primary anaerobic degrader of hexadecane (Zengler et al., 1999). However, this hydrophilic bacterium may be able to produce biosurfactants, which would allow it to access a water-insoluble substrate through its solubilization.

4.4.3.2. Alphaproteobacteria

Of the seven genera, which preferred the aqueous phase (Sphingobium,

Tistrella, Donghicola, Oceanibaculum, Mesorhizobium, Hyphomonas and

Ensifer), some have been described as hydrocarbon degraders. Sphingobium species were reported to consume both water-insoluble aromatic hydrocarbons and such water-soluble substances as benzoate (Liang and Lloyd-Jones, 2010).

Being associated with the aqueous phase, Sphingobium may prefer consuming water-soluble organics in the MHGC oil field. Association of Tistrella with an aqueous phase indicates that its primary metabolic substrate(s) must be water- soluble; however, this bacterium was found to degrade phenanthrene when another bacterium, Sphingomonas was present (Zhao et al., 2008); if it indeed consumes water-insoluble substances in the MHGC field, then it must be able to produce biosurfactants for accessing them. The same conclusion can be

112 made regarding Oceanibaculum, some species of which have been characterized by Lai et al. (2009) and Dong et al. (2010). Donghicola has been thought to play a role in oil degradation. However, it could also use water- soluble acetate (Tan et al., 2009).

4.4.3.3. Bacteroidetes and Synergistetes

Petrimonas was found to have representation in the ACP about twice as large as in the OCP (Table 13: #31). The species Petrimonas sulfuriphila was shown by Grabowski et al. (2005b) to consume acetate and other organic acids.

The entire phylum of Bacteroidetes, to which Petrimonas belongs, was about one and half times more abundant in the ACP than the OCP (Table 11), showing that it is represented in the MHGC oil field largely by hydrophilic bacteria whose major metabolic substrates are, most likely, organic acids.

The clostridium Thermovirga lienii isolated by Dahle and Birkeland

(2006) from a North Sea oil well was shown to consume some organic acids, and this fact corresponds well to the finding that Thermovirga is a bacterium associated with the aqueous phase.

4.4.4. Representation of diazotrophs

Members of the order Rhizobiales, generally known to be diazotrophs, were found either to be nearly equally represented in the OC and the AC (Table

13: #16 Kaistia #19 Hyphomicrobium and #22 Rhizobium, the last genus being one of the most represented bacteria in both the OC and the AC) or associated with the aqueous phase (Table 13: #32 Ensifer and #37 Mesorhizobium).

Although ammonium was found in the MHGC field in submillimolar

113 concentrations, anammox combined with preferential denitrification of the injected nitrate (Cornish-Shartau et al., 2010) can lead to the lack of NH3, at least in some zones of the oil reservoir. Nitrogen is more soluble in oil than in water. However, it still has relatively high solubility in the aqueous phase (it is about ten times more soluble in water than hydrogen). Therefore, whether they are attached to oil or water, Rhizobiales, together with some other bacteria (e.g.

Azovibrio and Azonexus), indeed, may convert N2 produced by nitrate-reducing bacteria to NH3. Interestingly, several recent publications indicated the ability of some species of Rhizobium to degrade polyaromatic hydrocarbons (Kaiya et al.,

2012; Wen et al., 2011; Zhang et al., 2012).

4.5. Conclusions and implications

It was shown that a distinct microbial community is attached to oil. Many primary hydrocarbon degraders, especially among the Gamma- and

Alphaproteobacteria, were found to reside mainly at an oil-water interface.

Hydrogenotrophs also appeared to associate with oil, suggesting that the oil phase is a source of hydrogen. Although oil degradation in the MHGC field is thought to be anaerobic, it is worthwhile to note with respect to aerobic hydrocarbon degradation, that O2, like hydrogen, is more soluble in oil than in water, providing an additional incentive for some aerobic hydrocarbon degraders to associate with oil.

The composition of oil has been reported to influence microbial adhesion to it. In particular, the presence of asphaltenes and resins may decrease affinity of microbial cells to hydrocarbons, whereas the presence of toluene may lead to its increase (Zoueki et al., 2010). It was shown in Chapter 3 that LO was much

114 poorer in light hydrocarbons and particularly in toluene than EO (Figure 10); at the same time, gas-driven recovery of LO was found to be enhanced through stimulation of indigenous oil field microflora with nitrate to a much lesser extent than the recovery of EO (Table 1). Based on the information provided by Zoueki et al. (2010), it is reasonable to think that this was the case because fewer microbes might be residing at a LO-water interface than at an EO-water interface.

115

Chapter 5: Native and recombinant

rhamnolipid producers as MEOR-agents

5.1. Introduction

For decades, chemically synthesized surfactants have been used for

EOR and for oil spill clean-up. However, because of their toxicity and resistance to degradation, they can cause serious environmental problems (Mulligan,

2005). Lower toxicity and higher biodegradability of biosurfactants, compared to synthetic chemical surfactants (Mohan et al., 2006; Rosenberg and Ron, 1999), make them an attractive alternative. The high biodegradability of biosurfactants may also be considered a disadvantage because it may pose stringent limits to the time during which their surface activity can be retained.

Rhamnolipids are biosurfactants capable of reducing oil-water IFT by up to one order of magnitude (Wang et al., 2007). Although IFT reduction by two to three orders of magnitude is generally required for displacement of a significant fraction of residual oil (Kowalewski et al., 2006; Gray et al., 2008), they nevertheless were shown to have potential for EOR and bioremediation applications (Lang and Wullbrandt, 1999; Wang et al., 2007; Cha et al., 2008).

Their capability of increasing adhesion of some microorganisms to hydrocarbons and, thus, enhancing hydrocarbon biodegradation (Zhang and

Miller, 1994) may add value to their use.

Pseudomonas aeruginosa is a well-known rhamnolipid producer.

However, being an opportunistic pathogen (Cha et al., 2008), it can hardly be used for biotechnological purposes. Non-pathogenic recombinant microorganisms may be used instead of P. aeruginosa (Ochsner et al., 1995;

116

Lang and Wullbrandt, 1999; Perfumo et al., 2006; Wang et al., 2007; Cha et al.,

2008), although there are issues with their use as well. First, it is difficult to make industrial ex situ rhamnolipid production economically viable because sufficient biosurfactant yields are not easily achievable (Mukherjee et al., 2006).

On the other hand, economic evaluation of many MEOR projects involving microbial biosurfactant production (Portwood, 1995), provides grounds for optimism in this respect. Second, production of biosurfactants in situ by recombinant microorganisms is often deemed unrealistic because of the opposing public opinion and legislation, which may prohibit release of such organisms to the environment (Urgun-Demirtas et al., 2006), despite the fact that environmental harm has never been shown to be caused by genetically modified organisms (De Lorenzo, 2010).

Ochsner et al. (1995) reported rhamnolipid production in recombinant pseudomonads, but not in a recombinant E. coli, despite the fact that it was capable of synthesizing an active rhamnosyltransferase and known to form the precursor substrate, dTDP-rhamnose (Glaser and Kornfeld, 1966). On the other hand, according to Cabrera-Valladares et al. (2006), the availability of dTDP- rhamnose restricts the production of rhamnolipids in E. coli. Although another important precursor, 3-hydroxydecanoyl-3-hydroxydecanoate might be missing in E. coli (Ochsner et al., 1995), alternative fatty acid(s) could potentially be used by this bacterium to synthesize rhamnolipids. Wang et al. (2007) used transposome-mediated chromosome integration to introduce P. aeruginosa

PAO1 rhlAB genes into P. aeruginosa PAO1-rhlA- and E. coli BL 21 to achieve mono-rhamnolipid production in recombinant strains. Results obtained by

Cabrera-Valladares et al. (2006) indicate that RhlA is responsible for

117 biosynthesis of 3-(3-hydroxyalkanoyloxy)alkanoic acids and the fatty acid dimer moieties of rhamnolipids. According to Ochsner et al. (1994), RhlB is responsible for the coupling of the activated sugar (dTDP-rhamnose) to a hydroxyfatty acid molecule. Similarly to results obtained by Ochsner et al.

(1995) and Wang et al. (2007), Cha et al. (2008) achieved mono-rhamnolipid production in a recombinant P. putida, which received the P. aeruginosa EMS1 rhlAB genes together with the rhlRI quorum sensing system regulating gene expression in response to fluctuations in cell-population density (Ochsner and

Reiser, 1995; Miller and Bassler, 2001).

Wang et al. (2007) characterized the capability of a purified rhamnolipid solution to reduce oil-water IFT and studied oil recovery from a sandpack when the same solution was applied. The EOR effects of supernatants from the cultures of rhamnolipid-producers were not reported by Wang et al. (2007).

However, the effects of such supernatants may be of interest for industrial purposes. Cha et al. (2008) determined CMC of cell free supernatants containing rhamnolipids and studied their emulsifying properties. However, the capability of supernatants to release trapped oil was not investigated.

In the current study, recombinant E. coli carrying rhlAB genes on plasmid pNOT19 was constructed. Model columns (section 3.2.3) were used to compare

EOR effects of culture supernatants (CSs) of the recombinant E. coli, two strains of P. aeruginosa and chemically synthesized surfactants.

Results presented in this chapter are currently in press (Kryachko et al.,

2012b). Construction of the recombinant E. coli strain was done in collaboration with Dr. Gerrit Voordouw and Johanna Voordouw; rhamnolipid extraction and surface pressure measurements were conducted in collaboration with Dr. Elmar

118

J. Prenner, Patrick Lai and Safia Nathoo; TLC analysis was conducted in collaboration with Patrick Lai; hexadecane emulsification test and monitoring of oil and water production from model columns were performed in collaboration with Safia Nathoo.

5.2. Materials and methods

5.2.1. Constructing recombinant E. coli and isolating CSs

To develop a recombinant E. coli strain able to produce rhamnolipids,

Pseudomonas aeruginosa PA14 DNA was isolated with Marmur’s (1961) method. The rhlAB genes encoding rhamnosyltransferase 1, which is responsible for production of mono-rhamnolipids (Lang and Wullbrandt, 1999), were PCR-amplified with the primers RhlAB-1b-Hind

(agttaagcttcATGCGGCGCGAAAGTCTGTTGG, forward primer) and RhlAB-R4

(gactggtaccTTCCAGGACGGCGAACACGC, reverse primer). The amplified 5.1 kbp rhlAB fragment was inserted into plasmid pNOT19 carrying the ampicillin resistance gene. The plasmid was digested at HindIII and KpnI restriction sites using respective restriction enzymes (New England Biolabs) and ligated with the rhlAB fragment to produce recombinant plasmid pF1bR4 (Figure 34). Upon correct insertion, the target gene would be expressed constitutively from the pNOT19 lac promoter. Plasmid pF1bR4 was used for E. coli TG2 transformation to obtain recombinant strain E. coli pF1bR4. Ca2+-dependent transformation of

E. coli was performed according to Inoue et al. (1990). Selected colonies were cultured into 5 mL of TY medium (10 g/L of Bactotryptone, Becton, Dickinson &

Co; 5 g/L of yeast extract, Becton, Dickinson & Co; 5 g/L of NaCl; pH 7.4)

119

BlpI

RhlB RhlA

Figure 34. Recombinant plasmid pF1bR4.

120

Size (base pairs) ______

23130

9416

6557

4361

2322 2027

564

A B C D E F

Figure 35. Agarose-gel-electrophoresis analysis of digesting pF1bR4 plasmid with restriction enzymes: A – marker (λ/HindIII); (B) – undigested plasmid; C – plasmid digested with BlpI (digesting only RhlB; see Figure 34; this band is below/ahead of the band in the lane B indicating plasmid linearization); D – plasmid digested with BlpI and HindIII (presence of the lower band indicates presence of RhlAB on the plasmid; see Figure 34); E – plasmid digested with BlpI and KpnI (presence of the lower band indicates presence of RhlAB on the plasmid; see Figure 34); F – plasmid digested with KpnI (leading to plasmid linearization).

121 containing 0.1 mg/mL of ampicillin. Plasmid DNA was isolated using a QIAprep

Spin Miniprep Kit (QIAGEN) and digested with HindIII, KpnI and BlpI restriction enzymes (New England Biolabs) to ensure the presence of pF1bR4. These digestions produced DNA bands of the expected sizes (Figure 35). The presence of the right insert was also proven through Sanger sequencing of pF1b2R4 with the universal primers M13F and M13R-17, which was conducted at the UCDNA Services at the Faculty of Medicine of the University of Calgary.

After sequence verification, E. coli pF1bR4 was stored as a glycerol stock at minus 80ºC. The wild-type rhamnolipid-producers P. aeruginosa PA14 and P. aeruginosa PDO 111 (the last one being a rhlR mutant with the switched off rhl quorum sensing system, but still containing all genes necessary for rhamnolipid production; Brint and Ohman, 1995), E. coli pF1bR4 and a reference strain E. coli TG2 unable to produce rhamnolipids were grown for 24 hours in TY medium at 37°C in the incubator-shaker (New Brunswick Scientific Inc., series

25) and centrifuged at 3000 rpm (1500 x g) for 12 minutes to produce CSs.

5.2.2. Rhamnolipid extraction

Lipids in CSs were precipitated by adjusting the pH to 2 using 4 M H2SO4

(Fisher Scientific, Canada) and incubating overnight. The pellet was washed three times by centrifugation at 2000 x g for 20 minutes, followed by resuspension in 0.05 M Na2CO3 (Fisher Scientific, Canada), pH 8.62. The final pellet was resuspended in sodium bicarbonate (Fisher Scientific, Canada), pH

8. A mixture of chloroform:methanol (2:1, v/v) was used to extract the lipids. The organic solutions were pooled and added to pre-weighed vials before the mixtures were dried down under argon gas flow. The film masses were

122 determined through measuring differences between the masses of the empty vials and the vials with the films, followed by reconstituting the films to 1 mg/mL solutions in chloroform:methanol (7:3, v/v). These solutions were subsequently used for surface pressure measurements in a Langmuir trough (details are provided in section 5.2.4).

5.2.3. MALDI-TOF analysis

MALDI-TOF mass spectra were recorded using an AB Sciex MALDI-

TOF/TOF 5800 system operating in reflectron mode. 2,5-Dihydrobenzoic acid dissolved in 90% methanol and 10% water was used as the matrix. Samples, including a 1 mg/mL commercial rhamnolipid solution (95% pure R-95 HPLC-

MS grade, Sigma-Aldrich) as a standard, were dissolved in chloroform:methanol

(7:3, v/v). The matrix was spotted on the plate in the amount of 0.5 µL and allowed to dry at room temperature; then 0.5 µL of a sample was applied and the spot was again allowed to dry. Acquisitions of 1200 shots were summed.

M/Z software (Proteometrics, LLC) was used for analyzing mass-spectra.

5.2.4. Surface pressure measurements

Surface pressure – area isotherms were obtained using a 20 x 10 cm

Langmuir trough with a subphase volume of 110 mL (KSV Nima, UK); 20 µL of a 1 mg/mL solution was deposited on a sub-phase of high purity water. Film compression was conducted at 96 cm2/min using a single Teflon barrier; surface pressure was recorded using a Wilhelmy surface balance (KSV Nima, UK).

