- Email: [email protected]
Study of enhanced oil recovery by rhamnolipids in a homogeneous 2D micromodel
Journal of Petroleum Science and Engineering 128 (2015) 212–219
Contents lists available at ScienceDirect
Journal of Petroleum Science and Engineering journal homepage: www.elsevier.com/locate/petrol
Study of enhanced oil recovery by rhamnolipids in a homogeneous 2D micromodel Hossein Amani n Faculty of Chemical Engineering, Babol Noshirvani University of Technology, Babol, Iran
art ic l e i nf o
a b s t r a c t
Article history: Received 17 September 2014 Accepted 18 February 2015 Available online 26 February 2015
In this study, homogeneous glass micromodel is used for investigation of oil recovery by rhamnolipid. The intention of this study was to investigate whether the rhamnolipid mixture could be produced in commercial quantities for enhanced oil recovery (EOR) projects in bioreactor and prove of its potential use as an effective material for field application. In this work, the ability of Pseudomonas aeruginosa HATH to grow and produce rhamnolipid on sunflower as a sole carbon source under nitrogen limitation was shown. The production of Rha-C10-C10 and Rha2-C10-C10 was confirmed by thin layer chromatography (TLC) and high-performance liquid chromatography (HPLC) analysis. The yield of rhamnolipid per biomass (YRL/x), rhamnolipid per sunflower oil (YRL/s), and the biomass per sunflower oil (Yx/s) for bioreactor were obtained about 0.54 g g 1, 0.059 g g 1, and 0.11 g g 1, respectively. The rhamnolipid mixture obtained was able to reduce the surface and interfacial tension of water to 26 and 2 mN/m, respectively. Produced rhamnolipid is an effective surfactant at very low concentrations over a wide range of temperatures, pHs and salt concentrations and also has the ability to emulsify oil, which is essential for enhanced oil recovery. With a critical micelle concentration, 5% of original oil in place was recovered after water flooding from a micromodel. This result suggests rhamnolipids as appropriate model biosurfactants for microbial enhanced oil recovery. & 2015 Elsevier B.V. All rights reserved.
Keywords: bioreactor biosurfactant fermentation microbial enhanced oil recovery micromodel rhamnolipid
1. Introduction Microbial enhanced oil recovery (MEOR) is the use of microbes in petroleum reservoirs to enhance the amount of oil that can be produced (Bordoloi and Konwar, 2008; Sen, 2008; Amani et al., 2010a,b; Al-Bahry et al., 2013; Youssef et al., 2013). The microbes use hydrocarbons as a food for their metabolic processes and excrete natural and non-toxic bio-products such as biosurfactants and biopolymers. Some of the advantages of biosurfactants over conventional surfactants include lower toxicity, high biodegradability, tolerance to at extreme temperatures, pH and salinity and their potential applications in environmental protection and management (Joshi et al., 2008; Amani et al., 2010a,b; Chrzanowski et al., 2012; Ismail et al., 2012; Ławniczak et al., 2013; Youssef et al., 2013). Rhamnolipids are among the best known biosurfactants and have been proven to be very promising in enhanced oil recovery (Banat, 1995; Wang et al., 2007; Sen, 2008; Amani et al., 2010a,b; Hörmann et al., 2010; Müller et al., 2010; Youssef et al., 2013). The rhamnolipids can change the physical and chemical properties of the crude oil and stimulate oil–water–rock interactions that improve oil recovery. They are the most effective biosurfactants with the ability to reduce the water surface tension
n
Tel.: þ 98 1113232071; fax: þ 98 1113220570. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.petrol.2015.02.030 0920-4105/& 2015 Elsevier B.V. All rights reserved.
from 72 to 30 mN/m as well as interfacial tension reduction in water/ oil system from 43 to below 1 mN/m at trace concentrations (Amani et al., 2010a,b; Hörmann et al., 2010). The following positive effects in a reservoir have been documented: (i) biosurfactants reduce the interfacial tensions between oil/rock and oil/water, improving oil flow, (ii) if reservoir is oil wet, biosurfactants increase the wettability of the rock toward water, causing the rock to be preferentially wet by water, dislodging the oil, again helping it flow more freely, and (iii) biosurfactants reduce the interfacial tension between water and oil, and therefore a lower hydrostatic pressure is required to move the liquid entrapped in the pores to overcome the capillary effect. Rhamnolipid always consists of one or two units of rhamnose linked to one or two fatty acid chains with C8–C14 carbon atoms, which may or may not be saturated (Hörmann et al., 2010; Müller et al., 2010). Four different rhamnolipid homologs, produced by Pseudomonas aeruginosa, have been identified and characterized (Hörmann et al., 2010; Müller et al., 2010). The focus of the proposed study is production of rhamnolipids, purification and identification of rhamnolipid, and a visual study of enhanced oil recovery by rhamnolipid flooding in a homogeneous 2D micromodel. In recent years, micromodels have been served as an excellent laboratory instrument to investigate and understand the mechanisms of biosurfactant flooding in removing the residual oil. Micromodels provide the opportunity to observe fluid flow within reservoirs
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
(Stewart and Kim, 2004; Soudmand-asali et al., 2007; Sayegh and Fisher, 2009). Several investigations have been carried out on the in situ bacteria flooding but the ex situ rhamnolipid production and addition to the micromodel as agents for MEOR has little been studied. In order to develop a suitable technology for ex situ MEOR processes, it is essential to carry out tests about ex situ enhanced oil recovery by rhamnolipids. Therefore this work tries to fill the gap. Another intention of this study was to investigate whether the rhamnolipid mixture could be produced in commercial quantities for EOR processes in the bioreactor and to prove its potential use as an effective material for reservoirs in Iran.
2. Materials and methods 2.1. Microorganism The biosurfactant producing strains were isolated from petroleum contaminated garage site using oil spreading method as described by Youssef et al. (2007). The isolation and screening of the biosurfactant producing bacteria were performed in the Tehran University Institute of Biotechnology. Those isolates which showed bigger zone of clearance on oil layer and reduced surface tension (ST) below 30 mN/m were selected for further experiments. One strain (HATH) was identified as P. aeruginosa according to its 16SrRNA sequences by Pasteur Institute (Paris, France). The strain was kept in laboratory of Industrial Microbiology at Tehran University (Tehran, Iran) and nominated as P. aeruginosa HATA. 2.2. Media and culture conditions Lysogeny broth (LB) was used for pre-cultivation step one. For the culture, a nitrogen-limited production medium consisting of 250 g/L sunflower oil (Bellasan, Aldi Süd, Rastatt, Germany) and a Ca-free mineral salt solution with 1.5 g/L NaNO3, 0.05 g/L MgSO4 7H2O, 0.1 g/L KCl, containing a 0.1 M sodium phosphate buffer at pH 6.5 was used. A total of 1 mL/L of trace element solution was added. The trace element solution contained 2.0 g/L sodium citrate 2H2O, 0.28 g/L FeCl3 6H2O, 1.4 g/L ZnSO4 7H2O, 1.2 g/L CoCl2 6H2O, 1.2 g/L CuSO4 5H2O, and 0.8 g/L MnSO4 H2O. Trace elements were filter-sterilized through a 0.22 μm membrane filter (Carl Roth GmbH, Karlsruhe, Germany). Production medium was adjusted to pH 6.5 with HCl (6 N) and NaOH (1 N). Mineral salt solutions, phosphate sources, and sunflower oil were autoclaved separately for all experiments (Müller et al., 2010). All shake flask cultures were incubated in a shake incubator chamber (Multitron II, HT Infors, Bottmingen, Switzerland). First 25 mL of LB in a 100-mL baffled shake flask was inoculated with a total volume of 100 μL from the glycerol stock solution of P. aeruginosa HATH and, incubated for 24 h at 37 1C and 120 rpm. The second pre-cultures also containing 25 mL LB were cultivated in a 100-mL baffled shake flask and inoculated using a total of 0.5 mL from the 24-h LB culture. This culture was incubated for 24 h at 37 1C, 120 rpm. The production was started after inoculation with the second LB pre-culture resulting in OD580 of 0.05 in the aqueous phase at the start of the bioreactor cultivation. 2.3. Bioreactor experiment Fermentation was carried out in an integrated stirred tank bioreactor with accessories and automatic systems for dissolved oxygen (DO), pH, impeller speed, aeration rate, and temperature (Infors AG, Heado fice, Switzerland). The reactor used was a 2.5-L batch stirred bioreactor with three baffle plates provided to reduce vertex effects and enhance the mixing. Two Rushton impellers
213