Surface pressure and area readings were compiled by a control unit

(microprocessor interface IU4) and Nima 7.2 software (KSV Nima, UK). The

123 molecular weight of the rhamnolipids was taken as 577 g/mol, an average for the mono- and di-rhamnolipids (Torrens et al., 1998; Özdemir and Malayoglu,

2004). All trials were conducted at least in triplicates. All isotherms were highly reproducible; differences between replicates did not exceed 3%.

5.2.5. TLC analysis

TLC analysis was used to separate and visualize mono- and di- rhamnolipids from the CS extracts and the commercial rhamnolipid solution. A

TLC plate was initially placed in a solvent mixture of chloroform:methanol:water

(65:15:2, v/v/v) to remove any contaminating material and then was allowed to dry. Lipid extracts and the commercial rhamnolipid in chloroform:methanol (7:3, v/v) were spotted onto the silica gel plate 60A KC18F (Whatman, USA) using capillary tubes. To visualize the plate, the developing agent anthrone (Sigma

Aldrich) was used. Anthrone in the amount of 0.1 g was dissolved in 0.5 mL of

18 M H2SO4, diluted with 9.5 mL of pure ethanol (Priya and Usharani, 2009) and sprayed onto the plate before it was heated at 85°C to visualize the spots.

5.2.6. Hexadecane emulsification test

Equal amounts of hexadecane (Sigma Aldrich Inc., St. Louis, MO, USA) and CS (3 ml each) were added to a 10 mL glass test tube. The mixture was agitated with a Vortex mixer (Baxter Diagnostics Inc., Deerfield, IL, USA) for 2 minutes and allowed to stand for 24 hours. Then the stability and height of the emulsified phase were examined.

124

5.2.7. Monitoring oil and water production from model columns

5.2.7.1. Oil samples

Experiments were conducted with two samples of heavy oil, EO and LO

(the same as those mentioned in section 3.2.1), EO being rich in alkanes and

LO being rich in polyaromatic hydrocarbons.

5.2.7.2. Column assembly

Barrels of 30 mL polypropylene syringes (Becton, Dickinson and

Company or Henke Sass Wolf) were used for making the columns. The columns were assembled in the same way as described in section 3.2.3. The average pore volume (PV) of a column was 13.4±0.3 mL, the porosity was

40.5±0.5%, and the absolute permeability of a column was 30 D.

5.2.7.3. Saturating columns with SPW

This was conducted as described in section 3.2.4.

5.2.7.4. Saturating columns with oil

This was conducted according to the procedure described in section

3.2.5. Average residual water saturation (Swr) in the columns saturated with oil was found to be 1.2±0.18 mL (9% of PV). The oil-saturated columns were aged at 20ºC for seven to ten days to create wettability conditions maximally similar to those in an oil reservoir.

125

5.2.7.5. Eluting oil with water

This was conducted in accordance with the method described in section

3.2.6. Elution (referred to as a primary SPW flood in Table 15) was conducted at a rate of 6.5 mL per hour until essentially no more oil was produced. The average amount of EO trapped after primary SPW flood for the columns mentioned in Tables 15 and 16 was 3.07±0.74 mL (average original EO in place was 12.34±0.32 mL), and the average amount of LO trapped after primary SPW flood was 3.94±1.22 mL (average original LO in place was 12.09±1.33 mL).

5.2.7.6. Experiments with columns containing trapped oil

Seven columns (set 1) with trapped LO were used for short-time incubations with surfactants at 20°C. One PV of the following solutions was injected into columns #1–7: #1 – 1% Triton X-100; #2 – 1% CTAB (cetyl trimethyl ammonium bromide); #3 – 1% SDS (sodium dodecyl sulphate); #4 –

CS from P. aeruginosa PA14 (native rhamnolipid producer) culture; #5 – CS from E.coli pF1bR4 (recombinant bacterium carrying genes for mono- rhamnolipid production) culture; #6 – CS from P. aeruginosa PDO111 (having repressed transcription of rhamnolipid production genes) culture; and #7 – CS from E. coli TG2 (the reference strain unable to produce rhamnolipids) culture.

After a 24-hour incubation, each column was flooded with 3 PV of SPW with the

Mandel-Gilson Minipulse 3 peristaltic pump at a flow rate of 6 mL per hour (this flood is referred to as a secondary flood in Tables 15 and 16); mixtures of produced water and oil were collected into 50 mL polypropylene tubes and the volumes of oil and water were determined as described in section 3.2.8. The injection-incubation-elution cycle was repeated six times. After the seventh

126 injection, crimp-sealed 10 mL vials filled with helium or argon were attached to all the columns for collecting liquids and gases; then the columns were placed in an incubator with a temperature of 30°C located in an anaerobic hood with an atmosphere of 90% N2 and 10% CO2 to study the effects of long-time (79- to 88- day) incubations.

The following solutions, each in the amount of 1 PV, were injected in three columns containing trapped EO (set 2): 1% Triton X-100, 1% CTAB or CS from E.coli TG2 culture. After 24-hour (short-time) incubations, each column was flooded with 3 PV of SPW using a Mandel-Gilson Minipulse 3 peristaltic pump at a flow rate of 6 mL per hour (this flood is referred to as a secondary flood in Tables 15 and 16); mixtures of produced water and oil were collected into 50 mL polypropylene tubes and the volumes of oil and water were determined. The injection-incubation-elution cycle was repeated five times. After the sixth injection, crimp-sealed 10 mL vials filled with helium or argon were attached to all the columns; then the columns were placed in the anaerobic hood in an incubator with a temperature of 30°C to study the effects of the long- time incubations (“procedure 1” in Figure A4).

The experiments with surfactants were also conducted with four more columns containing trapped EO (set 3). These columns were aged for two months at 20°C before surfactant treatment (and, therefore, the recovery of oil from them was likely harder to achieve than from the columns belonging to the two other sets, which were not aged with trapped oil). Using a Mandel-Gilson

Minipulse 3 peristaltic pump, they were flooded at a flow rate of 12 mL/hour with

10 PV of CSs from P. aeruginosa PA14, E. coli pF1bR4 or P. aeruginosa

PDO111 cultures or 1% SDS (this flood is referred to as a secondary flood in

127

Tables 15 and 16). After the end of flooding, crimp-sealed 10 ml vials filled with helium or argon were attached to all the columns; then the columns were placed in the anaerobic hood in an incubator with a temperature of 30°C to study the effects of long-time incubations (“procedure 2” in Figure A4; Figure 49).

Following 39- to 45-day incubation, 10 mL vials were removed for measuring amounts of produced oil and water, 1 PV of the corresponding solution was injected into each column manually with a syringe (at the flow rate of about 15 mL/min) and new 10 mL vials filled with helium or argon were attached. Each column underwent two rounds of long-time incubations.

5.2.7.7. Determination of the amounts of produced oil and water

This was done as described in section 3.2.8.

5.2.7.8. Determination of the concentrations of gases produced during long-time incubations

Concentrations of gases (N2, CO2, CH4 and H2) produced into 10 mL vials during long-time incubations were measured as described in section

3.2.10.

5.3. Results

5.3.1. Properties of CSs

The highest amount of lipids was produced by P. aeruginosa PA14, followed by P. aeruginosa PDO111, E. coli pF1bR4 and E. coli TG2 (Table 14).

MALDI-TOF analysis demonstrated that P. aeruginosa PA14 CS extract contained previously described di- and mono-rhamnolipids (Figure 37; Price et

128 al., 2009; Nie et al., 2010); P. aeruginosa PDO111 and E. coli pF1bR4 CS extracts did not contain known rhamnolipids, but both of them contained an ion with the m/z (molecular weight to charge) ratio equaling 361.96 – 361.99, which matched a minor component of the commercial rhamnolipid (Figures 44 and

45). No ions matching known rhamnolipids or other components of the commercial rhamnolipid were found in the E. coli TG2 CS extract, which still contained 3.79 mg/L of organics (Table 14; Figures 40 and 46). MALDI-TOF mass-spectra for all CS extracts and the commercial rhamnolipid in m/z ranges where rhamnolipid peaks are expected to be found are shown in Figures 36 –

40 and Figures 42 – 46. TLC analysis also demonstrated the presence of di- and mono-rhamnolipids in the P. aeruginosa PA14 CS extract. However, rhamnolipids were not identifiable in any other samples subjected to TLC analysis (Figure 47).

CSs from P. aeruginosa PA14 and E. coli pF1bR4 emulsified hexadecane (the latter to a lesser extent than the former), whereas the supernatant obtained from E. coli TG2 (the reference strain) had hardly any emulsification properties (Figure 41).

Surface pressure isotherms (Figure 48) clarified some properties of the extracts. Higher surface pressure means a stronger surface tension lowering capability; the shape of an isotherm often reveals some other properties such as molecular packing of a sample of interest.

The rhamnolipid standard showed a gradual rise in pressure followed by an inflection at 25 mN/m, which was expected given that the standard was a mixture of both mono- and di-rhamnolipids. The maximum surface pressure obtained for the standard rhamnolipid solution was 38 mN/m (Figure 48).

129

Table 14. Lipid concentrations in CSs.

Volume of CS Mass of Calculated CS sample used for extracted concentration of extraction, mL lipids, mg lipids in CS, mg/L

P. aeruginosa PA14 200 2.58 12.90 P. aeruginosa PDO111 200 1.05 5.24 E. coli pF1bR4 200 1.04 5.19 E. coli TG2 (reference) 150 0.57 3.79

Figure 36. MALDI-TOF mass-spectrum for the commercial rhamnolipid solution in the m/z range where most rhamnolipid peaks are expected to be found.

130

Figure 37. MALDI-TOF mass-spectrum for P. aeruginosa PA14 CS extract in the m/z range where most rhamnolipid peaks are expected to be found.

Figure 38. MALDI-TOF mass-spectrum for E. coli pF1bR4 CS extract in the m/z range where most rhamnolipid peaks are expected to be found (no peaks for known rhamnolipids can be seen).

131

Figure 39. MALDI-TOF mass-spectrum for P. aeruginosa PDO111 CS extract in the m/z range where most rhamnolipid peaks are expected to be found (no peaks for known rhamnolipids can be seen).

Figure 40. MALDI-TOF mass-spectrum for E. coli TG2 CS extract in the m/z range where most rhamnolipid peaks are expected to be found (no peaks for known rhamnolipids can be seen).

132

Figure 41. Emulsification of hexadecane by P. aeruginosa PA14, E.coli pF1bR4 and E.coli TG2 CSs.

133

Figure 42. MALDI-TOF peak for an ion with the m/z ratio equaling 361.96 – 361.99 in the commercial rhamnolipid solution.

Figure 43. MALDI-TOF peak for an ion with the m/z ratio equaling 361.96 – 361.99 in the P. aeruginosa PA14 CS extract.

134

Figure 44. MALDI-TOF peak for an ion with the m/z ratio equaling 361.96 – 361.99 in the E. coli pF1bR4 CS extract.

Figure 45. MALDI-TOF peak for an ion with the m/z ratio equaling 361.96 – 361.99 in the P. aeruginosa PDO111 CS extract.

135

Figure 46. MALDI-TOF peak for an ion with the m/z ratio equaling 361.96 – 361.99 in the E. coli TG2 CS extract.

136

Figure 47. Results of the TLC analysis of P. aeruginosa PA14, E. coli F1bR4, P. aeruginosa PDO111 and E. coli TG2 CS extracts, and the commercial rhamnolipid solution (“Standard”); di- and mono-rhamnolipid bands are seen only in the P. aeruginosa PA14 CS extract and the commercial rhamnolipid solution (“Standard”) lanes.

137

Figure 48. Surface pressure as a function of the area per molecule: results of monolayer experiments completed with lipid extracts from P. aeruginosa PA14, P. aeruginosa PDO111, E. coli pF1bR4 and E. coli TG2 (reference) CSs, the commercial rhamnolipid solution (“Rhamnolipid Standard”) and TY medium.

138

The early onsets of the isotherms for the extracts from P. aeruginosa

PA14 and E. coli pF1bR4 CSs (Figure 48) suggested that the molecules occupied large areas and their packing in the monolayer was not tight. This was also indicated by the slow process of leveling off, during which some compounds were likely forced out of the monolayer. The isotherm for P. aeruginosa PA14 CS extract reached high surface pressure of 43 mN/m, while that from E. coli pF1bR4 CS reached only 34 mN/m. The E. coli pF1bR4 CS extract isotherm showed an extended plateau compared to the others; at smaller areas, it also showed a rapid increase in pressure.

In the case of P. aeruginosa PDO111 CS extract, the shape of the surface pressure isotherm showed a more rapid increase in pressure suggesting the presence of a tightly packed film; some material was likely squeezed out of the monolayer when smaller areas-per-molecule were reached

(this is shown by the plateau starting at 47 mN/m; Figure 48).

The maximal surface pressure of E. coli TG2 (reference strain) CS extract was not lower than that of P. aeruginosa PA14 CS extract. TY medium alone had some surface activity as well (Figure 48).

5.3.2. Recovery of trapped oil from sandpack columns

5.3.2.1. Recovery of oil during CS injections and floods

During several rounds of injections and floods, some oil was produced from all columns containing LO (set 1). Injections of P. aeruginosa PA14 CS,

1% Triton X-100, 1% SDS and E. coli pF1bR4 CS led to larger amounts of produced oil compared to the reference strain (E. coli TG2) CS. Application of

139

Figure 49. EO production from set 3 columns containing one of the three CSs or 1% SDS during long-time incubation. Discharged oil can be seen in the vials attached to columns with E. coli F1bR4 and P. aeruginosa PDO111 CSs at the end of a long-time incubation round; discharged water can also be seen in the vials attached to the other columns.

140

Table 15. Production of oil and water from columns containing trapped LO and EO (sets1, 2 and 3) under different conditions. For set 1 and 2 columns, 3 PV of SPW was used for each secondary flood (“procedure 1” in Figure A4) and 1 PV of CS or 1 PV of 1% surfactant solution (Triton X-100, CTAB or SDS) was used for each injection; long-time incubations started right after secondary floods. For set 3 columns, 10 PV of CS or 1% SDS solution was used for the secondary flood (“procedure 2” in Figure A4) and 1 PV of CS or 1 PV of 1% SDS was used for an injection preceding each long-time incubation round; long-time incubations started in two months after the end of secondary floods (following secondary floods, these columns were incubated for two months in an aerobic atmosphere at 20˚C; no effluents were produced during this process; then long-time incubations were conducted as described in the text). Volumes of trapped oil after primary and secondary floods are shown as well.

Produced and trapped Oil produced Trapped oil Oil produced liquids Trapped oil Primary during after secondary Oil produced during after primary SPW secondary floods and during long- CS/solution Water SPW floods flood floods and injections time incubation injections produced Set (Vot1) reco- injections between the between during long- very, between the floods rounds of long- time (OOIP- floods (Vot2) time incubation incubation Oil type Vot1)/ (mL) and injected solution (% of OOIP (% of (% of (% of (% of (mL) OOIP) (mL) OOIP) (mL) OOIP) (mL) OOIP) (mL) OOIP) LO - E.coli TG2 CS 1 4.27 35.9 0.64 0.38 3.1 3.89 32.7 0 0 0.03 0.3 0 (reference) 1 LO - 1% Triton X-100 5.38 43.3 0.57 1.40 11.2 3.98 32.0 0.12 1.0 0.77 6.2 0.40 1 LO - 1% CTAB 3.13 26.6 0.73 0.43 3.6 2.70 22.9 0 0 0.02 0.2 0.05 1 LO - 1% SDS 4.23 35.6 0.64 0.96 8.0 3.27 27.5 0 0 0.99 8.3 0 1 LO - PA 14 - CS 5.37 45.0 0.55 2.81 23.5 2.56 21.5 0.06 0.5 0.02 0.1 1.50 1 LO - E. coli pF1bR4 - CS 3.00 24.4 0.76 0.89 7.2 2.11 17.1 0.12 1.0 0.07 0.6 0.13 1 LO - PDO111 - CS 2.17 17.5 0.83 0.39 3.1 1.78 14.3 0 0 0.00 0 1.65 EO - E.coli TG2 CS 2 3.51 29.1 0.71 0.42 3.5 3.09 25.6 0.03 0.2 0.05 0.4 0 (reference) 2 EO - 1% Triton X-100 3.78 31.6 0.68 0.77 6.5 3.01 25.1 0.02 0.1 0.22 1.9 0.30 2 EO - 1% CTAB 4.13 34.3 0.66 0.50 4.1 3.63 30.1 0 0 0.02 0.2 0 3 EO - 1% SDS 2.94 24.0 0.76 0 0 2.94 24.0 0.04 0.3 0.02 0.1 1.22 3 EO - PA 14 - CS 2.43 19.1 0.81 0 0 2.43 19.1 0.03 0.2 0.02 0.1 1.80 3 EO - E. coli pF1bR4 - CS 2.37 18.7 0.81 0.01 0.1 2.36 18.6 0.23 1.8 0.01 0.1 0 3 EO - PDO111 - CS 2.31 18.2 0.82 0 0 2.31 18.2 0.52 4.1 0 0 1.60

141

P. aeruginosa PA14 CS led to production of 7.5 times more oil than in the case of the reference strain; when 1% Triton X-100 was used, 3.5 times more oil was produced than in the case of E. coli TG2 CS, and 1% SDS and E. coli pF1bR4 CS were about 2.5 times more effective than E. coli TG2 CS (Table

15). Injections of 1% CTAB and P. aeruginosa PDO111 CS led to similar amounts of produced oil as an injection of E. coli TG2 CS (Table 15).

Among three columns with EO, which were subjected to a number of rounds of injections and floods (set 2), the column with 1% Triton X-100 solution produced almost twice as much oil as did columns with 1% CTAB and E. coli

TG2 CS (Table 15).

Among four columns with EO, which were subjected to floods with CSs or chemical surfactant solutions instead of SPW (set 3), only the column that received E. coli pF1bR4 CS (Table 15) produced some oil during floods and solution injections between floods. However, its amount was almost negligible.

The other three columns from this set did not produce any oil during this process (Table 15). As it was indicated in section 5.2.7.6, all set 3 columns were aged for two months at 20°C before any surfactant treatment.

5.3.2.2. Recovery of oil during long-time incubations and media replenishments

Some oil was produced from all set 1 columns during rounds of long-time incubation (when no floods or injections were applied) or solution injections/replenishments between incubation rounds (Table 15). During replenishments of chemical surfactants, oil was produced in emulsified droplets, and during replenishments of CSs it was produced rather as a single phase.

142

During incubations, some LO was produced from set 1 columns with 1% Triton

X-100, E. coli pF1bR4 CS and P. aeruginosa PA14 CS. However, in these cases, the amounts of discharged LO did not exceed its amount that can be discharged from a column containing just CSB-K medium without nitrate (up to

0.3 mL during 112 days; see information on the column IV-2 in Table 1). During solution injections/replenishments between rounds of long-time incubation, the largest amount of LO was produced from the set 1 column with 1% SDS (Table

15). Set 2 columns (with EO) discharged minimal amounts of oil and water during long-time incubations (Table 15).

Among set 3 columns (with EO), one with P. aeruginosa PDO111 CS discharged the largest amount of oil during long-time incubations, followed by the columns with E. coli pF1bR4 CS, P. aeruginosa PA14 CS and 1% SDS. The column with P. aeruginosa PDO111 CS discharged two times more oil than the one with E. coli pF1bR4 CS and more than ten times larger amount of oil than the columns with P. aeruginosa PA14 CS and 1% SDS (Table 15). The amount of EO that can be produced during long-time incubations from a column containing just CSB-K medium without nitrate (up to 0.16 mL during 300-400 days; see information on columns I-3, II-3 and III-3 in Table 1) was exceeded three times by its amount discharged from the column with P. aeruginosa

PDO111 CS and also exceeded 1.5 times by its amount discharged from the column with E. coli pF1bR4 CS (it is also necessary to note that these columns were incubated for much shorter time than 300 days; Table 16).

143

5.3.3. Gas production during long-time incubations

All columns produced 1 to 4 mM of CO2 and 5 to 54 mM of N2 into 10 mL vials attached to the columns during single rounds of long-time incubations

(Table 16). Traces of methane and hydrogen were detected in most vials as well. Control vials with helium or argon not attached to columns contained only traces of N2 and CO2 at the ends of incubation rounds, showing that that there was no gas leakage through vial stoppers.

5.4. Discussion

5.4.1. Determinants of the surface activity of CSs

Growth-associated production of rhamnolipids is known to occur mostly during the late exponential phase of culture growth (Cha et al., 2008). This phase would be observed earlier if the rich TY medium is used instead of a minimal salt medium; for this reason, the cultures were grown for only 24 hours.

CSs were prepared at a lower centrifugation g-force than that used by

Wang et al. (2007) and Cha et al. (2008) to let some cells remain in CSs to account for the possibility that they influence an oil-water IFT and/or wettability and/or mechanical properties of oil-water interfaces (Kowalewski et al., 2006;

Kang et al., 2008a). This was necessary to model an oil reservoir where bacterial cells would inevitably surround an oil-water interface and to examine if the presence of rhamnolipids, in addition to cells and their parts, would benefit an oil recovery process.

MALDI-TOF analysis demonstrated the presence of rhamnolipids in the

P. aeruginosa PA14 CS extract (Figure 37). The E. coli pF1bR4 CS extract contained some components, which were absent in E. coli TG2 (reference

144 strain) CS extract, and at least one of them (m/z = 361.96 – 361.99) was found to be identical to a minor component of the commercial rhamnolipid; this molecule was also present in small amounts in P. aeruginosa PA14 and P. aeruginosa PDO111CSs (Figures 44 and 45). Therefore, the production of this molecule in E. coli pF1bR4 may be attributed to the activity of the product of rhlAB expression (rhamnosyltransferase 1).

The surface pressure of the rhamnolipid-containing extract (from P. aeruginosa PA14 CS) was not higher than the surface pressure of other extracts (Figure 48), including that from the reference strain (E. coli TG2) CS.

Thus, surface pressures of all extracts were determined not only by the activity of rhamnolipids, but also by the activity of other lipids (which were either secreted by microbes or were parts of microbial cells). Nevertheless, the emulsification experiments (Figure 41) showed that CSs of P. aeruginosa PA14

(the native rhamnolipid producer) and E. coli pF1bR4 (the recombinant strain) were more powerful hydrocarbon emulsifiers than the E. coli TG2 (the reference strain) CS.

5.4.2. Effects of short-time incubations and CS/solution floods

A CS/surfactant solution injection combined with a 24-hour incubation of columns and subsequent SPW flood turned out to be an efficient approach only when certain CSs/surfactants were applied. In particular, in the case of EO

(Table 15), 1% Triton X-100 was the only agent helping to enhance oil recovery when short-time incubations and solution floods were applied. In the case of LO

(Table 15), P. aeruginosa PA14 CS was the most efficient agent, followed by

1% Triton X-100, 1% SDS and E. coli pF1bR4 CS. Thus, when short-time

145 incubations and SPW floods were applied to columns with LO, rhamnolipids

(contained in P. aeruginosa PA14 CS) turned out to be more efficient EOR agents than chemical surfactants. Among chemical surfactants, the non-ionic surfactant (Triton X-100) and the anionic surfactant (SDS) were more efficient than the cationic surfactant (CTAB), in agreement with the findings described by

Liu et al. (2011). Inefficiency of CTAB for oil recovery from a column with silica sand may be explained by its attraction to the negatively charged surfaces of sand granules and rendering them more oil-wet, thus, making oil production more difficult (this mechanism was also discussed in section 1.5.2). On the contrary, anionic and non-ionic surfactants can minimally adsorb to sand granule surfaces, so that they are unlikely to significantly change wettability, while they were capable of reducing oil-water IFT. As expected, chemical surfactants and a CS containing rhamnolipids in a high concentration (P. aeruginosa PA 14 CS), generally, were more effective EOR agents during short-time incubations and water floods than CSs containing biosurfactants in lower concentrations (E. coli pF1bR4 and P. aeruginosa PDO111 CSs).

Enhancement of the recovery of LO (21º API gravity; rich in polyaromatic hydrocarbons) during short-time incubations and floods was greater than that of

EO (16º API gravity; rich in alkanes) when 1% Triton X-100 was applied. In addition, P. aeruginosa PA14 CS, 1% SDS and E. coli pF1bR4 CS helped enhance recovery of LO, but not EO (Table 15). It cannot be ruled out that the recovery of oil was generally more difficult from the four columns with trapped

EO aged for two months (section 5.2.7.6) than from all others, which were not aged when they contained only trapped oil (long-time aging could render sand

146

Table 16. Oil, water and gas (N2 and CO2) production from columns containing (bio)surfactant solutions during long-time incubations in an anaerobic hood at 30˚C. Column set 1 contained LO and the column sets 2 and 3 contained EO. One round of incubations lasted for 39-45 days; media were replenished at the beginning of each round of incubations. No water flood was applied during incubations. The mechanism of liquid production during long-time incubation is explained in the text.

Volume of oil trapped Total Total Average Average Original after volume of volume of conc. of Pore Total conc. of N in oil in secondary oil water 2 CO in a vial Column Type of Injected volume OOIP/ incub. Voi/ Voi/ Voi/ Vwi/ Voi/PV + a vial after 2 place floods and produced produced after one set oil solution (PV), PV time, PV OOIP Vot2 PV Vwi/PV one round of (OOIP), injections during during round of mL days incub., mL between incub. incub. incub., mmol/L the floods (Voi), mL (Vwi), mL mmol/L (Vot2), mL

1 LO TG2-CS 13.22 11.91 0.90 3.89 79 0.00 0.00 0.00 0.00 0.00 0.00 0.00 18.1±14.25 2.1±1.27 1 LO 1% Triton X-100 13.24 12.42 0.94 3.98 85 0.12 0.01 0.01 0.03 0.40 0.03 0.04 17.8±7.28 1±0.00 1 LO 1% CTAB 13.16 11.78 0.90 2.70 85 0.00 0.00 0.00 0.00 0.05 0.00 0.00 40.3±20.08 2.2±0.16 1 LO 1% SDS 13.18 11.88 0.90 3.27 79 0.00 0.00 0.00 0.00 0.00 0.00 0.00 13.2±7.67 1.8±0.02 1 LO PA 14-CS 13.09 11.93 0.91 2.56 85 0.06 0.00 0.00 0.02 1.50 0.11 0.12 27.7±0.48 5.1±4.47 1 LO pF1bR4-CS 13.33 12.32 0.92 2.11 85 0.12 0.01 0.01 0.06 0.13 0.01 0.02 15.5±14.60 1.8±0.05 1 LO PDO111-CS 13.31 12.42 0.93 1.78 85 0.00 0.00 0.00 0.00 1.65 0.12 0.12 36.3±29.67 3.7±0.78 2 EO TG2-CS 13.46 12.08 0.90 3.09 79 0.03 0.00 0.00 0.01 0.00 0.00 0.00 7.4±4.27 2.4±0.02 2 EO 1% Triton X-100 13.31 11.98 0.90 3.01 79 0.02 0.00 0.00 0.01 0.30 0.02 0.02 16.9±11.02 2.3±1.17 2 EO 1% CTAB 13.34 12.05 0.90 3.63 79 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22.7±15.48 1.8±1.28 3 EO 1% SDS 13.47 12.24 0.91 2.94 88 0.04 0.00 0.00 0.01 1.22 0.09 0.09 25±15.72 2±0.97 3 EO PA14-CS 13.91 12.7 0.91 2.43 88 0.03 0.00 0.00 0.01 1.80 0.13 0.13 25±17.75 3±0.66 3 EO pF1bR4-CS 14.04 12.66 0.90 2.36 88 0.23 0.02 0.02 0.10 0.00 0.00 0.02 26±12.5 2.7±1.08 3 EO PDO111-CS 13.83 12.66 0.92 2.31 88 0.52 0.04 0.04 0.22 1.60 0.12 0.15 19±13.63 3.2±3.15

147 strongly oil-wet, which is the least favourable wettability state, as it was discussed in section 1.5.2; Rao et al., 2006). Results of not only short-time incubations combined with floods, but also long-time incubations (see below) indicate that the efficiency of the application of (bio)surfactants (and/or bacterial cells and/or cell parts) for EOR-purposes may depend on the composition of trapped oil. IFT reduction was the likely mechanism of surfactant-assisted oil production under brine floods. Favourable wettability changes could, in principle, contribute as well. However, as it was already mentioned in this section and in section 1.5.2, silica sand wettability alterations by anionic and, to a lesser extent, non-ionic surfactants, might be problematic; and wettability alterations of silica sand caused by cationic surfactants could be unfavourable.

Oil emulsification could contribute to enhancement of oil production during replenishments of surfactant solutions. These replenishments were conducive to oil emulsification as they were conducted at high flow rates (about

15 mL/min) and, hence, likely provided sufficient shear energy (McAuliffe, 1973;

Bryan and Kantzas, 2007).

5.4.3. Effects of long-time incubations

During long-time incubations, flood/injection-independent oil and/or water discharge from the columns could be observed because of the counter-diffusion phenomenon and due to the drop of the solubility of gases in oil and water after columns were transferred from a 20ºC into a 30ºC environment (this is discussed in section 3.4). Growth of the injected bacteria (as parts of CSs) was not expected to occur because no electron acceptor was added (except for

148 small amounts of oxygen dissolved in CSs); in addition, it was not known if the injected strains were capable of living in a crude oil containing environment.