attached to the same shaft were used for agitation. Air was sparged through a pipe sparger placed below the bottom impeller. A foam collector was connected to the top of the bioreactor to recover the biosurfactant by withdrawing the foam produced during the process. Stirrer speed was set fix at 600 rpm, temperature at 37 1C and the aeration rates at 2 vvm. For the cultivation, 0.55 L of the production medium was used. The trace element solution was added at cultivation times of 0, 20, 40, 70, and 120 h, as described before (Müller et al., 2010). The production was carried out for 9 days. According to the analysis purposes, sampling was undertaken on a daily basis. 2.4. Analysis of sunflower oil, rhamnolipid and biomass Culture suspension was mixed vigorously with n-hexane 1:1 (v/v) and centrifuged (4600 g, 4 1C, 30 min) for separation of cells, aqueous and n-hexane phase. The n-hexane phase was used for gravimetric determination of sunflower oil concentrations, after evaporation of n-hexane. We can only analyze the extractable portion of the sunflower oil and at t¼0 h, it is only the theoretical value of 250 g/L. For rhamnolipd measurement, an aliquot of the aqueous phase was acidified with 85% phosphoric acid 1:100 (v/v) to adjust a pH of about 2–3, leading to precipitation of the rhamnolipids. Rhamnolipids were extracted twice with ethyl acetate 1:1.25 (v/v). A 10 mL of ethyl acetate extracts was used for rhamnolipid identification by TLC. The TLC was performed according to Müller et al. (2010). A rhamnolipid standard was prepared from Jeneil JBR425 (Jeneil Biosurfactants Company, Saukville, United States). Appropriate amounts of ethyl acetate extracts were evaporated and used for rhamnolipid quantification by HPLC (Hörmann et al., 2010; Müller et al., 2010). For biomass measurement, the conventional cell dry weight measurement method was used to determine biomass dry weight. Following the centrifugation of sample and separation of hexane phase and aqueous phase from the biomass, the biomass was washed once in 0.9% NaCl solution (4600 g, 4 1C, 30 min) in distillated water and then it was transferred into a pre-tared vial. The bacterial dry weight was determined after drying at 1051 C for 24 h. A complete working scheme for analytics is shown in Fig. 1. The surface tension (ST), interfacial tension (IFT) and critical micelle concentration measurement (CMC) were measured at 251 C by a digital tensiometer (Kruss, K10ST, Germany) using the ring method. IFT measurements were carried out against crude oil (API¼ 341, acid number ¼0.2 mg KOH/g oil, base number ¼ 0.05 mg KOH/g oil). The CMC was determined by measuring surface tension at different concentrations of rhamnolipid in distillated water up to a constant value of surface tension saturated (Amani et al., 2010a,b; Hörmann et al., 2010). 2.5. Stability studies Rhamnolipid produced by P. aeruginosa HATH was used for stability studies. Samples of rhamnolipid at CMC concentration (5 cc of the produced rhamnolipid was poured into a test tube containing 5 mL of crude oil (water oil ratio¼1)) were checked for stability of surface activity (ST) under different environmental conditions: High temperatures (40, 60, 70, 80, 90 and 100 1C for 60 min and at 120 1C for 20 min), different salt concentrations (NaCl: 0, 1, 2, 4, 6, 8, 10, 15 and 20%, w/v) and a wide range of pH (2, 4, 6, 7, 8, 10 and 12). The pH was adjusted with HCl (6 N) and NaOH (1 N). 2.6. Emulsification and oil spreading studies Emulsification activity of the rhamnolipid solutions at CMC concentration was determined by measuring the emulsion index (E24) at 25 1C. In brief, 4 ml of crude oil was poured separately into
214
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Taking sample from bioreactor and add 1 volume of n-hexane for extraction of oil
4700rpm; 4°C; 30min
Gravimetric analysis of oil in n-hexane phase
Mix 2mL aqueous phase with 20µL H PO
Wash pellet once with 0,9% saline
2,5mL EtAc for RL extraction
10min at 7500rpm
remove supernatant for rhamnolipid analysis
Repeat extraction
HPLC analysis
Thin layer chromatography
Fig. 1. Working scheme for analytics.
a test tube containing 4 ml of biosurfactant solution. A biosurfactant solution consisted of the following composition: 120 mg rhamnolipid (CMC), 0.051 g NaHCO3, 75 g NaCl, 0.61 g KCl, 9.2 g CaCl2 and 7.6 g MgCl2 per liter, close to the specified reservoir brine. After being vigorously mixed for 2 min, the test tube was kept still for 24 h and the heights of emulsion, oil and aqueous zones were measured. The emulsion index (E24) was then calculated from the ratio of the height of the emulsion zone to the total height of the oil, emulsion, and aqueous zones (Yeh et al., 2005). To investigate the stability condition, test tubes were kept still for 15 days and then measurement of the emulsion index was undertaken again. In order to investigate the effects of the biosurfactant produced on the oil and aqueous phases, oil spreading technique was carried out according to Youssef et al. (2013). Briefly, fifty milliliter of distilled water was added to the petri dishes followed by the addition of 100 μl of crude oil to the surface of water. Then 10 μl of the rhamnolipid solutions at CMC concentration was dropped on to the crude oil surface. The diameter of clear zone on the oil surface was measured and compared to 10 μl of distilled water as negative control. 2.7. Determination of optimum salinity The salinity plays an important role in achieving low IFT between both the microemulsion and oil, and the microemulsion and brine. The optimum salinity is defined as the salinity where the IFTs are equal (Healy et al., 1976; Roshanfekr and Johns, 2011). The optimal salinity is alternatively defined as the salinity where the solubilization ratios for brine and oil are equal (Roshanfekr and Johns, 2011). In this research, the water to oil ratio value (WOR¼1)and biosurfactant concentration (CMC) were kept constant to find a possible trend by changing the oil and water solubilization ratios at different salinities. The solubilization ratio is defined as δi ¼ Vi/Vs, where Vi is the volume of oil or water solubilized and Vs is the volume of surfactant in the microemulsion phase. Tests were conducted at WOR¼1 in borosilicate 10 cc pipettes. Each pipette contained 1.2 wt% biosurfactant,
crude oil and brines of X% NaCl. The pipettes were sealed with a torch and samples were rotated end-to-end for 5 days and then left to stand quiescent. The oil and water solubilization ratios were measured after one week (equilibrium was assumed). 2.8. Micromodel studies and experimental setup As the average oil reservoir temperature is about 80 1C, the application of rhamnolipid in enhanced oil recovery was evaluated using micromodel at 80 1C. Glass etched micromodel was constructed for this investigation. The glass micromodel is a two dimensional flow channel network etched in a glass to simulate the fluid flow in porous media. Glass is composed mainly of sand (silicates, SiO2) and silicates which show slight adsorption of anionic surfactants, but adsorb strongly cationic surfactants (Bubela, 1989). In this work, we assume that the glass micromodel has a slight adsorption of biosurfactant in pores because the produced rhamnolipid is an anionic biosurfactant (Sreekala and Gina,1994). Irregular and dead-pore network patterns were drawn according to Soudmand-asali et al. (2007). In our work, one identically patterned micromodel was used. The network pattern of the micromodel has been shown in Fig. 2. The physical and hydraulic properties of the micromodels such as porosity (%), etched thickness (mm), permeability (mDarcy) and pore volume (cm3) were 41.5, 0.057, 100 and 0.335 respectively. Crude oil (API¼ 341) was used as the oleic phase. The crude oil was obtained from the College of Chemical Engineering of University of Tehran, Iran. Glass micromodels are usually considered strongly water-wet because their surface chemistry is similar to that of clean sandstone (Grattoni et al., 2002). In our work, we assume that no wettability alteration occurs in the experiments. To investigate oil displacement, first the model was saturated with 5% (by weight) NaCl solution, and then the crude oil injected at the rate of 4 mL/h until the condition of connate water saturation was reached (the crude oil displaced the brine until no more brine was produced). At the end of imbibition process, brine was injected at a rate of
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
215
0.5 ml/h until no more oil was produced in the effluent. At this point, we injected the rhamnolipid of P. aeruginosa HATH at CMC concentration for a period of 2 days and a flow rate of 0.5 mL/h instead of brine. Difference between oil residual saturation in experiments was equal to the amount of recovery factor due to rhamnolipid injection after water flooding. The experimental setup has been shown in Fig. 3.
3. Results and discussions 3.1. Production of rhamnolipid in bioreactor The cultivation of P. aeruginosa HATH at 37 1C and 600 rpm in the integrated bioreactor was performed three times. The profiles of biomass concentration, rhamnolipid production, and sunflower oil concentration versus time in the mentioned conditions are presented in Fig. 4. Fig. 4 shows that the maximum biomass of about 22 g/L was reached after 84 h of growth, giving a yield of biomass on sunflower oil (Yx/s) of 0.11 g g 1. Approximately biomass, sunflower oil and rhamnolipid concentration were stable after 100 h. After 24 h of cultivation rhamnolipid concentrations of 0.05 g/L were quantified. Rhamnolipid production continued up to 120 h of cultivation and reached a maximum of approximately 12 g/L. Specific growth rate was 0.04 h 1, ascertained from the slope of the plot of Ln (biomass) versus time (from 10 to 40 h).The rhamnolipid production was not growth related, which is in accordance to literature data and a typical behavior of secondary metabolite (Hörmann et al., 2010; Müller et al., 2010). The maximum of the production rate was calculated for between 38 h and 42 h of cultivation (0.19 g L 1 h 1) in the transition towards stationary phase. However, small quantities of rhamnolipid were already produced in the early growth phase. The yield of
Fig. 3. Experimental setup for oil displacement study.
Fig. 4. Time course profiles of P. aeruginosa HATH cell growth, sunflower oil concentration, and rhamnolipid concentration at 600 rpm, 2 vvm, and 37 1C in the integrated bioreactor containing 0.55 l medium.
rhamnolipid per biomass (YRL/x) and the production yield (YRL/s) was 0.54 g g 1 and 0.059 g g 1, respectively. For a better comparison between different researches, one parameter such as Y was calculated and presented in Table 1. 3.2. Characterization of biosurfactant produced by P. aeruginosa HATH
Fig. 2. Porous pattern of the micromodel.
A rhamnolipid standard was prepared from JBR (Jeneil Biosurfactants Co., Saukville, USA). The JBR is a mixture of an undefined number of rhamnolipid congeners of P. aeruginosa mainly containing di- and mono-rhamnolipids with one and two β-hydroxydecanoic acids, respectively (four different rhamnolipid homologs). To confirm the rhamnolipid production, the TLC of commercially available rhamnolipid (JBR425, Jeneil Biosurfactant Company) and rhamnolipid of this study were compared. Consequently, the rhamnolipids of P. aeruginosa appear on the TLC plate by decreasing hydrophobicity: Rha-C10-C10 (R1), Rha-C10 (R2), Rha2-C10-C10 (R3) and Rha2-C10 (R4). The TLC analysis of culture supernatants and their respective organic extracts of the strain displayed two spots for P. aeruginosa HATH which could be stained positively for sugar and fatty acid moieties. An example is shown in Fig. 5. Thin-layer chromatography (TLC) results suggest that the isolated surface-active products from P. aeruginosa HATH contain glycolipids (R1 and R3) which showed a similar retention behavior
216
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Table 1 Comparison of the results obtained for batch bioreactor cultivation of P. aeruginosa strains with comparable studies. P. aeruginosa strain
Carbon source (g/L)
Process
Process time (h)
Maximum rhamnolipid concentration (g/L)
Y (g L 1 h 1)
Ref.
PAO1 HATH UI 29791 DSM 7107 DSM 7108
Sunflower oil (250) Sunflower oil (250) Corn oil (75) Soybean (125) Soybean (163)
Batch Batch Batch Batch Batch
90 120 192 166 267
39 12 46 78 112
0.43 0.19 0.24 0.47 0.42
Müller et al. This study Müller et al. Müller et al. Müller et al.
(30 L) (2.5 L) (14 L) (30 m3) (300 L)
RL 1
RL 2
RL 3
RL 4
Jeneil standard
produced rhamnolipid
Fig. 5. TLC of surface-active product from P. aeruginosa HATH and from commercially available rhamnolipid (JBR425, Jeneil Biosurfactant Company).
to the commercially available Jeneil rhamnolipid mixture. The identity of the two glycolipids was further confirmed by HPLC analysis. To confirm the rhamnolipid production, the HPLC chromatogram of pure rhamnolipid standard solutions of Rha-C10-C10 and Rha2-C10-C10 and rhamnolipid of this study were compared. With this method, the production of rhamnolipids with similar retention times as Rha-C10-C10 and Rha2-C10-C10 by P. aeruginosa HATH could be proved. Smaller peaks apart from these could either be related to other congener types or also occur due to fatty acids or mono/diglycerides, which are generated by the cleavage of the carbon source. These substances are also present in the organic extracts and they are probably also derivatised by 4-bromo phenacyl bromide, due to their acid function. However, significant HPLC peaks were detected for our products. Peaks with a similar retention time to the di-rhamnolipid and mono-rhamnolipid of a commercially available rhamnolipid were found. Therefore we think our strain has produced a glycolipid or rhamnolipid. 3.3. Effect of produced rhamnolipid on surface tension, emulsification activity and oil spreading The production of biosurfactant by P. aeruginosa HATH showed excellent reduction of water surface tension from 72 to 25 mN/m. Our results also showed that the minimum value of interfacial tension (IFT) for produced rhamnolipid was 2 mN/m. The reduction of IFT between oil and water is very important in oil recovery because capillary number is defined as the ratio of viscous to capillary forces.
(2010) (2010) (2010) (2010)
Capillary number increases with decrease in interfacial forces. Increases in capillary number lower the residual oil saturation in the core and increase residual oil recovery (Sen, 2008). For further investigation we determined the CMC of the obtained rhamnolipid mixture. The CMC of the rhamnolipid was 120 mg/L. Emulsification activity of the produced rhamnolipid at CMC concentration was determined by measuring the emulsion index (E24). A maximum emulsion index of 85% was reached for the crude oil. The potential of the biosurfactant oil dispersion was also evaluated using the rhamnolipid produced by the P. aeruginosa HATH. The rhamnolipid at CMC concentration showed high dispersability of the crude oil. Fig. 6 shows that if 10 μl of the produced rhamnolipid is gently placed on the center of the oil layer, the crude oil is displaced and a clearing zone is formed. The diameter of this clearing zone on the oil surface correlates to surfactant activity, also called oil displacement activity. Diameter variations during one week are also shown in Fig. 6. As shown in the figure, the crude oil emulsion remained stable even after one week. However, the results showed that the produced rhamnolipid is capable of reducing the surface and interfacial tensions of water to significant lower value, and having excellent emulsifier and dispersion properties. These properties suggest that biosurfactants isolated from P. aeruginosa HATH are appropriate candidates for enhanced oil recovery. 3.4. Study of stability To check the stability of biosurfactant, samples of rhamnolipid (120 mg/L) were subjected to different conditions. At pH 2.0–4.0, we observed much higher ST; since the biosurfactant is not soluble at such acidic conditions, it tends to precipitate. The effect of pH on biosurfactant activity showed that when pH of biosurfactant decreased, the surface tension increased. It could be the result of biosurfactant precipitation in acidic condition. Thus precipitated and structurally distorted biosurfactant loses its capability of reducing surface tension. However, the effects of pH on surface tension showed that the rhamnolipid has an effective surface activity from pH ¼4 to pH ¼10.0. Several reports confirm the stability of biosurfactant at different pH values mostly in the alkaline medium (Pimientar et al. 1997; Batista et al., 2005; Amani et al., 2010b; Lovaglio et al., 2011; Raheba and Hajipoura, 2011, Amani et al., 2013). We also observed similar stability trend in our biosurfactant. Studies on the effect of heat treatment showed that no change in biosurfactant surface activities has occurred and there is no difference in surface tension before and after this treatment in all cases. The CMC value remained stable after exposure to high temperature (80–100 1C) even after 1 day. When placed to autoclave (120 1C for 20 min) the surface activity was also maintained. So it could be interpreted as a good heat resistance of rhamnolipid produced by P. aeruginosa HATH. Several studies confirm the stability of biosurfactants under extreme conditions of temperature. Amani et al.
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
217
Fig. 6. Oil spreading of produced rhamnolipid (CMC concentration) in a glass culture plate (90 mm diameter) which contained 50 ml water in the bottom and 100 μl of crude oil (API ¼ 34) at the surface. Dispersants' activities to the crude oil observed.
surface tension was measured. The results showed that the values of the ST resisted well against salt concentration up to 25 g/L. We know oil reservoirs are one of the harsh environments, where temperature can range from 20 to 90 1C, normal salinity to hypersalinity and pH over a wide range. So, these findings suggest that the rhamnolipid produced by the P. aeruginosa HATH is a superior surfactant for application under the extreme conditions which usually prevail in enhanced oil recovery. In other words the produced rhamnolipid meets those criteria and showed stability over a wide range of environmental factors, while retaining its surface activity. 3.5. Interfacial tension and solubilization parameters
Fig. 7. Solubilization parameter versus salinity (WOR ¼ 1, T ¼ 25 1C).
Fig. 7 represents the corresponding behavior of the solubilization parameters with different salinities. From Fig. 7, the oil solubilization ratio increases and water solubilization ratio decreases as salinity increases. The results show that the measured solubilization ratio at the optimum salinity for the crude oil is 10.5, while the optimum salinity is 2.4 wt%. The optimum salinity occurs where these two curves intersect at 2.4 wt% (24,000 ppm). At this condition (optimum salinity), the IFT between the microemulsion and the excess oil phase, and the IFT between the microemulsion and water phase were measured. Our results showed that IFTs were the same and about 0.25 mN/m. As discussed in Section 3.3, if interfacial tension is small, capillary number will be large enough to let the residual oil saturation go to zero. This is one of the main mechanisms for enhanced oil recovery with biosurfactants. However, the phase behavior of microemulsions is very important to enhanced oil recovery because it can be used as an indicator of ultra-low interfacial tension. 3.6. Study of oil recovery after waterflooding
Fig. 8. Relationship between water and rhamnolipid flooding with pore volume in micromodel.
(2013) reported that biosurfactant produced by P. aeruginosa MM1011 was stable at 120 1C. Bordoloi and Konwar (2008) exposed the biosurfactant produced by P. aeruginosa strains to a temperature of 100 1C for different time periods of 5–60 min and found that the biosurfactant activity remained unaffected with respect to ST changes. For salinity effect, various amounts of NaCl were added to rhamnolipid at CMC concentration and mixed completely and then
To check the effect of biosurfactant on enhanced oil recovery, micromodel experiments were carried out. The initial oil saturation (Soi) of the micromodel was calculated to be 71% after oil flooding. A 120 mg/L solution of the rhamnolipid (CMC concentration) was prepared and used to enhance oil recovery in micromodel. In our work, first a brine solution was introduced at 0.5 mL/h into the micromodel so that this aqueous phase now occupies all of the pore space. Then, crude oil (4 mL/h) displaced the brine in the micromodel until no more brine was produced. The brine was injected again at 0.5 mL/h into the micromodel until no more oil came out. So at this point, the micromodel with trapped oil was ready for flooding with the produced rhamnolipid. Finally, the rhamnolipid was injected at 0.5 mL/h. The oil recovery was recorded during the brine and the rhamnolipid injection steps. Fig. 8 shows relationship between water and rhamnolipid flooding with pore volume injected in micromodel. According to Fig. 8, it was found that after injecting
218
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
4. Conclusions The intention of our research was to investigate whether our rhamnolipid has high enough quantities of enhanced oil recovery in the laboratory scale to be used as an effective material for field application. In this work, rhamnolipid production and growth characteristics of P. aeruginosa HATH using sunflower oil as substrates were studied. Maximal rhamnolipid concentration of 12 g/L was attained after 84 h of cultivation and, at this point, most of the sunflower oil was utilized. The production of rhamnolipid was confirmed by TLC and HPLC analysis. The excellent ST and IFT reducing characteristic of biosurfactant produced by P. aeruginosa HATH and its stability over a wide range of temperature, pH and different salt concentrations imply the possibility of using this biosurfactant in enhanced oil recovery. The produced biosurfactant is capable of having excellent emulsifier and dispersion properties. Finally, the ability in recovering the oil from oilsaturated glass was also demonstrated. Nearly 5% of original oil in place was recovered using the rhamnolipid at CMC concentration after water flooding in micromodel. This work was demonstration of successfulness of use of rhamnolipids for industrial applications, especially in MEOR.