The amounts of oil and water produced during long-time incubations varied depending on the applied surface active agents. The largest enhancement of solution-gas-driven oil production was seen with the set 3 columns containing trapped EO (Table 15). In this case, during long-time incubations, the column with P. aeruginosa PDO111 CS discharged 22.4% of trapped EO or 4.1% of original oil in place (OOIP) and the column with E. coli pF1bR4 CS discharged 9.9% of trapped EO or 1.8% of OOIP; the column with

E. coli pF1bR4 CS also discharged 5.7% of trapped LO or 1% of OOIP (Tables

15 and 16). Therefore, recovery of oil during counter-diffusion-induced gas drive was enhanced only when rhamnolipids were present in very low concentrations, if present at all. This may seem somewhat surprising, but one has to note that all CSs contained surface active components other than rhamnolipids (e.g. cell parts, which were likely responsible for high surface activity of the devoid of rhamnolipids E. coli TG2 CS extract, demonstrated in Figure 48). It is also important to note that surfactants in low concentrations can still reduce an oil- water IFT, and live and/or dead bacterial cells or their parts can also contribute to its reduction; Kowalewski et al., 2006).

Cells from E. coli TG2 CS not carrying rhlAB genes did not help enhance solution-gas-driven oil recovery. This lets one think that only those cells or their parts, which were hydrophobic enough to be capable of attachment to an oil- water interface, could lead to MEOR-effects in this case. Presence of rhlAB in

E. coli pF1bR4 and P. aeruginosa PDO111, while not leading to production of previously described rhamnolipids (Price et al., 2009; Nie et al., 2010), was still

149 likely translated into sufficient hydrophobicity of their cell surfaces. Indeed,

Figures 44 and 45 indicate that CSs from the cultures of these strains did contain a compound with the same m/z ratio as one of the compounds of the commercial rhamnolipid (i.e. the molecule with the m/z equalling 361.96–

361.99; section 5.3.1); at the same time, this compound was not found in E. coli

TG2 CS (Figure 46). Thus, rhamnosyltransferase 1 might be produced in small amounts in E. coli pF1bR4 (this is in agreement with findings reported by

Ochsner et al., 1995) and in P. aeruginosa PDO111, and the product(s) of its activity might enhance attachment of cells or their parts to an oil-water interface (Zhang and Miller, 1994).

P. aeruginosa PA14 CS and chemical surfactants did not help enhance solution-gas-driven oil recovery (Table 15) likely because of the following. P. aeruginosa PA14, having a quorum sensing system (Ochsner and Reiser, 1995;

Miller and Bassler, 2001), produced rhamnolipids in a concentration (Table 14) sufficient for tight covering of gas bubble and oil droplet surfaces, with hydrophilic heads pointed to the aqueous phase and the hydrophobic tails pointed to a gas phase or an oil phase respectively (Figure 1A). CTAB, SDS and Triton X-100, each of which was present in a concentration higher than its

CMC, could also tightly cover gas bubbles and oil droplets. Consequently, spreading of an oil droplet over a gas bubble in the way described by Moosai and Dawe (2003) would become impossible because of the electrostatic repulsion between hydrophilic heads of a surfactant covering both surfaces

(Figure 1A). Oil recovery can, therefore, be impeded in these cases (possibility of this was discussed in section 1.5.2). Additionally, foams could be formed in

P. aeruginosa PA14 CS and in 1% SDS, Triton X-100 and CTAB solutions.

150

Presence of foam can reduce gas mobility to zero (Bernard and Holm,

1964) and prevent gas-driven oil mobilization. When gas has zero mobility, then liquids may be displaced because of their swelling and/or gas expansion

(sections 1.5.1 and 1.5.2), with water likely being preferentially produced due to having higher mobility than that of heavy oil. On the other hand, it is known that under higher flow rates, particularly in a surfactant alternating gas process, foaming can be used to improve mobility ratios (Farajzadeh et al., 2009).

When concentrations of rhamnolipids, cells and/or their parts are lower, like in the cases with E. coli pF1bR4 and P. aeruginosa PDO111 CSs, their presence on oil droplet surfaces would not lead to repulsion between the droplets and gas bubbles. Therefore, the droplets could spread over the bubbles, and the presence of surfactants, cells and/or cell parts would facilitate gas-driven oil production through reduction of an oil–water IFT (Figure 1B).

Foam formation would not likely to occur in these cases.

Interestingly, E. coli pF1bR4 CS enhanced the recovery of EO (16º API gravity; rich in alkanes) to much greater extent than the recovery of LO (21º API gravity; poor in alkanes, but rich in polyaromatic hydrocarbons) and P. aeruginosa PDO111 CS enhanced the recovery of only EO, but not LO (Tables

15 and 16), indicating again (see section 5.4.2) that the efficiency of the application of biosurfactants and/or bacterial cells and/or cell surface parts for

EOR-purposes may depend on the composition of trapped oil.

5.5. Conclusions and implications

Injection of a rhamnolipid solution into a column/oil reservoir with trapped oil with a subsequent short-time incubation and water/SPW flood may enhance

151 oil recovery provided the rhamnolipid concentration is high enough to sufficiently reduce an oil-water IFT. In some cases (e.g. when the oil rich in polyaromatic hydrocarbons and poor in alkanes is being recovered), rhamnolipids may be more efficient EOR agents than chemical surfactants.

In an oil reservoir subjected to solution gas drive, the presence of rhamnolipids, hydrophobic bacterial cells and/or their parts in concentrations sufficient for oil-water IFT reduction, but low enough not to allow repulsion between oil droplets and gas bubbles, may also enhance oil recovery.

Potentially useful in this respect may be the oil-associated microflora like that from the field of origin of EO (Chapter 4). External gas drive, as opposed to solution gas drive, is less likely to lead to a success because of the great probability of flushing only high permeability zones and omitting low- permeability zones (Kumar et al., 2000) where most oil is usually trapped.

According to our results, introduction of rhlAB genes into E. coli is not sufficient for production of large amounts of rhamnolipids by a recombinant strain, although the presence of these genes may still affect cell surface hydrophobicity. One of the ways for improvement of rhamnolipid production by the recombinant strain is the incorporation of the P. aeruginosa rml-BDAC operon allowing increased production of dTDP-rhamnose (Cabrera-Valladares et al., 2006).

Further experiments with multiple non-pathogenic hosts and broad host range vectors may help in the construction of recombinant microorganisms capable of efficient in or ex situ biosurfactant production.

152

Chapter 6: Main conclusions and directions for future research

All hypotheses (section 2.2) were supported by the experimental findings described in the thesis. In particular,

- stimulation of indigenous oil field microflora with nitrate did not lead to enhancement of water-flood-driven recovery of heavy oil trapped in model columns, indicating that numbers and sizes of microorganisms grown with nitrate were not sufficient to plug high permeability zones and that the influx of water likely led to diluting biosurfactants possibly produced by indigenous microflora during the time of incubation with nitrate, so that oil-water IFT was not sufficiently reduced;

- stimulation of indigenous oil field microflora with nitrate led to enhancement of counter-diffusion-induced-gas-driven recovery of heavy oil rich in alkanes (EO) and, to a lesser extent, of heavy oil rich in polyaromatic hydrocarbons (LO) trapped in model columns, suggesting that very low concentrations of microbially produced surfactants were required for enhancement of gas-driven oil recovery;

- primary hydrocarbon degraders were found to be adhering to the oil phase of the mixture of produced water with EO; they were most likely responsible for the MEOR-effect of nitrate injections mentioned above, as their cells and metabolites, due to their localization, were capable of influencing

(reducing) an oil-water IFT; hydrogenotrophs were also found to adhere to an oil phase, likely because of greater availability of dissolved hydrogen in oil than in water; consumers of water-soluble substrates were found mostly in the aqueous phase of the same produced water-EO mixture;

153

- the presence of a cell-containing supernatant from the culture of a native rhamnolipid producer in a water-flooded column with trapped LO enhanced oil recovery to a greater extent than the presence of chemical surfactants;

- a supernatant from the culture of the native rhamnolipid-producer P. aeruginosa PA14 and 1% solutions of CTAB, SDS and Triton X-100 did not help enhance oil recovery under counter-diffusion-induced gas drive likely because

(bio-)surfactants taken in relatively high concentrations either (a) prevented spreading of oil droplets over gas bubbles due to electrostatic repulsion between them or (b) stopped gas flow through foam formation;

- cell-containing supernatants from cultures of P. aeruginosa PDO111 and E. coli pF1bR4, containing small amounts of biosurfactants, helped enhance oil recovery under counter-diffusion-induced gas drive suggesting that biosurfactants taken in low concentrations were capable of sufficient reduction of oil-water IFT, while spreading of oil droplets over gas bubbles was not prevented and no or little foams were formed.

Based on these conclusions, it may be recommended that 20 mM nitrate is injected in the MHGC field (the field of origin of EO) right before the reservoir pressurization if solution gas drive is used as an oil production mechanism. It is important that the reservoir is kept under pressure below the minimum miscibility pressure for a given gas, especially if it is carbon dioxide, which is known to be capable of de-activating microflora under high pressures (Dillow et al., 1999). The reservoir is to be incubated for 30-40 days to allow growth of microorganisms. Afterwards, the reservoir may be de-pressurized to create solution gas drive. In principle, it could also be possible to pressurize and de-

154 pressurize the reservoir after its incubation for 30-40 days with 20 mM nitrate.

The results presented in Chapter 3 suggest that it is likely that solution-gas- driven EO recovery will be enhanced by nitrate injection due to biosurfactant production and growth of oil-associated biomass. Alternatively, to enhance EO recovery, injection of a (bio)surfactant in a concentration below CMC, instead of nitrate injection, may be applied in combination with solution gas drive (Chapter

5). In this case, prolonged reservoir incubation will not be required. Additional research has to be conducted to find out if these approaches could be effective in the LAS field (field of origin of LO). In particular, it will be worthwhile to replicate experiments with trapped LO described in Chapters 3 and 5.

In the future, using steel columns in experiments similar to those described in Chapter 3, but involving gas (e.g. CO2 or CH4) injection and long- time incubation at high pressure (e.g. at several hundred psi) may help elucidate the extent at which high pressure may interfere with nitrate-mediated microbial enhancement of solution-gas-driven oil recovery.

In the LAS field, the application of rhamnolipids in concentrations exceeding CMC in combination with a water flood may lead to more efficient oil recovery than the application of chemical surfactants. Additional experiments are required to be conducted to determine if the application of rhamnolipids in the combination with a water flood may be efficient in the MHGC field. However, generally, this approach may be economically prohibitive due to the high cost of industrial biosurfactant production (Mukherjee et al., 2006) and the need to apply them in large amounts during water-flood-driven oil recovery (Gray et al.,

2008; Chapter 5).

155

To better understand the mechanism of counter-diffusion-induced gas drive, in the future it will be worthwhile to use micromodels for conducting experiments similar to those involving long-time incubations of miniature model columns with trapped oil (sections 3.2.7 and 5.2.7.6). During such experiments, it will be important to observe formation and behaviour of gas bubbles in a porous matrix containing trapped oil droplets.

The analysis of compositions of microbial communities associated with oil and water in columns containing trapped oil and nitrate at the end of long- time incubation will help get more information about microorganisms responsible for nitrate-mediated MEOR. It will also be interesting to compare compositions of these communities with those isolated from the oil-containing produced water (Chapter 4). Some other recommended future experiments include determination of oil-water IFT, amounts of biomass (e.g. through measuring protein contents) and levels of expression of some well-known microbial genes responsible for biosurfactant production (e.g. using qPCR) in columns incubated with and without nitrate.

156

References

Abbasnezhad, H., Gray, M.R., Foght, J.M., 2008. Two different mechanisms for

adhesion of Gram-negative bacterium, Pseudomonas fluorescens LP6a,

to an oil–water interface. Colloids and Surfaces B: Biointerfaces 62, 36–

41.

Agrawal, A., Park, H.S., Nathoo, S., Gieg, L.M., Jack, T.R., Miner, K., Ertmoed,

R., Benko, A., Voordouw, G., 2012. Toluene depletion in produced oil

contributes to souring control in a field subjected to nitrate injection.

Environmental Science and Technology 46, 1285−1292.

Akasaka, H., Ueki, A., Hanada, S., Kamagata, Y., Ueki, K., 2003.

Propionicimonas paludicola gen. nov., sp. nov., a novel facultatively

anaerobic Gram-positive, propionate-producing bacterium isolated from

plant residue in irrigated rice field soil. International Journal of Systematic

and Evolutionary Microbiology 53, 1991–1998.

Al-Awadhi, H., Sulaiman, R.H., Mahmoud, H.M., Radwan, S.S., 2007.

Alkaliphilic and halophilic hydrocarbon-utilizing bacteria from Kuwaiti

coasts of the Arabian Gulf. Applied Microbiology and Biotechnology 77,

183-186.

Almeida, P. F., Moreira, R. S., Almeida, R. C. C., Guimaraes, A. K., Carvalho,

A. S., Quintella, C., Esperidia, M. C. A., Taft, C. A., 2004. Selection and

application of microorganisms to improve oil recovery. Engineering in Life

Sciences 4, 319–325.

157

Alshmakhy, A.B., Maini, B.B., 2009. Effect of foaminess on the performance of

solution gas drive in heavy oil reservoirs. Journal of Canadian Petroleum

Technology 48, 36–41.

Andreoni, V., Bernasconi, S., Colombo, M., van Beilen, J.B., Cavalca, L., 2000.

Detection of genes for alkane and naphthalene catabolism in

Rhodococcus sp. strain 1BN. Environmental Microbiology 2, 572–577.

Atlas, R.M., Bartha, R., 1992. Hydrocarbon biodegradation and oil spill

bioremediation. Advances in Microbial Ecology 12, 287–338.

Bae, H.-S., Moe, W.M., Yan, J., Tiago, I., da Costa, M.S., Rainey, F.A., 2006.

Propionicicella superfundia gen. nov., sp. nov., a chlorosolvent-tolerant

propionate-forming, facultative anaerobic bacterium isolated from

contaminated groundwater. Systematic and Applied Microbiology 29,

404–413.

Bae, J. H., Chambers, K. T., Lee, H. O., 1996. Microbial profile modification with

spores. SPE Reservoir Engineering 11, 163–167.

Banat, I.M., 1995. Biosurfactants production and possible uses in microbial

enhanced oil recovery and oil pollution remediation: A review.

Bioresource Technology 51, 1–12.

Banat, I. M., Makkar, R. S., Cameotra, S. S., 2000. Potential commercial

applications of microbial surfactants. Applied Microbiology and

Biotechnology 53, 495–508.

Beckman, J.W., 1926. The action of bacteria on mineral oil. Industrial

Engineering Chemical News 4, 3.

Beckmann, S., Lueders, T., Krüger, M., von Netzer, F., Engelen, B., Cypionka,

H., 2011. Acetogens and acetoclastic Methanosarcinales govern

158

methane formation in abandoned coal mines. Applied and Environmental

Microbiology 77, 3749–3756.

Beeder, J., Nilsen, R.K., Thorstenson, T., Torsvik T., 1996. Penetration of

sulfate reducers through a porous North Sea oil reservoir. Applied and

Environmental Microbiology 62, 3551–3553.