References
Fig. 9. Photographs of micromodel when it is completely saturated by oil. (a) Micromodel after water flooding and (b) micromodel after rhamnolipid flooding at CMC concentration.
6 PVof brine at 0.5 mL/h, no more oil was produced. Extra oil recovery occurred after injecting 2 PV of the produced biosurfactant at 0.5 mL/h. As shown in the figure, the recovery value of oil after water flooding was obtained as 30% of original oil in place, whereas nearly 5% of original oil in place was recovered using rhamnolipid after water flooding. The experiment was performed three times. Our oil displacement experiment demonstrates that a very low concentration (120 mg/L) of the rhamnolipid can recover a significant fraction of trapped oil from a micromodel. There are many pictures about our experiment. Two of them have been shown in Fig. 9. Fig. 9a is a picture of the micromodel after injecting 6 PVof brine at 0.5 mL/h and Fig. 9b is a picture of the micromodel after injecting 2 PV of the produced biosurfactant at 0.5 mL/h. As this figure shows for rhamnolipid of P. aeruginosa HATH, the oil recovery is more than that observed by water flooding. These pictures indicated that rhamnolipid mobilized oil in the micromodel and should have a significant role in enhanced oil recovery. Therefore, this microorganism produces biosurfactant that can lower oil–water interfacial tension. The drop in interfacial tension increases the capillary number. The increased capillary numbers are associated with reduction of residual oil saturation. There are also some publications that use glass micromodels to study the MEOR process (Soudmand-asali et al., 2007; Amani et al., 2010b). Soudmand-asali et al. (2007) carried out experiments with etched glass micromodels to investigate the MEOR process in fractured porous media. They injected one pore volume of the mixture of bacterial solution and nutrient solution into the micromodel. However, they did not do anything to purify their biosurfactant. Therefore our work tries to fill the gap. The excellent ST and IFT reducing characteristic of biosurfactant produced by P. aeruginosa HATH and its stability over a wide range of temperature, pH and different salt concentrations imply the possibility of using this biosurfactant in enhanced oil recovery.
Al-Bahry, S.N., Al-Wahaibi, Y.M., Elshafie, A.E., Al-Bemani, A.S., Joshi, S.J., AlMakhmari, H.S., Al-Sulaimani, H.S., 2013. Biosurfactant production by Bacillus subtilis B20 using date molasses and its possible application in enhanced oil recovery. Int. Biodeterior. Biodegrad. 81, 141–146. Amani, H., Mehrnia, M.R., Haghighi, M., Sarrafzadeh, M.H., Soudi, M.R., 2010a. Scale up and application of biosurfactant from Bacillus subtilis in enhanced oil recovery. Appl. Biochem. Biotechnol. 162, 510–523. Amani, H., Sarrafzadeh, M.H., Haghighi, M., Mehrnia, M.R., 2010b. Comparative study of biosurfactant producing bacteria in MEOR applications. Pet. Sci. Eng. 75, 209–214. Amani, H., Miller, M.M., Syldatk, C., Hausmann, R., 2013. Production of microbial rhamnolipid by Pseudomonas aeruginosa MM1011 for ex situ enhanced oil recovery. Appl. Biochem. Biotechnol. 170, 1080–1093. Banat, I.M., 1995. Biosurfactants production and possible uses in microbial enhanced oil recovery and oil pollution remediation: a review. Bioresour. Technol. 51, 1–12. Batista, S., Mounteer, A., Amorim, F., Totola, M., 2005. Isolation and characterization of biosurfactant/bioemulsifier producing bacteria from petroleum contaminated sites. Bioresour. Technol. 97, 868–875. Bordoloi, N.K., Konwar, B.K., 2008. Microbial surfactant-enhanced mineral oil recovery under laboratory conditions. Colloids Surf. B 63, 73–82. Bubela, B., 1989. Geobiology and microbiologically enhanced oil recovery. In: Donaldson, E.C., Chilingarian, G.V., Yen, T.F. (Eds.), Microbial Enhanced Oil Recovery. Development in Petroleum Science. Elsevier, Amsterdam, pp. 75–98. Chrzanowski, Ł., Ławniczak, Ł., Czaczyk, K., 2012. Why do microorganisms produce rhamnolipids? World J. Microbiol. Biotechnol. 28, 401–419. Hörmann, B., Müller, M.M., Syldatk, C., Hausmann, R., 2010. Rhamnolipid production by Burkholderia plantarii DSM 9509. Eur. J. Lipid Sci. Technol. 112, 674–680. Grattoni, C.A., Al-Sharji, H.H., Dawe, R.A., Zimmerman, R.W., 2002. Segregated pathways mechanism for oil and water flow through an oil-based gelant. Soc. Pet. Eng. J. 35, 183–190. Healy, R.N., Reed, R.L., Stenmark, D.G., 1976. Multiphase microemulsion systems. Soc. Pet. Eng. J. 16, 147–160. Ismail, W., Al-Rowaihi, I.S., Al-Humam, A.A., Hamza, R.Y., El Nayal, A.M., Bououdina, M., 2012. Characterization of a lipopeptide biosurfactant produced by a crudeoil-emulsifying Bacillus sp. l–15. Int. Biodeterior. Biodegrad. 84, 168–178. Joshi, S., Bharucha, C., Jha, S., Yadav, S., Nerurkar, A., Desai, A.J., 2008. Biosurfactant production using molasses and whey under thermophilic conditions. Bioresour. Technol. 99, 195–199. Ławniczak, Ł., Marecik, R., Chrzanowski, Ł., 2013. Contributions of biosurfactants to natural or induced bioremediation. Appl. Microbiol. Biotechnol. 97, 2327–2339. Lovaglio, R.B., Santos, F.J., Jafelicci, M.J., Contiero, J., 2011. Rhamnolipid emulsifying activity and emulsion stability: pH rules. Colloids Surf. B 85, 301–305. Müller, M.M., Hörmann, B., Syldatk, C., Hausmann, R., 2010. Pseudomonas aeruginosa PAO1 as a model for rhamnolipid production in bioreactor systems. Appl. Microbiol. Biotechnol. 87, 167–174. Pimientar, A.L., Díaz, M.P., Carvajal, F.G., Grosso, J.L., 1997. Production of biosurfactants (rhamnolipids) by Pseudomonas aeruginosa isolated from colombian sludges. Cienc. Tecnol. Futuro 1, 95–108. Raheba, J., Hajipoura, M.J., 2011. The stable rhamnolipid biosurfactant production in genetically engineered pseudomonas strain reduced energy consumption in biodesulfurization. Energy Sources 33, 2113–2121.
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Roshanfekr, M., Johns, R.T., 2011. Prediction of optimum salinity and solubilization ratio for microemulsion phase behavior with live crude at reservoir pressure. Fluid Phase Equilib. 304, 52–60. Sayegh, S.G., Fisher, D.B., 2009. Enhanced oil recovery by CO2 flooding in homogeneous and heterogeneous 2D micromodels. J. Can. Pet. Technol. 48, 30–36. Sen, R., 2008. Biotechnology in petroleum recovery: the microbial EOR. Prog. Energy Combust. Sci. 34, 714–724. Soudmand-asali, A., Ayatollahi, S., Mohabatkar, H., Zareie, M., Shariatpanahi, F., 2007. The in situ microbial enhanced oil recovery in fractured porous media. Pet. Sci. Eng. 58, 61–72. Sreekala, T., Gina, S.S., 1994. Effect of anionic biosurfactant on hexadecane partitioning in multiphase systems. Environ. Sci. Technol. 28 (12), 1993–2000. Stewart, T.L., Kim, D.S., 2004. Modeling of biomass-plug development and propagation in porous media. Biochem. Eng. J. 17, 107–119.
219
Yeh, M.S., Wei, Y.H., Chang, J.S., 2005. Enhanced production of surfactin from Bacillus subtilis by addition of solid carriers. Biotechnol. Prog. 21, 1329–1334. Youssef, N., Simpson, D.R., Duncan, K.E., McInerney, M.J., Flomsbee, M., Fincher, T., Knapp, R.M., 2007. In-situ biosurfactant production by Bacillus strains injected into a limestone petroleum reservoir. Appl. Environ. Microbiol. 73, 1239–1247. Youssef, N., Simpson, D.R., McInerney, M.J., Duncan, K.E., 2013. In-situ lipopeptide biosurfactant production by Bacillus strains correlates with improved oil recovery in two oil wells approaching their economic limit of production. Int. Biodeterior. Biodegrad. 81, 127–132. Wang, Q.H., Fang, X.D., Bai, B.J., Liang, X.L., Shuler, P.J., Goddard, W.A., Tang, Y.C., 2007. Engineering bacteria for production of rhamnolipid as an agent for enhanced oil recovery. Biotechnol. Bioeng. 98, 842–853.
Contents lists available at ScienceDirect
Journal of Petroleum Science and Engineering journal homepage: www.elsevier.com/locate/petrol
Study of enhanced oil recovery by rhamnolipids in a homogeneous 2D micromodel Hossein Amani n Faculty of Chemical Engineering, Babol Noshirvani University of Technology, Babol, Iran
art ic l e i nf o
a b s t r a c t
Article history: Received 17 September 2014 Accepted 18 February 2015 Available online 26 February 2015
In this study, homogeneous glass micromodel is used for investigation of oil recovery by rhamnolipid. The intention of this study was to investigate whether the rhamnolipid mixture could be produced in commercial quantities for enhanced oil recovery (EOR) projects in bioreactor and prove of its potential use as an effective material for field application. In this work, the ability of Pseudomonas aeruginosa HATH to grow and produce rhamnolipid on sunflower as a sole carbon source under nitrogen limitation was shown. The production of Rha-C10-C10 and Rha2-C10-C10 was confirmed by thin layer chromatography (TLC) and high-performance liquid chromatography (HPLC) analysis. The yield of rhamnolipid per biomass (YRL/x), rhamnolipid per sunflower oil (YRL/s), and the biomass per sunflower oil (Yx/s) for bioreactor were obtained about 0.54 g g 1, 0.059 g g 1, and 0.11 g g 1, respectively. The rhamnolipid mixture obtained was able to reduce the surface and interfacial tension of water to 26 and 2 mN/m, respectively. Produced rhamnolipid is an effective surfactant at very low concentrations over a wide range of temperatures, pHs and salt concentrations and also has the ability to emulsify oil, which is essential for enhanced oil recovery. With a critical micelle concentration, 5% of original oil in place was recovered after water flooding from a micromodel. This result suggests rhamnolipids as appropriate model biosurfactants for microbial enhanced oil recovery. & 2015 Elsevier B.V. All rights reserved.
Keywords: bioreactor biosurfactant fermentation microbial enhanced oil recovery micromodel rhamnolipid
1. Introduction Microbial enhanced oil recovery (MEOR) is the use of microbes in petroleum reservoirs to enhance the amount of oil that can be produced (Bordoloi and Konwar, 2008; Sen, 2008; Amani et al., 2010a,b; Al-Bahry et al., 2013; Youssef et al., 2013). The microbes use hydrocarbons as a food for their metabolic processes and excrete natural and non-toxic bio-products such as biosurfactants and biopolymers. Some of the advantages of biosurfactants over conventional surfactants include lower toxicity, high biodegradability, tolerance to at extreme temperatures, pH and salinity and their potential applications in environmental protection and management (Joshi et al., 2008; Amani et al., 2010a,b; Chrzanowski et al., 2012; Ismail et al., 2012; Ławniczak et al., 2013; Youssef et al., 2013). Rhamnolipids are among the best known biosurfactants and have been proven to be very promising in enhanced oil recovery (Banat, 1995; Wang et al., 2007; Sen, 2008; Amani et al., 2010a,b; Hörmann et al., 2010; Müller et al., 2010; Youssef et al., 2013). The rhamnolipids can change the physical and chemical properties of the crude oil and stimulate oil–water–rock interactions that improve oil recovery. They are the most effective biosurfactants with the ability to reduce the water surface tension
n
Tel.: þ 98 1113232071; fax: þ 98 1113220570. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.petrol.2015.02.030 0920-4105/& 2015 Elsevier B.V. All rights reserved.
from 72 to 30 mN/m as well as interfacial tension reduction in water/ oil system from 43 to below 1 mN/m at trace concentrations (Amani et al., 2010a,b; Hörmann et al., 2010). The following positive effects in a reservoir have been documented: (i) biosurfactants reduce the interfacial tensions between oil/rock and oil/water, improving oil flow, (ii) if reservoir is oil wet, biosurfactants increase the wettability of the rock toward water, causing the rock to be preferentially wet by water, dislodging the oil, again helping it flow more freely, and (iii) biosurfactants reduce the interfacial tension between water and oil, and therefore a lower hydrostatic pressure is required to move the liquid entrapped in the pores to overcome the capillary effect. Rhamnolipid always consists of one or two units of rhamnose linked to one or two fatty acid chains with C8–C14 carbon atoms, which may or may not be saturated (Hörmann et al., 2010; Müller et al., 2010). Four different rhamnolipid homologs, produced by Pseudomonas aeruginosa, have been identified and characterized (Hörmann et al., 2010; Müller et al., 2010). The focus of the proposed study is production of rhamnolipids, purification and identification of rhamnolipid, and a visual study of enhanced oil recovery by rhamnolipid flooding in a homogeneous 2D micromodel. In recent years, micromodels have been served as an excellent laboratory instrument to investigate and understand the mechanisms of biosurfactant flooding in removing the residual oil. Micromodels provide the opportunity to observe fluid flow within reservoirs
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
(Stewart and Kim, 2004; Soudmand-asali et al., 2007; Sayegh and Fisher, 2009). Several investigations have been carried out on the in situ bacteria flooding but the ex situ rhamnolipid production and addition to the micromodel as agents for MEOR has little been studied. In order to develop a suitable technology for ex situ MEOR processes, it is essential to carry out tests about ex situ enhanced oil recovery by rhamnolipids. Therefore this work tries to fill the gap. Another intention of this study was to investigate whether the rhamnolipid mixture could be produced in commercial quantities for EOR processes in the bioreactor and to prove its potential use as an effective material for reservoirs in Iran.
2. Materials and methods 2.1. Microorganism The biosurfactant producing strains were isolated from petroleum contaminated garage site using oil spreading method as described by Youssef et al. (2007). The isolation and screening of the biosurfactant producing bacteria were performed in the Tehran University Institute of Biotechnology. Those isolates which showed bigger zone of clearance on oil layer and reduced surface tension (ST) below 30 mN/m were selected for further experiments. One strain (HATH) was identified as P. aeruginosa according to its 16SrRNA sequences by Pasteur Institute (Paris, France). The strain was kept in laboratory of Industrial Microbiology at Tehran University (Tehran, Iran) and nominated as P. aeruginosa HATA. 2.2. Media and culture conditions Lysogeny broth (LB) was used for pre-cultivation step one. For the culture, a nitrogen-limited production medium consisting of 250 g/L sunflower oil (Bellasan, Aldi Süd, Rastatt, Germany) and a Ca-free mineral salt solution with 1.5 g/L NaNO3, 0.05 g/L MgSO4 7H2O, 0.1 g/L KCl, containing a 0.1 M sodium phosphate buffer at pH 6.5 was used. A total of 1 mL/L of trace element solution was added. The trace element solution contained 2.0 g/L sodium citrate 2H2O, 0.28 g/L FeCl3 6H2O, 1.4 g/L ZnSO4 7H2O, 1.2 g/L CoCl2 6H2O, 1.2 g/L CuSO4 5H2O, and 0.8 g/L MnSO4 H2O. Trace elements were filter-sterilized through a 0.22 μm membrane filter (Carl Roth GmbH, Karlsruhe, Germany). Production medium was adjusted to pH 6.5 with HCl (6 N) and NaOH (1 N). Mineral salt solutions, phosphate sources, and sunflower oil were autoclaved separately for all experiments (Müller et al., 2010). All shake flask cultures were incubated in a shake incubator chamber (Multitron II, HT Infors, Bottmingen, Switzerland). First 25 mL of LB in a 100-mL baffled shake flask was inoculated with a total volume of 100 μL from the glycerol stock solution of P. aeruginosa HATH and, incubated for 24 h at 37 1C and 120 rpm. The second pre-cultures also containing 25 mL LB were cultivated in a 100-mL baffled shake flask and inoculated using a total of 0.5 mL from the 24-h LB culture. This culture was incubated for 24 h at 37 1C, 120 rpm. The production was started after inoculation with the second LB pre-culture resulting in OD580 of 0.05 in the aqueous phase at the start of the bioreactor cultivation. 2.3. Bioreactor experiment Fermentation was carried out in an integrated stirred tank bioreactor with accessories and automatic systems for dissolved oxygen (DO), pH, impeller speed, aeration rate, and temperature (Infors AG, Heado fice, Switzerland). The reactor used was a 2.5-L batch stirred bioreactor with three baffle plates provided to reduce vertex effects and enhance the mixing. Two Rushton impellers
213