Behlulgil, K., Mehmetoglu, M. T., 2002. Bacteria for improvement of oil

recovery: A laboratory study. Energy Sources 24, 413–421.

Belyaev, S. S., Ivanov, V. M., 1990. Contemporary geochemical activity of

microorganisms in the oilfields, and the possibility of its regulation.

Geochimiya 11, 1618–1625.

Bernard, G.G., Holm, L.W., 1964. Effect of foam on permeability of porous

media to gas. SPE 983. SPE Annual Fall Meeting, Houston, TX.

Bi, Z., Liao, W., Qi, L., 2004. Wettability alteration by CTAB adsorption at

surfaces of SiO2 film or silica gel powder and mimic oil recovery. Applied

Surface Science 221, 25–31.

Bødtker, G., Hvidsten, I. V., Barth, T., Torsvik T., 2009. Hydrocarbon

degradation by Dietzia sp. A14101 isolated from an oil reservoir model

column. Antonie van Leeuwenhoek 96, 459–469.

Bouchez-Naitali, M., Vandecasteele J. P., 2008. Biosurfactants, an help in the

biodegradation of hexadecane? The case of Rhodococcus and

Pseudomonas strains. World Journal of Microbiology and Biotechnology

24, 1901–1907.

Bregnard, T.P.A., Häner, A., Höhener, P., Zeyer, J., 1997. Anaerobic

degradation of pristane in nitrate-reducing microcosms and enrichment

cultures. Applied and Environmental Microbiology 63, 2077–2081.

159

Brint, J. M., Ohman, D. E., 1995. Synthesis of multiple exoproducts in

Pseudomonas aeruginosa is under the control of RhlR-RhlI, another set

of regulators in strain PAO1with homology to the autoinducer-responsive

LuxR-LuxI family. Journal of Bacteriology 177, 7155–7163.

Brown, F. G., 1992. Microbes: The practical and environmental safe solution to

production problems, Enhanced Production, and Enhanced Oil

Recovery. SPE 23955. SPE Permian Basin Oil and Gas Recovery

Conference, Midland, TX.

Brown, L. R., Vadie, A. A., Stephens, J. O., 2002. Slowing production decline

and extending the economic life of an oil field: New MEOR technology.

SPE Reservoir Evaluation and Engineering 5, 33–41.

Bryan, J., Kantzas, A., 2007. Enhanced heavy-oil recovery by alkali-surfactant

flooding. SPE 110738. SPE Annual Technical Conference and

Exhibition, Anaheim, CA.

Bryant, R. S., 1988. Microbial enhanced oil recovery and compositions thereof.

Patent 4905761. U.S. Patent Office.

Bryant, R. S., 1991. MEOR screening criteria fit 27% of U.S. oil reservoirs. Oil

and Gas Journal 15, 56–59.

Bryant, R.S., Burchfield, T.E., 1989. Review of microbial technology for

improving oil recovery. SPE Reservoir Engineering 4, 151–154.

Bryant, R.S., Burchfield, T.E., Chase, K.L., Bertus, K.M., Stepp, A.K., 1989.

Optimization of microbial formulations for oil recovery: Mechanisms of oil

mobilization, transport of microbes and metabolites, and effects of

additives. SPE 19686. 64th Annual Technical Conference and Exhibition,

San Antonio TX.

160

Bryant, R. S., Douglas, J., 1988. Evaluation of microbial systems in porous

media for EOR. SPE Reservoir Engineering 3, 489–495.

Bryant, R. S., Stepp, A. K., Bertus, K. M., Burchfield, T. E., Dennis, M., 1993.

Microbial enhanced waterflooding field pilots. Developments in

Petroleum Science 39, 289–306.

Burton, S., 2004. Isobaric counter diffusion. ScubaEngineer.com, online at

http://www.scubaengineer.com/isobaric_counter_diffusion.htm, accessed

on November 10, 2012.

Cabrera-Valladares, N., Richardson, A.-P., Olvera, C., Treviño, L. G., Déziel, E.,

Lépine, F., Soberón-Chávez, G., 2006. Monorhamnolipids and 3-(3-

hydroxyalkanoyloxy) alkanoic acids (HAAs) production using Escherichia

coli as a heterologous host. Applied Microbiology and Biotechnology 73,

187–194.

Cai, H.-Y., Shaw, J.M., Chung, K.H., 2001. Hydrogen solubility measurements

in heavy oil and bitumen cuts. Fuel 80, 1055–1063.

Cameotra, S. S., Makkar, R. S., 1998. Synthesis of biosurfactants in extreme

conditions. Applied Microbiology and Biotechnology 50, 520–529.

Camper, A. K., Hayes, J. T., Strurman, P. J., Jones, W. L., Cunningham, A. B.,

1993. Effects of motility and adsorption rate coefficient on transport of

bacteria through saturated porous media. Applied and Environmental

Microbiology 59, 3455–3462.

Cha, M., Lee, N., Kim, M., Kim M., Lee, S., 2008. Heterologous production of

Pseudomonas aeruginosa EMS1 biosurfactant in Pseudomonas putida.

Bioresource Technology 99, 2192–2199.

161

Chaîneau, C.H., Morel, J., Dupont, J., Bury, E., Oudot, J., 1999. Comparison of

the fuel oil biodegradation potential of hydrocarbon-assimilating

microorganisms isolated from a temperate agricultural soil. Science of

the Total Environment 227, 237–247.

Chakraborty, R., Coates, J.D., 2004. Anaerobic degradation of monoaromatic

hydrocarbons. Applied Microbiology and Biotechnology 64, 437–446.

Chang, Y., 1987. Preliminary studies assessing sodium pyrophosphate effects

on microbially mediated oil recovery. Annals of the New York Academy

of Sciences 506, 296–307.

Chen, S., Hickey, W.J. 2011. Development of tools for genetic analysis of

phenanthrene degradation and nanopod production by Delftia sp.Cs1-4.

Frontiers in Microbiology 2, 1–10.

Coates, J.D., Bhupathiraju, V.K., Achenbach, L.A., Mclnerney, M.J., Lovley,

D.R., 2001. Geobacter hydrogenophilus, Geobacter chapellei and

Geobacter grbiciae, three new, strictly anaerobic, dissimilatory Fe(III)-

reducers. International Journal of Systematic and Evolutionary

Microbiology 51, 581–588.

Cornish-Shartau, S.L., Yurkiw, M., Lin, S., Grigoryan, A. A., Lambo, A., Park,

H.S., Lomans, B.P., van der Biezen, E., Jetten, M.S.M, Voordouw, G.,

2010. Ammonium concentrations in produced waters from a

mesothermic oil field subjected to nitrate injection decrease through

formation of denitrifying biomass and anammox activity. Applied and

Environmental Microbiology 76, 4977–4987.

162

Cunningham, A. B., Sharp, R. R., Caccavo, Jr., F., Gerlach, R., 2007. Effects of

starvation on bacterial transport through porous media. Advances in

Water Resources 30, 1583–1592.

Daniels, R., Reynaert, S., Hoekstra, H., Verreth, C., Janssens, J., Braeken, K.,

Fauvart, M., Beullens, S., Heusdens, C., Lambrichts, I., De Vos, D. E.,

Vanderleyden, J., Vermant, J., Michiels, J., 2006. Quorum signal

molecules as biosurfactants affecting swarming in Rhizobium etli. PNAS

103, 14965–14970.

Davey, M.E., Caiazza, N.C., O'Toole, G.A., 2003. Rhamnolipid surfactant

production affects biofilm architecture in Pseudomonas aeruginosa PAO

1. Journal of Bacteriology 185, 1027–1036.

Davidova, I., Hicks, M.S., Fedorak, P.M., Suflita, J.M., 2001. The influence of

nitrate on microbial processes in oil industry production waters. Journal

of Industrial Microbiology and Biotechnology 27, 80–86.

Davis, D. A., Lynch, H. C., Varley, J., 1999. The production of surfactin in batch

culture by Bacillus subtilis ATCC 21332 is strongly influenced by the

conditions of nitrogen metabolism. Enzyme and Microbial Technology 25,

322–329.

Dahle, H., Birkeland, N.-K., 2006. Thermovirga lienii gen. nov., sp. nov., a novel

moderately thermophilic, anaerobic, amino-acid-degrading bacterium

isolated from a North Sea oil well. International Journal of Systematic

and Evolutionary Microbiology 56, 1539–1545.

De Lorenzo, V., 2010. Environmental biosafety in the age of synthetic biology:

Do we really need a radical new approach? Bioessays 32, 926–931.

163

Dennis, D.M., Hitzman, D.O., 2007. Advanced nitrate-based technology for

sulphide control and improved oil recovery. SPE 106154. SPE

International Symposium on Oilfield Chemistry, Houston, TX.

Desai, J.D., Banat, I.M., 1997. Microbial production of surfactants and their

commercial potential. Microbiology and Molecular Biology Reviews 6,

47–64.

Desouky, S. M., Abdel-Daim, M. M., Sayyouh, M. H., Dahab, A. S., 1996.

Modeling and laboratory investigation of microbial enhanced oil recovery.

Journal of Petroleum Science and Engineering15, 309–320.

Dillow, A.K., Dehghani, F., Hrkach, J.S., Foster, N.R., Langer, R., 1999.

Bacterial inactivation by using near- and supercritical carbon dioxide.

PNAS 96, 10344–10348.

Dong, C., Lai, Q., Chen, L., Sun, F., Shao, Z., Yu, Z., 2010. Oceanibaculum

pacificum sp. nov., isolated from hydrothermal field sediment of the

south-west Pacific Ocean. International Journal of Systematic and

Evolutionary Microbiology 60, 219–222.

Dostalek, M., Spurny, M., 1958. Bacterial release of oil: Preliminary trials in an

oil deposit. Folia Biologica 4, 166–172.

Farajzadeh, R., Andrianov, A., Zitha, P.L.J., 2009. Foam assisted enhanced oil

recovery at miscible and immiscible conditions. SPE 126410. Kuwait

International Petroleum Conference and Exhibition, Kuwait City.

Firoozabadi, A., 2001. Mechanisms of solution gas drive in heavy oil reservoirs.

Journal of Canadian Petroleum Technology 40, 15–20.

164

Firoozabadi, A., Aronson, A., 1999. Visualization and measurement of gas

evolution and flow of heavy and light oil in porous media. SPE Reservoir

Evaluation and Engineering 2, 550–557.

Fitch, W.M., 1971. Toward defining the course of evolution: Minimum change

for a specific tree topology. Systematic Zoology 20, 406–416.

Fontes, D. E., Mills, A. L., Hornberger, G. M., Herman, J. S., 1991. Physical and

chemical factors influencing transport of microorganisms through porous

media. Applied and Environmental Microbiology 57, 2473–2481.

Frias, J. A., Richman, J. E., Wackett, L. P., 2009. C29 olefinic hydrocarbons

biosynthesized by Arthrobacter species. Applied and Environmental

Microbiology 75, 1774–1777.

Fujiwara, K., Sugai, Y., Yazawa, N., Ohno, K., Hong, C.X., Enomoto, H., 2004.

Biotechnological approach for development of microbial enhanced oil

recovery technique. Studies in Surface Science and Catalysis 151, 405–

445.

Gautam, K. K., Tyagi, V. K., 2006. Microbial surfactants: A review. Journal of

Oleo Science 55, 155–166.

Gentleman, R., 2008. Bioinformatics with R. Chapman and Hall/CRC, Boca

Raton, FL.

Gieg, L.M., Duncan, K.E., Suflita, J.M., 2008. Bioenergy production via

microbial conversion of residual oil to natural gas. Applied and

Environmental Microbiology 74, 3022–3029.

Gieg, L.M., Jack, T.R., Foght, J.M., 2011. Biological souring and mitigation in

oil reservoirs. Applied Microbiology and Biotechnology 92, 263–282.

165

Glaser, L., Kornfeld, S., 1966. Formation of dTDP-L-rhamnose from dTDP-D-

glucose. Methods in Enzymology 8, 303–306.

Grabowski, A., Nercessian, O., Fayolle, F., Blanchet, D., Jeanthon, C., 2005a.

Microbial diversity in production waters of a low-temperature

biodegraded oil reservoir. FEMS Microbiology Ecology 54, 427–443.

Grabowski, A., Tindall, B.J., Bardin, V., Blanchet, D., Jeanthon, C., 2005b.

Petrimonas sulfuriphila gen. nov., sp. nov., a mesophilic fermentative

bacterium isolated from a biodegraded oil reservoir. International Journal

of Systematic and Evolutionary Microbiology 55, 1113–1121.

Graves, D.J., Idicula, J., Lambertsen, C. J., Quinn, J. A., 1972. Bubble

formation in physical and biological systems: a manifestation of

counterdiffusion in composite media. Science 179, 582–584.

Graves, D.J., Idicula, J., Lambertsen, C. J., Quinn, J. A., 1973. Bubble

formation resulting from counterdiffusion supersaturation: a possible

explanation for isobaric inert gas “urticaria” and vertigo. Physics in

Medicine and Biology 18, 256–264.

Gray, M., Yeung, A., Fought, J., Yarranton H., 2008. Potential microbial

enhanced oil recovery processes: A critical analysis. SPE114676. SPE

Annual Technical Conference and Exhibition, Denver, CO.

Green, M.S., Flanegan, K.C., Gilcrease, P.C., 2008. Characterization of a

methanogenic consortium enriched from a coalbed methane well in the

Powder River Basin, USA. International Journal of Coal Geology 76, 34–

45.

Green, D. W., Willhite, G. P., 1998. Enhanced oil recovery. Society of

Petroleum Engineers, Richardson, TX.

166

Grundmann, O., Behrends, A., Rabus, R., Amann, J., Halder, T., Heider, J.,

Widdel, F., 2008. Genes encoding the candidate enzyme for anaerobic

activation of n-alkanes in the denitrifying bacterium, strain HxN1.

Environmental Microbiology 10, 376–85.

Guerra-Santos, L., Kappeli, O., Fiechter, A., 1984. Process development for the

production of biosurfactants. In: Proceedings of the Third European

Congress in Biotechnology, Vol. 1, Munich, Verlag Chemie, pp. 507–512.

Guerra-Santos, L., Kappeli, O., Fiechter, A., 1986. Dependence of

Pseudomonas aeruginosa continuous culture biosurfactant production on

nutritional and environmental factors. Applied Microbiology and

Biotechnology 24, 443–448.

Gullapalli, I. L., Bae, J. H., Heji, K., Edwards, A., 2000. Laboratory design and

field implementation of microbial profile modification process. SPE

Reservoir Evaluation and Engineering 3, 42–49.

Hall, C., Tharakan, J., Hallock, J., Cleveland, C., Jefferson, M., 2003.

Hydrocarbons and the evolution of human culture. Nature 426, 318–322.

Hamady, M., Lozupone, C., Knight, R., 2010. Fast UniFrac: Facilitating high-

throughput phylogenetic analyses of microbial communities including

analysis of pyrosequencing and PhyloChip data. ISME Journal 4, 17–27.