attached to the same shaft were used for agitation. Air was sparged through a pipe sparger placed below the bottom impeller. A foam collector was connected to the top of the bioreactor to recover the biosurfactant by withdrawing the foam produced during the process. Stirrer speed was set fix at 600 rpm, temperature at 37 1C and the aeration rates at 2 vvm. For the cultivation, 0.55 L of the production medium was used. The trace element solution was added at cultivation times of 0, 20, 40, 70, and 120 h, as described before (Müller et al., 2010). The production was carried out for 9 days. According to the analysis purposes, sampling was undertaken on a daily basis. 2.4. Analysis of sunflower oil, rhamnolipid and biomass Culture suspension was mixed vigorously with n-hexane 1:1 (v/v) and centrifuged (4600 g, 4 1C, 30 min) for separation of cells, aqueous and n-hexane phase. The n-hexane phase was used for gravimetric determination of sunflower oil concentrations, after evaporation of n-hexane. We can only analyze the extractable portion of the sunflower oil and at t¼0 h, it is only the theoretical value of 250 g/L. For rhamnolipd measurement, an aliquot of the aqueous phase was acidified with 85% phosphoric acid 1:100 (v/v) to adjust a pH of about 2–3, leading to precipitation of the rhamnolipids. Rhamnolipids were extracted twice with ethyl acetate 1:1.25 (v/v). A 10 mL of ethyl acetate extracts was used for rhamnolipid identification by TLC. The TLC was performed according to Müller et al. (2010). A rhamnolipid standard was prepared from Jeneil JBR425 (Jeneil Biosurfactants Company, Saukville, United States). Appropriate amounts of ethyl acetate extracts were evaporated and used for rhamnolipid quantification by HPLC (Hörmann et al., 2010; Müller et al., 2010). For biomass measurement, the conventional cell dry weight measurement method was used to determine biomass dry weight. Following the centrifugation of sample and separation of hexane phase and aqueous phase from the biomass, the biomass was washed once in 0.9% NaCl solution (4600 g, 4 1C, 30 min) in distillated water and then it was transferred into a pre-tared vial. The bacterial dry weight was determined after drying at 1051 C for 24 h. A complete working scheme for analytics is shown in Fig. 1. The surface tension (ST), interfacial tension (IFT) and critical micelle concentration measurement (CMC) were measured at 251 C by a digital tensiometer (Kruss, K10ST, Germany) using the ring method. IFT measurements were carried out against crude oil (API¼ 341, acid number ¼0.2 mg KOH/g oil, base number ¼ 0.05 mg KOH/g oil). The CMC was determined by measuring surface tension at different concentrations of rhamnolipid in distillated water up to a constant value of surface tension saturated (Amani et al., 2010a,b; Hörmann et al., 2010). 2.5. Stability studies Rhamnolipid produced by P. aeruginosa HATH was used for stability studies. Samples of rhamnolipid at CMC concentration (5 cc of the produced rhamnolipid was poured into a test tube containing 5 mL of crude oil (water oil ratio¼1)) were checked for stability of surface activity (ST) under different environmental conditions: High temperatures (40, 60, 70, 80, 90 and 100 1C for 60 min and at 120 1C for 20 min), different salt concentrations (NaCl: 0, 1, 2, 4, 6, 8, 10, 15 and 20%, w/v) and a wide range of pH (2, 4, 6, 7, 8, 10 and 12). The pH was adjusted with HCl (6 N) and NaOH (1 N). 2.6. Emulsification and oil spreading studies Emulsification activity of the rhamnolipid solutions at CMC concentration was determined by measuring the emulsion index (E24) at 25 1C. In brief, 4 ml of crude oil was poured separately into
214
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Taking sample from bioreactor and add 1 volume of n-hexane for extraction of oil
4700rpm; 4°C; 30min
Gravimetric analysis of oil in n-hexane phase
Mix 2mL aqueous phase with 20µL H PO
Wash pellet once with 0,9% saline
2,5mL EtAc for RL extraction
10min at 7500rpm
remove supernatant for rhamnolipid analysis
Repeat extraction
HPLC analysis
Thin layer chromatography
Fig. 1. Working scheme for analytics.
a test tube containing 4 ml of biosurfactant solution. A biosurfactant solution consisted of the following composition: 120 mg rhamnolipid (CMC), 0.051 g NaHCO3, 75 g NaCl, 0.61 g KCl, 9.2 g CaCl2 and 7.6 g MgCl2 per liter, close to the specified reservoir brine. After being vigorously mixed for 2 min, the test tube was kept still for 24 h and the heights of emulsion, oil and aqueous zones were measured. The emulsion index (E24) was then calculated from the ratio of the height of the emulsion zone to the total height of the oil, emulsion, and aqueous zones (Yeh et al., 2005). To investigate the stability condition, test tubes were kept still for 15 days and then measurement of the emulsion index was undertaken again. In order to investigate the effects of the biosurfactant produced on the oil and aqueous phases, oil spreading technique was carried out according to Youssef et al. (2013). Briefly, fifty milliliter of distilled water was added to the petri dishes followed by the addition of 100 μl of crude oil to the surface of water. Then 10 μl of the rhamnolipid solutions at CMC concentration was dropped on to the crude oil surface. The diameter of clear zone on the oil surface was measured and compared to 10 μl of distilled water as negative control. 2.7. Determination of optimum salinity The salinity plays an important role in achieving low IFT between both the microemulsion and oil, and the microemulsion and brine. The optimum salinity is defined as the salinity where the IFTs are equal (Healy et al., 1976; Roshanfekr and Johns, 2011). The optimal salinity is alternatively defined as the salinity where the solubilization ratios for brine and oil are equal (Roshanfekr and Johns, 2011). In this research, the water to oil ratio value (WOR¼1)and biosurfactant concentration (CMC) were kept constant to find a possible trend by changing the oil and water solubilization ratios at different salinities. The solubilization ratio is defined as δi ¼ Vi/Vs, where Vi is the volume of oil or water solubilized and Vs is the volume of surfactant in the microemulsion phase. Tests were conducted at WOR¼1 in borosilicate 10 cc pipettes. Each pipette contained 1.2 wt% biosurfactant,
crude oil and brines of X% NaCl. The pipettes were sealed with a torch and samples were rotated end-to-end for 5 days and then left to stand quiescent. The oil and water solubilization ratios were measured after one week (equilibrium was assumed). 2.8. Micromodel studies and experimental setup As the average oil reservoir temperature is about 80 1C, the application of rhamnolipid in enhanced oil recovery was evaluated using micromodel at 80 1C. Glass etched micromodel was constructed for this investigation. The glass micromodel is a two dimensional flow channel network etched in a glass to simulate the fluid flow in porous media. Glass is composed mainly of sand (silicates, SiO2) and silicates which show slight adsorption of anionic surfactants, but adsorb strongly cationic surfactants (Bubela, 1989). In this work, we assume that the glass micromodel has a slight adsorption of biosurfactant in pores because the produced rhamnolipid is an anionic biosurfactant (Sreekala and Gina,1994). Irregular and dead-pore network patterns were drawn according to Soudmand-asali et al. (2007). In our work, one identically patterned micromodel was used. The network pattern of the micromodel has been shown in Fig. 2. The physical and hydraulic properties of the micromodels such as porosity (%), etched thickness (mm), permeability (mDarcy) and pore volume (cm3) were 41.5, 0.057, 100 and 0.335 respectively. Crude oil (API¼ 341) was used as the oleic phase. The crude oil was obtained from the College of Chemical Engineering of University of Tehran, Iran. Glass micromodels are usually considered strongly water-wet because their surface chemistry is similar to that of clean sandstone (Grattoni et al., 2002). In our work, we assume that no wettability alteration occurs in the experiments. To investigate oil displacement, first the model was saturated with 5% (by weight) NaCl solution, and then the crude oil injected at the rate of 4 mL/h until the condition of connate water saturation was reached (the crude oil displaced the brine until no more brine was produced). At the end of imbibition process, brine was injected at a rate of
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
215
0.5 ml/h until no more oil was produced in the effluent. At this point, we injected the rhamnolipid of P. aeruginosa HATH at CMC concentration for a period of 2 days and a flow rate of 0.5 mL/h instead of brine. Difference between oil residual saturation in experiments was equal to the amount of recovery factor due to rhamnolipid injection after water flooding. The experimental setup has been shown in Fig. 3.
3. Results and discussions 3.1. Production of rhamnolipid in bioreactor The cultivation of P. aeruginosa HATH at 37 1C and 600 rpm in the integrated bioreactor was performed three times. The profiles of biomass concentration, rhamnolipid production, and sunflower oil concentration versus time in the mentioned conditions are presented in Fig. 4. Fig. 4 shows that the maximum biomass of about 22 g/L was reached after 84 h of growth, giving a yield of biomass on sunflower oil (Yx/s) of 0.11 g g 1. Approximately biomass, sunflower oil and rhamnolipid concentration were stable after 100 h. After 24 h of cultivation rhamnolipid concentrations of 0.05 g/L were quantified. Rhamnolipid production continued up to 120 h of cultivation and reached a maximum of approximately 12 g/L. Specific growth rate was 0.04 h 1, ascertained from the slope of the plot of Ln (biomass) versus time (from 10 to 40 h).The rhamnolipid production was not growth related, which is in accordance to literature data and a typical behavior of secondary metabolite (Hörmann et al., 2010; Müller et al., 2010). The maximum of the production rate was calculated for between 38 h and 42 h of cultivation (0.19 g L 1 h 1) in the transition towards stationary phase. However, small quantities of rhamnolipid were already produced in the early growth phase. The yield of
Fig. 3. Experimental setup for oil displacement study.
Fig. 4. Time course profiles of P. aeruginosa HATH cell growth, sunflower oil concentration, and rhamnolipid concentration at 600 rpm, 2 vvm, and 37 1C in the integrated bioreactor containing 0.55 l medium.
rhamnolipid per biomass (YRL/x) and the production yield (YRL/s) was 0.54 g g 1 and 0.059 g g 1, respectively. For a better comparison between different researches, one parameter such as Y was calculated and presented in Table 1. 3.2. Characterization of biosurfactant produced by P. aeruginosa HATH
Fig. 2. Porous pattern of the micromodel.
A rhamnolipid standard was prepared from JBR (Jeneil Biosurfactants Co., Saukville, USA). The JBR is a mixture of an undefined number of rhamnolipid congeners of P. aeruginosa mainly containing di- and mono-rhamnolipids with one and two β-hydroxydecanoic acids, respectively (four different rhamnolipid homologs). To confirm the rhamnolipid production, the TLC of commercially available rhamnolipid (JBR425, Jeneil Biosurfactant Company) and rhamnolipid of this study were compared. Consequently, the rhamnolipids of P. aeruginosa appear on the TLC plate by decreasing hydrophobicity: Rha-C10-C10 (R1), Rha-C10 (R2), Rha2-C10-C10 (R3) and Rha2-C10 (R4). The TLC analysis of culture supernatants and their respective organic extracts of the strain displayed two spots for P. aeruginosa HATH which could be stained positively for sugar and fatty acid moieties. An example is shown in Fig. 5. Thin-layer chromatography (TLC) results suggest that the isolated surface-active products from P. aeruginosa HATH contain glycolipids (R1 and R3) which showed a similar retention behavior
216
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Table 1 Comparison of the results obtained for batch bioreactor cultivation of P. aeruginosa strains with comparable studies. P. aeruginosa strain
Carbon source (g/L)
Process
Process time (h)
Maximum rhamnolipid concentration (g/L)
Y (g L 1 h 1)
Ref.
PAO1 HATH UI 29791 DSM 7107 DSM 7108
Sunflower oil (250) Sunflower oil (250) Corn oil (75) Soybean (125) Soybean (163)
Batch Batch Batch Batch Batch
90 120 192 166 267
39 12 46 78 112
0.43 0.19 0.24 0.47 0.42
Müller et al. This study Müller et al. Müller et al. Müller et al.
(30 L) (2.5 L) (14 L) (30 m3) (300 L)
RL 1
RL 2
RL 3
RL 4
Jeneil standard
produced rhamnolipid
Fig. 5. TLC of surface-active product from P. aeruginosa HATH and from commercially available rhamnolipid (JBR425, Jeneil Biosurfactant Company).
to the commercially available Jeneil rhamnolipid mixture. The identity of the two glycolipids was further confirmed by HPLC analysis. To confirm the rhamnolipid production, the HPLC chromatogram of pure rhamnolipid standard solutions of Rha-C10-C10 and Rha2-C10-C10 and rhamnolipid of this study were compared. With this method, the production of rhamnolipids with similar retention times as Rha-C10-C10 and Rha2-C10-C10 by P. aeruginosa HATH could be proved. Smaller peaks apart from these could either be related to other congener types or also occur due to fatty acids or mono/diglycerides, which are generated by the cleavage of the carbon source. These substances are also present in the organic extracts and they are probably also derivatised by 4-bromo phenacyl bromide, due to their acid function. However, significant HPLC peaks were detected for our products. Peaks with a similar retention time to the di-rhamnolipid and mono-rhamnolipid of a commercially available rhamnolipid were found. Therefore we think our strain has produced a glycolipid or rhamnolipid. 3.3. Effect of produced rhamnolipid on surface tension, emulsification activity and oil spreading The production of biosurfactant by P. aeruginosa HATH showed excellent reduction of water surface tension from 72 to 25 mN/m. Our results also showed that the minimum value of interfacial tension (IFT) for produced rhamnolipid was 2 mN/m. The reduction of IFT between oil and water is very important in oil recovery because capillary number is defined as the ratio of viscous to capillary forces.
(2010) (2010) (2010) (2010)
Capillary number increases with decrease in interfacial forces. Increases in capillary number lower the residual oil saturation in the core and increase residual oil recovery (Sen, 2008). For further investigation we determined the CMC of the obtained rhamnolipid mixture. The CMC of the rhamnolipid was 120 mg/L. Emulsification activity of the produced rhamnolipid at CMC concentration was determined by measuring the emulsion index (E24). A maximum emulsion index of 85% was reached for the crude oil. The potential of the biosurfactant oil dispersion was also evaluated using the rhamnolipid produced by the P. aeruginosa HATH. The rhamnolipid at CMC concentration showed high dispersability of the crude oil. Fig. 6 shows that if 10 μl of the produced rhamnolipid is gently placed on the center of the oil layer, the crude oil is displaced and a clearing zone is formed. The diameter of this clearing zone on the oil surface correlates to surfactant activity, also called oil displacement activity. Diameter variations during one week are also shown in Fig. 6. As shown in the figure, the crude oil emulsion remained stable even after one week. However, the results showed that the produced rhamnolipid is capable of reducing the surface and interfacial tensions of water to significant lower value, and having excellent emulsifier and dispersion properties. These properties suggest that biosurfactants isolated from P. aeruginosa HATH are appropriate candidates for enhanced oil recovery. 3.4. Study of stability To check the stability of biosurfactant, samples of rhamnolipid (120 mg/L) were subjected to different conditions. At pH 2.0–4.0, we observed much higher ST; since the biosurfactant is not soluble at such acidic conditions, it tends to precipitate. The effect of pH on biosurfactant activity showed that when pH of biosurfactant decreased, the surface tension increased. It could be the result of biosurfactant precipitation in acidic condition. Thus precipitated and structurally distorted biosurfactant loses its capability of reducing surface tension. However, the effects of pH on surface tension showed that the rhamnolipid has an effective surface activity from pH ¼4 to pH ¼10.0. Several reports confirm the stability of biosurfactant at different pH values mostly in the alkaline medium (Pimientar et al. 1997; Batista et al., 2005; Amani et al., 2010b; Lovaglio et al., 2011; Raheba and Hajipoura, 2011, Amani et al., 2013). We also observed similar stability trend in our biosurfactant. Studies on the effect of heat treatment showed that no change in biosurfactant surface activities has occurred and there is no difference in surface tension before and after this treatment in all cases. The CMC value remained stable after exposure to high temperature (80–100 1C) even after 1 day. When placed to autoclave (120 1C for 20 min) the surface activity was also maintained. So it could be interpreted as a good heat resistance of rhamnolipid produced by P. aeruginosa HATH. Several studies confirm the stability of biosurfactants under extreme conditions of temperature. Amani et al.
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
217
Fig. 6. Oil spreading of produced rhamnolipid (CMC concentration) in a glass culture plate (90 mm diameter) which contained 50 ml water in the bottom and 100 μl of crude oil (API ¼ 34) at the surface. Dispersants' activities to the crude oil observed.
surface tension was measured. The results showed that the values of the ST resisted well against salt concentration up to 25 g/L. We know oil reservoirs are one of the harsh environments, where temperature can range from 20 to 90 1C, normal salinity to hypersalinity and pH over a wide range. So, these findings suggest that the rhamnolipid produced by the P. aeruginosa HATH is a superior surfactant for application under the extreme conditions which usually prevail in enhanced oil recovery. In other words the produced rhamnolipid meets those criteria and showed stability over a wide range of environmental factors, while retaining its surface activity. 3.5. Interfacial tension and solubilization parameters
Fig. 7. Solubilization parameter versus salinity (WOR ¼ 1, T ¼ 25 1C).
Fig. 7 represents the corresponding behavior of the solubilization parameters with different salinities. From Fig. 7, the oil solubilization ratio increases and water solubilization ratio decreases as salinity increases. The results show that the measured solubilization ratio at the optimum salinity for the crude oil is 10.5, while the optimum salinity is 2.4 wt%. The optimum salinity occurs where these two curves intersect at 2.4 wt% (24,000 ppm). At this condition (optimum salinity), the IFT between the microemulsion and the excess oil phase, and the IFT between the microemulsion and water phase were measured. Our results showed that IFTs were the same and about 0.25 mN/m. As discussed in Section 3.3, if interfacial tension is small, capillary number will be large enough to let the residual oil saturation go to zero. This is one of the main mechanisms for enhanced oil recovery with biosurfactants. However, the phase behavior of microemulsions is very important to enhanced oil recovery because it can be used as an indicator of ultra-low interfacial tension. 3.6. Study of oil recovery after waterflooding
Fig. 8. Relationship between water and rhamnolipid flooding with pore volume in micromodel.
(2013) reported that biosurfactant produced by P. aeruginosa MM1011 was stable at 120 1C. Bordoloi and Konwar (2008) exposed the biosurfactant produced by P. aeruginosa strains to a temperature of 100 1C for different time periods of 5–60 min and found that the biosurfactant activity remained unaffected with respect to ST changes. For salinity effect, various amounts of NaCl were added to rhamnolipid at CMC concentration and mixed completely and then
To check the effect of biosurfactant on enhanced oil recovery, micromodel experiments were carried out. The initial oil saturation (Soi) of the micromodel was calculated to be 71% after oil flooding. A 120 mg/L solution of the rhamnolipid (CMC concentration) was prepared and used to enhance oil recovery in micromodel. In our work, first a brine solution was introduced at 0.5 mL/h into the micromodel so that this aqueous phase now occupies all of the pore space. Then, crude oil (4 mL/h) displaced the brine in the micromodel until no more brine was produced. The brine was injected again at 0.5 mL/h into the micromodel until no more oil came out. So at this point, the micromodel with trapped oil was ready for flooding with the produced rhamnolipid. Finally, the rhamnolipid was injected at 0.5 mL/h. The oil recovery was recorded during the brine and the rhamnolipid injection steps. Fig. 8 shows relationship between water and rhamnolipid flooding with pore volume injected in micromodel. According to Fig. 8, it was found that after injecting
218
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
4. Conclusions The intention of our research was to investigate whether our rhamnolipid has high enough quantities of enhanced oil recovery in the laboratory scale to be used as an effective material for field application. In this work, rhamnolipid production and growth characteristics of P. aeruginosa HATH using sunflower oil as substrates were studied. Maximal rhamnolipid concentration of 12 g/L was attained after 84 h of cultivation and, at this point, most of the sunflower oil was utilized. The production of rhamnolipid was confirmed by TLC and HPLC analysis. The excellent ST and IFT reducing characteristic of biosurfactant produced by P. aeruginosa HATH and its stability over a wide range of temperature, pH and different salt concentrations imply the possibility of using this biosurfactant in enhanced oil recovery. The produced biosurfactant is capable of having excellent emulsifier and dispersion properties. Finally, the ability in recovering the oil from oilsaturated glass was also demonstrated. Nearly 5% of original oil in place was recovered using the rhamnolipid at CMC concentration after water flooding in micromodel. This work was demonstration of successfulness of use of rhamnolipids for industrial applications, especially in MEOR.