Harshey, R.M., 2003. Bacterial motility on a surface: Many ways to a common

goal. Annual Review of Microbiology 57, 249–273.

Harvey, R. W., Harms, H., 2001. Transport of microorganisms in the terrestial

subsurface: In situ and laboratory methods. In: Hurst, C.J., Knudsen,

G.R., McInerney, M.J., Stetzenback, L.D., Crawford, R.L. (Eds.), Manual

of Environmental Microbiology, ASM, Washington, DC, pp. 753–776.

167

Head, I.M., Jones, D.M., Larter, S.R., 2003. Biological activity in the deep

subsurface and the origin of heavy oil. Nature 426, 344–352.

Hirasaki, G.J., Miller, C.A., Puerto, M., 2011. Recent advances in surfactant

EOR. SPE 115386. SPE Annual Technical Conference and Exhibition,

Denver, CO.

Hitzman, D. O., 1988. Review of microbial enhanced oil recovery field trials. In:

Burchfield, F. E., Bryant, R. S. (Eds.), Proceedings of Symposium on

Application of Microorganisms to Petroleum Technology, National

Technical Information Service, Springfield, VA, pp. VI1–VI41.

Hitzman, D.O., Dennis, D.M., Hitzman, D.C., 2004. Recent successes: MEOR

using synergistic H2S prevention and increased oil recovery systems.

SPE 89453. SPE/DOE Symposium on Improved Oil Recovery, Tulsa,

OK.

Hitzman, D.O., Sperl, G.T., 1994. A new microbial technology for enhanced oil

recovery and sulphide prevention and reduction. SPE/DOE 27752.

SPE/DOE Ninth Symposium on Improved Oil Recovery, Tulsa, OK.

Holm, L.W., 1987. Evolution of the carbon dioxide flooding processes. Journal

of Petroleum Technology 39,1337–1342.

Holm, L.W., Josendal, V.A., 1974. Mechanisms of oil displacement by carbon

dioxide. Journal of Petroleum Technology 26,1427–1438.

Holm, L.W., O’Brien, L.J., 1971. Carbon dioxide test at the Mead-Strawn field.

Journal of Petroleum Technology 23, 431–442.

Homsy, G. M., 1987. Viscous fingering in porous media. Annual Review of Fluid

Mechanics 19, 271–311.

168

Huse, S., Huber, J., Morrison, H., Sogin, M., Welch, D., 2007. Accuracy and

quality of massively parallel DNA pyrosequencing. Genome Biology 8,

R143.

Huson, D.H., Richter, D.C., Rausch, C., Dezulian, T., Franz, M., Rupp, R.,

2007. Dendroscope: An interactive viewer for large phylogenetic trees.

BMC Bioinformatics 8, 460.

Inoue, H., Nojima, H., Okayama, H., 1990. High efficiency transformation of

Escherichia coli with plasmids. Gene 96, 23–28.

Ivanov, M. V., Belyaev, S. S., Borzenkov, I. A., Glumov, I. F., Ibatulin, P. B.,

1993. Additional oil production during field trials in Russia. In: Premuzic,

E., Woodhead, A. (Eds.), Microbial Enhancement of Oil Recovery –

Recent Advances, Amsterdam, Elsevier, pp. 373–381.

Ivanov, M. V., Belyaev, S. S., Zyakun, M. A., Bondar, A. V., Laurinavichus, S.

K., 1983. Microbiological formation of methane in the oil field

development. Geokhimiya 11, 1647–1654.

Jack, T. R., 1993. MORE to MEOR: An overview of microbial enhanced oil

recovery. In: Premuzic, E., Woodhead, A. (Eds.), Microbial Enhancement

of Oil Recovery – Recent Advances, Elsevier, Amsterdam, pp. 7–16.

Jackson, S.C., Alsop, A., Choban, E., D’Achille, B., Fallon, R., Fisher, J.,

Hendrickson, E., Hnatow, L., Keeler, S., Luckring, A., Nopper, R.,

Norvell, J., Perry, M., Rees, B., Suchanec, D., Wolstenholme, S.M.,

Thrasher, D., Pospisil, G., 2010. Microbial EOR – critical aspects learned

from the lab. SPE 129657. SPE Improved Oil Recovery Symposium,

Tulsa, OK.

169

Jang, L.-K., Chang, P. W., Findley, J. E., Yen, T. F., 1983. Selection of bacteria

with favorable transport properties through porous rock for the

application of microbial enhanced oil recovery. Applied and

Environmental Microbiology 46, 1066–1072.

Jenneman, G. E., Clark, B., 1994a. Injection of phosphorus nutrient sources

under acid conditions for subterranean microbial processes. Patent

5341875. U.S. Patent office.

Jenneman, G. E., Clark, B., 1994b. Utilization of phosphite salts as nutrients for

subterranean microbial processes. Patent 5327967. U.S. Patent office.

Jenneman, G. E., Knapp, R.M., McInerney, M.J., Menzie, D.E., Revus, D.E.,

1982. Experimental studies of in situ microbial enhanced oil recovery.

SPE/DOE 10789. SPE/DOE Third Joint Symposium on Enhanced Oil

Recovery of the Society of Petroleum Engineers, Tulsa, OK.

Jenneman, G. E., McInerney, M. J., Knapp, R. M., 1985. Microbial penetration

through nutrient-saturated Berea sandstone. Applied and Environmental

Microbiology 50, 383–391.

Jimenez, B. J., Rivas, P. R., Lopez, J. G., Pesciaroli, C., Barghini, P., Fenice

M., 2012. Immobilization of Delftia tsuruhatensis in macro-porous

cellulose and biodegradation of phenolic compounds in repeated batch

process. Journal of Biotechnology 157, 148–153.

Jinfeng, L., Lijun, M., Bozhong, M., Rulin, L., Fangtian, N., Jiaxi, Z., 2005. The

field pilot of microbial enhanced oil recovery in a high temperature

petroleum reservoir. Journal of Petroleum Science and Engineering 48,

265–271.

170

Kaiya, S., Utsunomiya, S., Suzuki, S., Yoshida, N., Futamata, H., Yamada, T.,

Hiraishi, A., 2012. Isolation and functional gene analyses of aromatic-

hydrocarbon-degrading bacteria from a polychlorinated-dioxin-

dechlorinating process. Microbes and Environments 27, 127–135.

Kang, Z., Yeung, A., Foght J.M., Gray, M.R., 2008a. Hydrophobic bacteria at

the hexadecane–water interface: Examination of micrometre-scale

interfacial properties. Colloids and Surfaces B: Biointerfaces 67, 59–66.

Kang, Z., Yeung, A., Foght, J.M., Gray, M.R., 2008b. Mechanical properties of

hexadecane-water interfaces with adsorbed hydrophobic bacteria.

Colloids and Surfaces B: Biointerfaces 62, 273–279.

Karaskiewicz, I., 1974. The application of microbiological method for secondary

oil recovery from the Carpathian crude oil reservoir. SLASK, Katowice,

pp. 1–67.

Kendall, M., 1938. A new measure of rank correlation. Biometrica 30, 81–89.

Khomenkov, V.G., Shevelev, A.B., Zhukov, V.G., Kurlovich, A.E., Zagustina,

N.A., Popov, V.O., 2005. Application of molecular systematics to study of

bacterial cultures consuming volatile organic compounds. Applied

Biochemistry and Microbiology 41, 176–84.

Kosheleva, I. A., Balashova, N.V., Izmalkova, T. Y., Filonov, A. E., Sokolov, S.

L., Slepen’kin, A. V., and Boronin, A. M., 2000. Degradation of

phenanthrene by mutant naphthalene-degrading Pseudomonas putida

strains. Microbiology 69, 663–669.

Kowalewski, E., Rueslåtten, I., Gilje, E., Sunde, E., Bødtker, G., Lillebø, B.L.P.,

Torsvik, T., Stensen, J.Å., Bjørkvik, B., Strand, K.A., 2005. Interpretation

171

of microbial oil recovery from laboratory experiments. 13th European

Symposium on Improved Oil Recovery, Budapest.

Kowalewski, E., Rueslåtten, I., Steen, K.H., Bødtker G., Torsæter O., 2006.

Microbial improved oil recovery – bacterial induced wettability and

interfacial tension effects on oil production. Journal of Petroleum Science

and Engineering 52, 275–286.

Kryachko, Y., Dong, X., Sensen, C.W., Voordouw, G., 2012a. Compositions of

microbial communities associated with oil and water in a mesothermic oil

field. Antonie van Leeuwenhoek 101, 493–506.

Kryachko, Y., Nathoo, S., Lai, P., Voordouw, J., Prenner, E.J., Voordouw, G.,

2012b. Prospects for using native and recombinant rhamnolipid

producers for microbially enhanced oil recovery. In press (International

Biodeterioration and Biodegradation).

Kulkarni, M.M., 2003. Immiscible and miscible gas-oil displacements in porous

media, MS thesis, Louisiana State University and A&M College, Baton

Rouge, LA, USA.

Kumar, R., Pooladi-Darvish, M., Okazawa, T., 2000. An investigation into

enhanced recovery under solution gas drive in heavy oil reservoirs. SPE

59336. SPE/DOE Improved Oil Recovery Symposium, Tulsa, OK.

Kuznetsov, S. I., Ivanov, M. V., Lyalikova, N. N., 1963. Introduction to

Geological Microbiology. McGraw-Hill Inc., New York.

Lai, M.C., Chen, S.C., Shu, C.M., Chiou, M.S., Wang, C.C., Chuang, M.J.,

Hong, T.Y., Liu, C.C., Lai, L.J., Hua, J.J., 2002. Methanocalculus

taiwanensis sp. nov., isolated from an estuarine environment.

172

International Journal of Systematic and Evolutionary Microbiology 52,

1799–1806.

Lai, M.C., Lin, C.C., Yu, P.H., Huang, Y.F., Chen, S.C., 2004. Methanocalculus

chunghsingensis sp. nov., isolated from an estuary and a marine

fishpond in Taiwan. International Journal of Systematic and Evolutionary

Microbiology 54, 183–189.

Lai, Q., Yuan, J., Wu, C., Shao, Z., 2009. Oceanibaculum indicum gen. nov., sp.

nov., isolated from deep seawater of the Indian Ocean. International

Journal of Systematic and Evolutionary Microbiology 59, 1733–1737.

Lambo, A.J., Noke, K., Larter, S.R., Voordouw, G., 2008. Competitive,

microbially-mediated reduction of nitrate with sulfide and aromatic oil

components in a low-temperature, Western Canadian oil reservoir.

Environmental Science and Technology 42, 8941–8946.

Lang, S., Wullbrandt, D., 1999. Rhamnose lipids – biosynthesis, microbial

production, and application potential. Applied Microbiology and

Biotechnology 51, 22–32.

Lappin-Scott, H. M., Cusack, F., Costerton, J. W., 1988. Nutrient resuscitation

and growth of starved cells in sandstone cores: A novel approach to

enhanced oil recovery. Applied and Environmental Microbiology 54,

1373–1382.

Lazar, I., 1978. Microbiological methods in secondary oil recovery. In:

Proceedings of the European Symposium on Enhanced Oil Recovery,

Institute of Offshore Engineering and Herriot-Watt University, Edinburgh,

pp. 279–287.

173

Lazar, I., 1991. MEOR field trials carried out over the world during the past 35

years. In: Donaldson, E. C. (Ed.), Microbial Enhancement of Oil

Recovery – Recent Advances. Elsevier Science, Amsterdam, pp. 485–

530.

Lazar, I., 1996. Microbial systems for enhancement of oil recovery used in

Romanian oil fields. In: Opa, N. V. (Ed.), Mineral Processing and

Extractive Metallurgy Review, Vol. 19, pp. 379–393.

Lazar, I., Petrisor, I. G., Yen, T. F., 2007. Microbial Enhanced Oil Recovery

(MEOR). Petroleum Science and Technology 25,1353–1366.

Lever, M.J., Paton, W.D.M., Smith, E.B., 1971. Decompression characteristics

of inert gases. In Lambertsen, C.J. (Ed.), Underwater Physiology,

Academic Press, New York, p. 125.

Li, Q., Kang, C., Wang, H., Liu, C., Zhang, C., 2002. Application of microbial

enhanced oil recovery technique to Daqing oil field. Biochemical

Engineering Journal 11, 197–199.

Liang, Q., Lloyd-Jones, G., 2010. Sphingobium scionense sp. nov., an aromatic

hydrocarbon-degrading bacterium isolated from contaminated sawmill

soil. International Journal of Systematic and Evolutionary Microbiology

60, 413–416.

Lin, S.C., 1996. Biosurfactant: Recent advances. Journal of Chemical

Technology and Biotechnology 63, 109–120.

Liu, W., Jin Y., Tan, X., Yeung, A., 2011. Altering the wettability of bitumen-

treated glass surfaces with ionic surfactants. Fuel 90, 2858–2862.

Lomans, B.P., Maas, R., Luderer, R., Op den Camp, H.J.M., Pol, A., van der

Drift, C., Vogels, G.D., 1999. Isolation and characterization of

174

Methanomethylovorans hollandica gen. nov., sp. nov., isolated from

freshwater sediment, a methylotrophic methanogen able to grow on

dimethyl sulfide and methanethiol. Applied and Environmental

Microbiology 65, 3641–3650.

López, D., Fischbach, M. A., Chu, F., Losick, R., Kolter, R., 2009. Structurally

diverse natural products that cause potassium leakage trigger

multicellularity in Bacillus subtilis. PNAS 106, 280–285.

Lozupone, C., Knight, R., 2005. UniFrac: A new phylogenetic method for

comparing microbial communities. Applied and Environmental

Microbiology 71, 8228–8235.

MacLeod, F. A., Lappin-Scott, H. M., Costerton J. W., 1988. Plugging of a

model rock system by using starved bacteria. Applied and Environmental

Microbiology 54, 1365–1372.

Maini, B.B., Sarma, H.K., George, A.E., 1993. Significance of foamy-oil

behaviour in primary production of heavy oils. Journal of Canadian

Petroleum Technology 32, 50–54.

Marmur, J., 1961. A procedure for the isolation of deoxyribonucleic acid from

microorganisms. Journal of Molecular Biology 3, 208–218.

Maudgalya, S., Knapp, R. M., McInerney, M. J., Nagle, D. P., Folsmbee, M. J.,

2004. Development of a bio-surfactant-based enhanced oil recovery

procedure. SPE 89473. SPE/DOE Improved Oil Recovery Symposium,

Richardson, TX.

McAuliffe, C.D., 1973. Oil-in-water emulsions and their flow properties in porous

media. Journal of Petroleum Technology 25, 727–733.

175

McGill, W.B., Rowell, M.J., 1980. Determination of oil content of oil

contaminated soil. Science of the Total Environment 14, 245–253.

McInerney, M.J., Nagle, D.P., Knapp, R.M., 2005a. Microbially enhanced oil

recovery: past, present, and future. In: Ollivier, B., Magot, M. (Eds.),

Petroleum microbiology, ASM Press, Washington, DC, pp. 215–237.