References
Fig. 9. Photographs of micromodel when it is completely saturated by oil. (a) Micromodel after water flooding and (b) micromodel after rhamnolipid flooding at CMC concentration.
6 PVof brine at 0.5 mL/h, no more oil was produced. Extra oil recovery occurred after injecting 2 PV of the produced biosurfactant at 0.5 mL/h. As shown in the figure, the recovery value of oil after water flooding was obtained as 30% of original oil in place, whereas nearly 5% of original oil in place was recovered using rhamnolipid after water flooding. The experiment was performed three times. Our oil displacement experiment demonstrates that a very low concentration (120 mg/L) of the rhamnolipid can recover a significant fraction of trapped oil from a micromodel. There are many pictures about our experiment. Two of them have been shown in Fig. 9. Fig. 9a is a picture of the micromodel after injecting 6 PVof brine at 0.5 mL/h and Fig. 9b is a picture of the micromodel after injecting 2 PV of the produced biosurfactant at 0.5 mL/h. As this figure shows for rhamnolipid of P. aeruginosa HATH, the oil recovery is more than that observed by water flooding. These pictures indicated that rhamnolipid mobilized oil in the micromodel and should have a significant role in enhanced oil recovery. Therefore, this microorganism produces biosurfactant that can lower oil–water interfacial tension. The drop in interfacial tension increases the capillary number. The increased capillary numbers are associated with reduction of residual oil saturation. There are also some publications that use glass micromodels to study the MEOR process (Soudmand-asali et al., 2007; Amani et al., 2010b). Soudmand-asali et al. (2007) carried out experiments with etched glass micromodels to investigate the MEOR process in fractured porous media. They injected one pore volume of the mixture of bacterial solution and nutrient solution into the micromodel. However, they did not do anything to purify their biosurfactant. Therefore our work tries to fill the gap. The excellent ST and IFT reducing characteristic of biosurfactant produced by P. aeruginosa HATH and its stability over a wide range of temperature, pH and different salt concentrations imply the possibility of using this biosurfactant in enhanced oil recovery.
Al-Bahry, S.N., Al-Wahaibi, Y.M., Elshafie, A.E., Al-Bemani, A.S., Joshi, S.J., AlMakhmari, H.S., Al-Sulaimani, H.S., 2013. Biosurfactant production by Bacillus subtilis B20 using date molasses and its possible application in enhanced oil recovery. Int. Biodeterior. Biodegrad. 81, 141–146. Amani, H., Mehrnia, M.R., Haghighi, M., Sarrafzadeh, M.H., Soudi, M.R., 2010a. Scale up and application of biosurfactant from Bacillus subtilis in enhanced oil recovery. Appl. Biochem. Biotechnol. 162, 510–523. Amani, H., Sarrafzadeh, M.H., Haghighi, M., Mehrnia, M.R., 2010b. Comparative study of biosurfactant producing bacteria in MEOR applications. Pet. Sci. Eng. 75, 209–214. Amani, H., Miller, M.M., Syldatk, C., Hausmann, R., 2013. Production of microbial rhamnolipid by Pseudomonas aeruginosa MM1011 for ex situ enhanced oil recovery. Appl. Biochem. Biotechnol. 170, 1080–1093. Banat, I.M., 1995. Biosurfactants production and possible uses in microbial enhanced oil recovery and oil pollution remediation: a review. Bioresour. Technol. 51, 1–12. Batista, S., Mounteer, A., Amorim, F., Totola, M., 2005. Isolation and characterization of biosurfactant/bioemulsifier producing bacteria from petroleum contaminated sites. Bioresour. Technol. 97, 868–875. Bordoloi, N.K., Konwar, B.K., 2008. Microbial surfactant-enhanced mineral oil recovery under laboratory conditions. Colloids Surf. B 63, 73–82. Bubela, B., 1989. Geobiology and microbiologically enhanced oil recovery. In: Donaldson, E.C., Chilingarian, G.V., Yen, T.F. (Eds.), Microbial Enhanced Oil Recovery. Development in Petroleum Science. Elsevier, Amsterdam, pp. 75–98. Chrzanowski, Ł., Ławniczak, Ł., Czaczyk, K., 2012. Why do microorganisms produce rhamnolipids? World J. Microbiol. Biotechnol. 28, 401–419. Hörmann, B., Müller, M.M., Syldatk, C., Hausmann, R., 2010. Rhamnolipid production by Burkholderia plantarii DSM 9509. Eur. J. Lipid Sci. Technol. 112, 674–680. Grattoni, C.A., Al-Sharji, H.H., Dawe, R.A., Zimmerman, R.W., 2002. Segregated pathways mechanism for oil and water flow through an oil-based gelant. Soc. Pet. Eng. J. 35, 183–190. Healy, R.N., Reed, R.L., Stenmark, D.G., 1976. Multiphase microemulsion systems. Soc. Pet. Eng. J. 16, 147–160. Ismail, W., Al-Rowaihi, I.S., Al-Humam, A.A., Hamza, R.Y., El Nayal, A.M., Bououdina, M., 2012. Characterization of a lipopeptide biosurfactant produced by a crudeoil-emulsifying Bacillus sp. l–15. Int. Biodeterior. Biodegrad. 84, 168–178. Joshi, S., Bharucha, C., Jha, S., Yadav, S., Nerurkar, A., Desai, A.J., 2008. Biosurfactant production using molasses and whey under thermophilic conditions. Bioresour. Technol. 99, 195–199. Ławniczak, Ł., Marecik, R., Chrzanowski, Ł., 2013. Contributions of biosurfactants to natural or induced bioremediation. Appl. Microbiol. Biotechnol. 97, 2327–2339. Lovaglio, R.B., Santos, F.J., Jafelicci, M.J., Contiero, J., 2011. Rhamnolipid emulsifying activity and emulsion stability: pH rules. Colloids Surf. B 85, 301–305. Müller, M.M., Hörmann, B., Syldatk, C., Hausmann, R., 2010. Pseudomonas aeruginosa PAO1 as a model for rhamnolipid production in bioreactor systems. Appl. Microbiol. Biotechnol. 87, 167–174. Pimientar, A.L., Díaz, M.P., Carvajal, F.G., Grosso, J.L., 1997. Production of biosurfactants (rhamnolipids) by Pseudomonas aeruginosa isolated from colombian sludges. Cienc. Tecnol. Futuro 1, 95–108. Raheba, J., Hajipoura, M.J., 2011. The stable rhamnolipid biosurfactant production in genetically engineered pseudomonas strain reduced energy consumption in biodesulfurization. Energy Sources 33, 2113–2121.
H. Amani / Journal of Petroleum Science and Engineering 128 (2015) 212–219
Roshanfekr, M., Johns, R.T., 2011. Prediction of optimum salinity and solubilization ratio for microemulsion phase behavior with live crude at reservoir pressure. Fluid Phase Equilib. 304, 52–60. Sayegh, S.G., Fisher, D.B., 2009. Enhanced oil recovery by CO2 flooding in homogeneous and heterogeneous 2D micromodels. J. Can. Pet. Technol. 48, 30–36. Sen, R., 2008. Biotechnology in petroleum recovery: the microbial EOR. Prog. Energy Combust. Sci. 34, 714–724. Soudmand-asali, A., Ayatollahi, S., Mohabatkar, H., Zareie, M., Shariatpanahi, F., 2007. The in situ microbial enhanced oil recovery in fractured porous media. Pet. Sci. Eng. 58, 61–72. Sreekala, T., Gina, S.S., 1994. Effect of anionic biosurfactant on hexadecane partitioning in multiphase systems. Environ. Sci. Technol. 28 (12), 1993–2000. Stewart, T.L., Kim, D.S., 2004. Modeling of biomass-plug development and propagation in porous media. Biochem. Eng. J. 17, 107–119.
219
Yeh, M.S., Wei, Y.H., Chang, J.S., 2005. Enhanced production of surfactin from Bacillus subtilis by addition of solid carriers. Biotechnol. Prog. 21, 1329–1334. Youssef, N., Simpson, D.R., Duncan, K.E., McInerney, M.J., Flomsbee, M., Fincher, T., Knapp, R.M., 2007. In-situ biosurfactant production by Bacillus strains injected into a limestone petroleum reservoir. Appl. Environ. Microbiol. 73, 1239–1247. Youssef, N., Simpson, D.R., McInerney, M.J., Duncan, K.E., 2013. In-situ lipopeptide biosurfactant production by Bacillus strains correlates with improved oil recovery in two oil wells approaching their economic limit of production. Int. Biodeterior. Biodegrad. 81, 127–132. Wang, Q.H., Fang, X.D., Bai, B.J., Liang, X.L., Shuler, P.J., Goddard, W.A., Tang, Y.C., 2007. Engineering bacteria for production of rhamnolipid as an agent for enhanced oil recovery. Biotechnol. Bioeng. 98, 842–853.