McInerney, M. J., Duncan, K. E., Youssef, N., Fincher, T., Maudgalya, S. K.,

Folmsbee, M. J., Knapp, R., Simpson, D. R., Ravi, N., Nagle D. P.,

2005b. Development of Microorganisms with Improved Transport and

Biosurfactant Activity for Enhanced Oil Recovery: Report to the

Department of Energy. DE-FC-02NT15321. Washington, DC.

Mikucki, J.A., Liu, Y., Delwiche, M., Colwell, F.S., Boone, D.R., 2003. Isolation

of a methanogen from deep marine sediments that contain methane

hydrates, and description of Methanoculleus submarinus sp. nov. Applied

and Environmental Microbiology 69, 3311–3316.

Miller, M.B., Bassler, B.L., 2001. Quorum sensing in bacteria. Annual Review of

Microbiology 55, 165–199.

Mireles II, J. R., Toguchi, A., Harshey, R. M., 2001. Salmonella enterica Serovar

Typhimurium Swarming Mutants with Altered Biofilm-Forming Abilities:

Surfactin Inhibits Biofilm Formation. Journal of Bacteriology 183, 5848–

5854.

Mohan, P.K., Nakhla, G., Yanful, E.K., 2006. Biokinetics of biodegradation of

surfactants under aerobic, anoxic and anaerobic conditions. Water

Resources 40, 533–540.

176

Moosai, R., Dawe, R.A., 2003. Gas attachment of oil droplets for gas flotation

for oily wastewater cleanup. Separation and Purification Technology 33,

303–314.

Mu, B., Wu, Z., Chen, Z., Wang, X., Ni, F., Zhou, J., 2002. Wetting behavior on

quartz surfaces by the microbial metabolism and metabolic products.

Seventh International Symposium on Wettability and Its Effect on Oil

Recovery, Tasmania, Australia.

Mukherjee, S., Das, P., Sen, R., 2006. Towards commercial production of

microbial surfactants. Trends in Biotechnology 24, 509–515.

Mulligan, C.N., 2005. Environmental applications for biosurfactants.

Environmental Pollution 133, 183–198.

Nahar, N., Alauddin, M., Quilty, B., 2000. Toxic effects of toluene on the growth

of activated sludge bacteria. World Journal of Microbiology and

Biotechnology 16, 307–311.

Nie, M., Yin, X., Ren C., Wang, Y., Xu, F., Shen, Q., 2010. Novel rhamnolipid

biosurfactants produced by a polycyclic aromatic hydrocarbon-degrading

bacterium Pseudomonas aeruginosa strain NY3. Biotechnology

Advances 28, 635–643.

Nuryadi, A., Kishita, A., Watanabe, N., Vilcaez, J., Kawai, N., 2011. EOR

simulation by in situ nitrogen production via denitrifying bacteria and

performance improvement by nitrogen alternating surfactant injection.

SPE 147823. SPE Asia Pacific Oil and Gas Conference and Exhibition,

Jakarta.

177

Obuekwe, C.O., Al-Jadi, Z.K., Al-Saleh, E.S., 2008. Comparative hydrocarbon

utilization by hydrophobic and hydrophilic variants of Pseudomonas

aeruginosa. Journal of Applied Microbiology 105,1876–1887.

Ochsner, U.A., Fiechter, A., Reiser, J., 1994. Isolation, characterization and

expression in Escherichia coli of the Pseudomonas aeruginosa rhlAB

genes encoding a rhamnosyltransferase involved in rhamnolipid

biosurfactant synthesis. Journal of Biological Chemistry 269, 19787–

19795.

Ochsner U.A., Reiser, J., 1995. Autoinducer-mediated regulation of rhamnolipid

biosurfactant synthesis in Pseudomonas aeruginosa. PNAS 92, 6424–

6428.

Ochsner, U.A., Reiser, J., Fiechter, A., Witholt, B., 1995. Production of

Pseudomonas aeruginosa rhamnolipid biosurfactants in heterologous

hosts. Applied and Environmental Microbiology 61, 3503–3506.

Ostos, A., Maini, B., 2004. Capillary number in heavy oil solution gas drive and

its relationship with gas-oil relative permeability curves. SPE 89430.

SPE/DOE Fourteenth Symposium on Improved Oil Recovery, Tulsa, OK.

Özdemir, G., Malayoglu, U., 2004. Wetting characteristics of aqueous

rhamnolipid solutions. Colloids and Surfaces B: Biointerfaces 39, 1–7.

Paria, S., Khilar, K.C., 2004. A review on experimental studies of surfactant

adsorption at the hydrophilic solid – water interface. Advances in Colloid

and Interface Science 110, 75–95.

Park, H.S., Lin, S., Voordouw, G., 2008. Ferric iron reduction by Desulfovibrio

vulgaris Hildenborough wild type and energy metabolism mutants.

Antonie van Leeuwenhoek 93, 79–85.

178

Park, M.-O., Heguri, K., Hirata, K., Miyamoto K., 2005. Production of

alternatives to fuel oil from organic waste by the alkane-producing

bacterium, Vibrio furnissii M1. Journal of Applied Microbiology 98, 324–

331.

Perfumo, A., Banat, I. M., Canganella, F., Marchant, R., 2006. Rhamnolipid

production by a novel thermophilic hydrocarbon-degrading

Pseudomonas aeruginosa AP02-1. Applied Microbiology and

Biotechnology 72,132–138.

Planckaert, M., 2005. Oil reservoirs and oil production. In: Ollivier, B., Magot, M.

(Eds.), Petroleum microbiology, ASM Press, Washington, DC, pp. 3–19.

Polson, E.J., Buckman, J.O., Bowen, D., Todd, A.C., Gow, M.M., Cuthbert, S.J.,

2002. An ESEM investigation into the effect of microbial biofilms on the

wettability of quartz. Seventh International Symposium on Wettability and

Its Effect on Oil Recovery, Tasmania, Australia.

Pooladi-Darvish, M., Firoozabadi, A., 1999. Solution-gas drive in heavy oil

reservoirs. Journal of Canadian Petroleum Technology 38, 54–61.

Portwood, J.T., 1995. A commercial microbial enhanced oil recovery

technology: Evaluation of 322 projects. SPE 29518. SPE Production

Operations Symposium, Oklahoma City, OK.

Price, N.P.J., Ray, K.J., Vermillion K., Kuo, T.-Min., 2009. MALDI-TOF mass

spectrometry of naturally occurring mixtures of monorhamnolipids and

dirhamnolipids. Carbohydrate Research 344, 204–209.

Prinzhofer A., 2012. Noble gas in oil and gas accumulations. In: Burnard, P.

(Ed.), The Noble Gases as Geochemical Tracers.

CRPG, Nancy, online at http://recherche.crpg.cnrs-

179

nancy.fr/IMG/pdf/Chapter_9_Prinzhofer_Final.pdf, accessed on October

22, 2012.

Priya, T., Usharani, G., 2009. Comparative study for biosurfactant production by

using Bacillus subtilis and Pseudomonas aeruginosa. Botany Research

International 2, 284–287.

Pruesse, E., Quast, C., Knittel, K., Fuchs, B., Ludwig, W., Peplies, J., Glöckner,

F.O., 2007. SILVA: a comprehensive online resource for quality checked

and aligned ribosomal RNA sequence data compatible with ARB. Nucleic

Acids Research 35, 7188–7196.

Rabus, R., Wilkes, H., Schramm, A., Harms, G., Behrends, A., Amann, R.,

Widdel, F., 1999. Anaerobic utilization of alkylbenzenes and n-alkanes

from crude oil in an enrichment culture of denitrifying bacteria affiliating

with the β-subclass of Proteobacteria. Environmental Microbiology 1,

145–157.

Rabus, R., Wilkes, H., Behrends, A., Armstroff, A., Fischer, T., Pierik, A.J.,

Widdel, F., 2001. Anaerobic initial reaction of n-alkanes in a denitrifying

bacterium: Evidence for (1-methylpentyl)succinate as initial product and

for involvement of an organic radical in n-hexane metabolism. Journal of

Bacteriology 183, 1707–1715.

Ramana, K.V., Karanth, N.G., 1989. Factors affecting biosurfactant production

using Pseudomonas aeruginosa CFTR-6 under submerged conditions.

Journal of Chemical Technology and Biotechnology 45, 249–257.

Rao, D.N., Ayirala, S.C., Abe, A.A., Xu, W., 2006. Impact of low-cost dilute

surfactants on wettability and relative permeability. SPE 99609.

SPE/DOE Symposium on Improved Oil Recovery, Tulsa, OK.

180

Rao, D.N., Girard, M., Sayegh, S.M., 1992. The influence of reservoir wettability

on waterflood and miscible flood performance. Journal of Canadian

Petroleum Technology 31, 47–55.

Rauf, M. A., Ikram, M., Tabassum, N., 2003. Enhanced oil recovery through

microbial treatment. Journal of Trace and Microprobe Techniques 21,

533–541.

Reinhold-Hurek, B., Hurek, T., 2006. The Genera Azoarcus, Azovibrio, Azospira

and Azonexus. In: Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-

H., Stackebrandt, E. (Eds.), The Prokaryotes, Vol 5., Springer, New York,

pp. 873–891.

Richter, M., Willey, J.M., Süβmuth, R., Jung, G., Fiedler, H.-P., 1998.

Streptofactin, a novel biosurfactant with aerial mycelium inducing activity

from Streptomyces tendae Tü 901/8c. FEMS Microbiology Letters 163,

165–171.

Ron, E. Z., Rosenberg, E., 2001. Natural roles of biosurfactants. Environmental

Microbiology 3, 229–236.

Ron, E.Z., Rosenberg, E., 2002. Biosurfactants and oil bioremediation. Current

Opinion in Biotechnology 13, 249–252.

Rosenberg, E., Ron, E.Z., 1999. High- and low-molecular-mass microbial

surfactants. Applied Microbiology and Biotechnology 52,154–162.

Rosenberg, M., 2006. Microbial adhesion to hydrocarbons: twenty-five years of

doing MATH. FEMS Microbiology Letters 262,129–134.

Rosenberg, M., Gutnick, D., Rosenberg, E., 1980. Adherence of bacteria to

hydrocarbons: a simple method for measuring cell-surface

hydrophobicity. FEMS Microbiology Letters 9, 29–33.

181

Ruckenstein, E., Nagarajan, R., 1975. Critical micelle concentration. A transition

point for micellar size distribution. The Journal of Physical Chemistry 79,

2622–2626.

Sarkar, A.K., Georgiou G., Sharma, M.M., 1994. Transport of bacteria in porous

media: I. An experimental investigation. Biotechnology and

Bioengineering 44, 489–497.

Sarkar, A.K., Sharma, M.M., Georgiou G., 1991. Compositional numerical

simulation of MEOR processes. In: Donaldson, E.C. (Ed.), Microbial

enhancement of oil recovery – recent advances, Elsevier, Amsterdam,

pp. 331–343.

Schloss, P.D., Westcott, S.L., 2011. Assessing and improving methods used in

operational taxonomic unit-based approaches for 16S rRNA gene

sequence analysis. Applied and Environmental Microbiology 77, 3219–

3226.

Schloss, P.D., Westcott, S.L., Ryabin, T., Hall J. R., Hartmann M., Hollister E.

B., Lesniewski R. A., Oakley B. B., Parks D. H., Robinson C. J., Sahl J.

W., Stres B., Thallinger G. G., Van Horn D. J., Weber C. F., 2009.

Introducing Mothur: Open-source, platform-independent, community-

supported software for describing and comparing microbial communities.

Applied and Environmental Microbiology 75, 7537–7541.

Schlumberger, 2012a. Interfacial tension. Oilfield Glossary, online at

http://www.glossary.oilfield.slb.com/Display.cfm?Term=interfacial%20ten

sion, accessed on December 2, 2012.

182

Schlumberger, 2012b. Wettability. Oilfield Glossary, online at

http://www.glossary.oilfield.slb.com/Display.cfm?Term=wettability,

accessed on December 2, 2012.

Senyukov, V. M., Yulbarisov, M. E., Taldykina, N. N., Shisherina, P. E., 1970.

Microbial method of treating a petroleum deposit containing highly

mineralized stratal waters. Mikrobiologiya 39, 705–710.

Seow, Y., Wood, M. J., 2009. Biological Gene Delivery Vehicles: Beyond Viral

Vectors. Molecular Therapy 17, 767–777.

Shahvali, M., Pooladi-Darvish, M., 2009. Dynamic modeling of solution-gas

drive in heavy oils. Journal of Canadian Petroleum Technology 48, 39–

46.

Sharma, P. K., McInerney, M. J., 1994. Effect of grain size on bacterial

penetration, reproduction, and metabolic activity in porous glass beads

chambers. Applied and Environmental Microbiology 60, 1481–1486.

Sharma, P. K., McInerney, M. J., Knapp, R. M., 1993. In situ growth and activity

and modes of penetration of Escherichia coli in unconsolidated porous

materials. Applied and Environmental Microbiology 59, 3686–3694.

Sheehy, A., 1992. Recovery of oil from oil reservoirs. Patent 5083610. U.S.

patent office.

Shen, C., Batycky, J., 1996. Some observations of mobility enhancemet of

heavy oils flowing through sand pack under solution gas drive. SPE 96-

27. Proceedings of the 47th Annual Technical Meeting of the Petroleum

Society, Calgary, AB.

Simpson, D.R., Natraj, N.R., McInerney, M.J., Duncan, K.E., 2011.

Biosurfactant-producing Bacillus are present in produced brines from

183

Oklahoma oil reservoirs with a wide range of salinities. Applied

Microbiology and Biotechnology 91, 1083–1093.

Smith, G.E., 1988. Fluid flow and sand production in heavy oil reservoirs under

solution-gas drive. SPE Production Engineering 3, 169–177.

Stepp, A.K., Bryant, R.S., Llave, F.M., Evans, D.B., Bailey, S.A., 1996. Microbial

methods for improved conformance control in porous media. SPE 35357.

SPE/DOE Improved Oil Recovery Symposium, Tulsa OK.

Sugihardjo, E.H., Partomo, L., Pratomo, S.W., 1999. Microbial core flooding

experiments using indigenous microbes. SPE 57306. SPE Asia Pacific

Improved Oil Recovery Conference, Kuala Lumpur.

Syldatk, C., Lang, S., Matulovic, U., Wagner, F., 1985. Production of four

interfacial active rhamnolipids from n-alkanes or glycerol by resting cells

of Pseudomonas species DSM 2874. Zeitschrift für Naturforschung C 40,

61–67.

Talabi, O., Pooladi-Darvish, M., Okazawa, T., 2003. Effect of rate and viscosity

on gas mobility during solution-gas drive in heavy oils. SPE 84032. SPE

Annual Technical Conference and Exhibition, Denver, CO.

Tan, T., Wang, B., Shao, Z., 2009. Donghicola xiamenensis sp. nov., a marine

bacterium isolated from seawater of the Taiwan Strait in China.

International Journal of Systematic and Evolutionary Microbiology 59,

1143–1147.

Tang, G., Firoozabadi, A., 1999. Gas and liquid-phase relative permeabilities for

cold production from heavy oil reservoirs. SPE 56540. SPE Annual

Technical Conference and Exhibition, Houston, TX.

184

Teramoto, M., Suzuki, M., Hatmanti, A., Harayama, S., 2010. The potential of

Cycloclasticus and Altererythrobacter strains for use in bioremediation of

petroleum-aromatic-contaminated tropical marine environments. Journal

of Bioscience and Bioengineering 110, 48–52.

Terry, R.E., 2001. Enhanced oil recovery. In: Meyers, R.A. (Ed.), Encyclopedia

of Physical Science and Technology, 3rd edition, Vol. 18, pp. 503–518.

The Engineering Toolbox, 2012. Solubility of gases in water. Online at

http://www.engineeringtoolbox.com/gases-solubility-water-d_1148.html,

accessed on November 10, 2012.

Torrens, J. L., Herman, D. C., Miller-Maier, R., 1998. Biosurfactant

(rhamnolipid) sorption and the impact on rhamnolipid-facilitated removal

of cadmium from various soils under saturated flow conditions.

Environmental Science and Technology 32, 776–781.

Town, K., Sheehy, A.J., Govreau, B.R., 2009. MEOR success in southern

Saskatchewan. SPE124319. SPE Annual Technical Conference and

Exhibition, New Orleans, LA.

Udegbunam, E.O., Adkins, J.P., Knapp, R.M., Mclnerney, M.J., Tanner, R.S.,

1991. Assessing the effects of microbial metabolism and metabolites on

reservoir pore structure. SPE 22846. The 66th Annual Technical

Conference and Exhibition, Dallas TX.

Updegraff, D. M., 1957. Recovery of Petroleum Oil. Patent 2807570. U.S.

Patent office.

Urgun-Demirtas, M., Stark, B., Pagilla, K., 2006. Use of genetically engineered

microorganisms (GEMs) for the bioremediation of contaminants. Critical

Reviews in Biotechnology 26:145–164.

185

U.S. Department of Energy, 2011. Enhanced oil recovery/CO2 injection. Fossil

Energy Office of Communications, online at

http://www.fe.doe.gov/programs/oilgas/eor/, accessed on October 3,

2012.

Van der Mei, H.C., Van de Belt-Gritter, B., Busscher, H.J., 1995. Implications of

microbial adhesion to hydrocarbons for evaluating cell surface

hydrophobicity. Colloids and Surfaces B: Biointerfaces 5, 117–126.

Van Hamme, J.D., Singh, A., Ward, O.P., 2003. Recent advances in petroleum

microbiology. Microbiology and Molecular Biology Reviews 67, 503–549.

Van Hamme, J.D., Singh, A., Ward, O.P., 2006. Surfactants in microbiology and

biotechnology: Physiological aspects (Part 1). Biotechnology Advances

24, 604–620.

Von Heningen, J., De Hann, A.J., Jansen, J.D., 1958. Process for the recovery

of petroleum from rocks. Netherlands Patent 80580.

Voordouw, G., 2011. Production-related petroleum microbiology: Progress and

prospects. Current Opinion in Biotechnology 22, 401–405.

Voordouw, G., Grigoryan, A.A., Lambo, A., Lin, S., Park, H.S., Jack, T.R.,

Miner, K., Arensdorf, J.J., 2009. Sulfide remediation by pulsed injection

of nitrate into a low temperature Canadian heavy oil reservoir.

Environmental Science and Technology 43, 9512–9518.

Wackett, L.P., Frias, J. A., Seffernick, J.L., Sukovich, D.J., Cameron S.M.,

2007. Genomic and biochemical studies demonstrating the absence of

an alkane-producing phenotype in Vibrio furnissii M1. Applied and

Environmental Microbiology 73, 7192–7198.

186

Wagner, M., 1985. Microbial enhancement of oil recovery from carbonate

reservoir with complex formation characteristics. In: Donaldson, E. C.

(Ed.), Microbial enhancement of oil recovery – recent advances, Elsevier,

Amsterdam, pp. 387–398.

Wagner, M., Lungerhausen, D., Murtada, H., Rosenthal, G., 1995. Development

and application of a new biotechnology of the molasses in-situ method;

detailed evaluation for selected wells in the Romashkino carbonate

reservoir. In: Bryant, R.S., Sublette, K.L. (Eds.), The Fifth International

Conference on Microbial Enhanced Oil Recovery and Related Problems

for Solving Environmental Problems, National Technical Information

Service, Springfield, VA, pp. 153–174.

Wagner, M., Lungerhausen, D., Nowak, U., Ziran B., 1993. Microbially improved

oil recovery from carbonate. In: Torma, A.E., Appel, M.L., Brierley C.I.

(Eds.), Biohydrometallurgical Technologies, Vol. II, Wallendale, PA,

TMS, Minerals-Metals-Materials, pp. 695–710.

Wang, X.Y., 1991. Advances in research, production and application of

biopolymers used for EOR in China. In: Donaldson, E. C. (Ed.), Microbial

enhancement of oil recovery: Recent advances, Development in

Petroleum Science Series, 31, Elsevier, Amsterdam.

Wang, Q., Fang, X., Bai, B., Liang, X., Shuler, P. J., Goddard III, W.A., Tang,

Y., 2007. Engineering bacteria for production of rhamnolipid as an agent

for enhanced oil recovery. Biotechnology and Bioengineering 4, 842–

853.

Wanger, G., Onstott T.C., Southam G., 2006. Structural and chemical

characterization of a natural fracture surface from 2.8 kilometers below

187

land surface: Biofilms in the deep subsurface. Geomicrobiology Journal

23, 443–452.

Wen, Y., Zhang, J., Yan, Q., Li, S., Hong, Q., 2011. Rhizobium

phenanthrenilyticum sp. nov., a novel phenanthrene-degrading bacterium

isolated from a petroleum residue treatment system. Journal of General

and Applied Microbiology, 57, 319–329.

Wilkes, H., Rabus, R., Fischer, T., Armstroff, A., Behrends, A., Widdel, F., 2002.

Anaerobic degradation of n-hexane in a denitrifying bacterium: Further

degradation of the initial intermediate (1-methylpentyl)succinate via C-

skeleton rearrangement. Archives of Microbiology 177, 235–243.

Willhite, G.P., Green D.W., 1998. Enhanced oil recovery. SPE Textbook series,

Vol. 6, SPE, Richardson, TX.

Yakimov, M.M., Genovese, M., Denaro, R., 2010. Thalassolituus. In: Timmis,

K.N. (Ed.), Handbook of hydrocarbon and lipid microbiology, Berlin,

Heidelberg, Springer-Verlag, pp. 1765–1772.

Yonebayashi, H., Ono, K., 1997. Microbial enhanced oil recovery field pilot in a

waterflooded reservoir. SPE 38070. SPE Asia Pacific Oil and Gas

Conference, Kuala Lumpur.

Youssef, N., Simpson, D. R., Duncan, K. E., McInerney, M. J., Folmsbee, M.,

Fincher, T., Knapp R. M., 2007. In situ biosurfactant production by

Bacillus strains injected into a limestone petroleum reservoir. Applied and

Environmental Microbiology 73, 1239–1247.

Youssef, N., Elshahed, M.S., McInerney, M.J., 2009. Microbial processes in oil

fields: Culprits, problems and opportunities. Advances in Applied

Microbiology 66, 141–251.

188

Youssef, N., Simpson, D.R., Duncan, K.E., McInerney, M.J., Folmsbee, M.,

Fincher, T., Knapp, R.M., 2007. In situ biosurfactant production by

Bacillus strains injected into a limestone petroleum reservoir. Applied and

Environmental Microbiology 73, 1239–1247.

Yue, J.C., Clayton, M.K.., 2005. A similarity measure based on species

proportions. Communications in Statistics (A) – Theory and Methods 34,

2123–2131.

Zengler, K., Richnow, H.H., Rosselló-Mora, R., Michaelis, W., Widdel, F., 1999.

Methane formation from long-chain alkanes by anaerobic

microorganisms. Nature 401, 266–269.

Zhang, R., Somasundaran, P., 2006. Advances in adsorption of surfactants and

their mixtures at solid/solution interfaces. Advances in Colloid and

Interface Science 123–126, 213–229.

Zhang, X., Li, B., Wang, H., Sui, X., Ma, X., Hong, Q., Jiang, R., 2012.

Rhizobium petrolearium sp. nov., isolated from oilcontaminated soil.

International Journal of Systematic and Evolutionary Microbiology 62,

1871–1876.

Zhang, X., Yue, S., Zhong, H., Hua, W., Chen, R., Cao, Y., Zhao, L., 2011. A

diverse bacterial community in an anoxic quinoline-degrading bioreactor

determined by using pyrosequencing and clone library analysis. Applied

Microbiology and Biotechnology 91, 425-434.

Zhang, Y., Miller, R.M., 1994. Effect of a Pseudomonas rhamnolipid

biosurfactant on cell hydrophobicity and biodegradation of octadecane.

Applied and Environmental Microbiology, 60, 2101–2106.

189

Zhang, Y., Xu, Z., Ji, P., Hou W., 1999. Microbial EOR laboratory studies and

application results in Daqing oil field. SPE 54332. SPE Asia Pacific Oil

and Gas Conference and Exhibition, Jakarta.

Zhao, H.-P., Wang, L., Ren, J.-R., Li, Z., Li, M., Gao, H.-W., 2008. Isolation and

characterization of phenanthrene-degrading strains Sphingomonas sp.

ZP1 and Tistrella sp. ZP5. Journal of Hazard Materials 152: 1293–1300.

Zhaowei, H., Peihui, H., Jianjun, L., Jianfei, C., Xumou, D., Menghua, G.,

Xinghong, C., 2008. The application of hydrocarbon-degrading bacteria

in Danqing’s low permeability, high paraffin content oilfields. 2008

SPE/DOE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA,

19-23 April 2008 – SPE 114344.

ZoBell, E. C., 1947. Bacterial release of oil from oil-bearing materials (Part 1).

World Oil 126, 36–47.

Zoueki, C.W., Ghoshal, S., Tufenkji, N., 2010. Bacterial adhesion to

hydrocarbons: role of asphaltenes and resins. Colloids and Surfaces B:

Biointerfaces 79, 219–226.

190

Appendices

Table A1. Volumes of solutions injected into set I – IV columns with trapped EO and LO before the beginnings of long-time incubation rounds. Columns I-1, I-2, II-1, II-2, III-1, III-2, IV-1, IV-3, IV-4 and IV-5 contained nitrate.

Volume of *Volume of Volume of Volume of an an CSB-K CSB-K enrichment enrichment medium medium culture culture injected injected injected Column Type Column injected before the before the before the set of oil number before the beginning beginning beginning beginning of each of the first of each of the first following incubation following incubation incubation round, mL incubation round, mL round, mL round, mL I EO I-1 11.5 2 13.5 0 I EO I-2 13.5 0 13.5 0 I EO I-3 13.5 0 13.5 0 II EO II-1 8.5 1.5 10 0 II EO II-2 10 0 10 0 II EO II-3 10 0 10 0 III EO III-1 8.5 1.5 10 0 III EO III-2 10 0 10 0 III EO III-3 10 0 10 0 IV LO IV-1 13.5 0 13.5 0 IV LO IV-2 13.5 0 13.5 0 IV LO IV-3 11.5 2 13.5 0 IV LO IV-4 13.5 0 13.5 0 IV LO IV-5 11.5 2 13.5 0

*In the case of an enrichment culture injection, a culture was mixed with CSB-K medium prior to an injection.

191

Table A2-1. Rounds of long-time incubations for column set I (columns I-1 and I-2 contained nitrate)

Days when new rounds started 0 39 64 131 195 237 Concentration of nitrate in columns I-1 and I-2 20 20 20 50 20 20 at the beginning of the round, mM Total incubation time, days 341

Table A2-2. Rounds of long-time incubations for column set II (columns II-1 and II-2 contained nitrate)

Days when new rounds started 0 62 90 156 220 262 366 Concentration of nitrate in columns II-1 and II-2 20 20 20 50 20 20 20 at the beginning of the round, mM Total incubation time, days 401

Table A2-3. Rounds of long-time incubations for column set III (columns III-1 and III-2 contained nitrate)

Days when new rounds started 0 49 101 165 207 Concentration of nitrate in columns III-1 and III-2 20 20 20 50 20 at the beginning of the round, mM Total incubation time, days 311

Table A2-4. Rounds of long-time incubations for column set IV (columns IV-1, IV-3, IV-4 and IV-5 contained nitrate) Days when new rounds started 0 37 68 Total incubation time, days 112

192

(a) Obtain columns with trapped oil ↓ (b) Transfer them to an anaerobic hood ↓ (c) Inject media with or without nitrate ↓ (d) Incubate @30ºC for one month or longer

(e1) Repeat (e2) Remove from steps (c) and (d) an anaerobic hood several times and subject to water flood (3 PV) ↓ Determine amounts of oil and water (f) Bring back to produced at steps (c), (d), (e) and (f); an anaerobic monitor nitrate and nitrite hood and repeat concentrations in produced water steps (c) ,(d) and (e2) several times

Figure A1. Experiments with set I – IV columns containing trapped oil and nitrate-containing or nitrate-deficient media.

I-1, II-1 and III-1: 1 PV of enrichment culture in CSB-K medium (15:85, v/v), 20mM NO3¯

I-2, II-2 and III-2: 1 PV of CSB-K medium (no enrichment culture), 20mM NO3¯

I-3, II-3 and III-3: 1 PV CSB-K medium, no NO3¯

*Replenishments with the media (1 PV) were conducted after one-month or longer-time incubations

Figure A2. Nitrate-containing and nitrate-deficient media introduced into set I – III columns (with trapped EO) prior to long-time incubation rounds.

193

IV-1: 1 PV of CSB-K, 20mM NO3¯ IV-2: 1 PV of CSB-K without NO3¯ IV-3: 1 PV of enrichment culture in CSB-K medium (15:85, v/v), 90mM NO3¯ IV-4: 1 PV of CSB-K, 500 mM NO3¯ IV-5: 1 PV of enrichment culture in CSB-K medium (15:85, v/v), 900mM NO3¯

*Replenishments with the media (1 PV) were conducted after one-month or longer-time incubations

Figure A3. Nitrate-containing and nitrate-deficient media introduced into set IV columns (with trapped LO) prior to long-time incubation rounds.

194

A column with trapped oil

PROCEDURE 1 PROCEDURE 2 a) inject 1PV of CS or surfactant solution; b) measure amount of produced oil; a) flood with 10 PV of CS or c) incubate for 24 hours; surfactant solution; d) flood with 3 PV of SPW; b) measure amount of produced e) measure amount of produced oil; oil f) repeat steps a)-e) several times

♦ incubate a column in an anaerobic hood @ 30ºC for several months (a long-time incubation); ♦ replenish with CS or surfactant solution monthly (1 PV)

Figure A4. Experiments involving introduction of CSs or surfactant solutions into columns containing trapped oil.

195