Recent updates in our understanding of the microbial metabolisms and their biochemical pathways lead to different applications in petroleum industries like enhancing oil recovery, biodegradation of oil-based waste, treatment of oil-field produced water, bioremediation of oil-spills, etc. At present the worldwide crude oil market is quite volatile with fluctuations in prices. Still, there is an urgent need for new, efficient and superior oil recovery technologies, known as enhanced oil recovery (EOR) techniques. Amongst several types of EOR technologies, microbial enhanced oil recovery (MEOR) is considered to be quite promising one, environmental friendly green-technology (Belyaev et al., 2004; Sen, 2008; Al-Sulaimani et al., 2011b; Zahner et al., 2011). Chemical surfactants and polymers are amongst highly utilized compounds in petroleum industries during EOR applications, for reducing surface tension (ST), interfacial tension (IFT), for generating oil-in water OR water-in-oil emulsions, heavy oil recovery by acting as a viscosifying agent, etc. However, recently the trend has been shifted toward applications of green-compounds or bio-products, mainly because of the environmental concerns. Biosurfactants are such biologically produced surfactants by microorganisms, plants, and animals. Biosurfactants are reported for reducing ST and IFT between oil/water/rock interface, altering the rock wettability, initiating oil/water emulsions, or changing the hydrophilic–lipophilic balance (HLB), thus enhancing the oil recovery (Desai and Banat, 1997; Sen, 2008; Banat et al., 2010). Biosurfactants have been the focus of extensive research because of several advantages over chemical surfactants: considerably lower toxicity, better biodegradability and synthesis from cheaper-agro-industrial waste or renewable raw materials (Marchant and Banat, 2012a,b). Several companies around the globe (UK, USA, Japan, Germany, China) are producing different types of biosurfactant (such as rhamnolipids, lipopeptides, sophorolipids, etc.), with focus on different applications (such as pharmaceuticals, cosmetics, antimicrobials and anticancer, bioremediation, EOR, etc.) (Sekon and Rahman, 2014). Even with all these new developments and increased interest by prominent business-houses, the economically availability and wide-spread applications of biosurfactants are still limited in petroleum industries. The major reason is the availability of comparatively cheaper chemical surfactants and higher cost of biosurfactant production. Several researchers have reported different ways to tackle higher production costs by applying statistical optimization methods or by use of cheaper agro-industrial waste products (Sen and Swaminathan, 2004; Joshi et al., 2007, 2008a; Makkar et al., 2011; Al-Bahry et al., 2013; Saimmai et al., 2013; Chooklin et al., 2014). Different types of biosurfactants are reported, amongst which lipopeptide type of biosurfactant produced by Bacillus group are widely studied. Surfactins and lichenysins produced by Bacillus subtilis and Bacillus licheniformis strains are reported for their high surface activities and other beneficial properties suitable for several applications like enhancing oil recovery (Jean-Marc et al., 2003).

Sultanate of Oman is oil producing Middle Eastern country, where various EOR technologies are employed to enhance the crude oil recovery from the declining reservoirs in the sultanate. To check the potential of biosurfactant based MEOR applications in Oman, we isolated and screened several spore forming bacteria and analyzed for biosurfactant production (Al-Sulaimani et al., 2011a). Amongst several isolates, B. licheniformis W16 showed biosurfactant production and was further selected. The biosurfactant production by this isolate was studied in nine different reported minimal media, and five different carbohydrates were tested to find suitable carbon source in selected better media. The biosurfactant was studied further for wettability alteration (contact angle determination), critical micelle dilution (CMD), extracted and characterized, stability under harsh conditions and MEOR studies by core-flood using Berea sandstone core-plugs under reservoir conditions.

Microbial enhanced oil recovery is regarded as one among the most promising green-technologies that can be potentially implemented with an exceptionally low operating cost. Several of the bacterial products are reported for playing a role in MEOR, where biopolymers and biosurfactants are reported as key players (Belyaev et al., 2004; Sen, 2008; Marchant and Banat, 2012a). Amongst different kind of biosurfactants, low molecular compounds – lipopeptides produced by spore-forming bacteria are regarded as having potential role in MEOR. Several morphologically different spore-forming bacteria were isolated from different oil contaminated soil samples collected from Oman, and screened for biosurfactant production. Out of those isolates, a facultative aerobic bacteria B. licheniformis W16 produced biosurfactant (Al-Sulaimani et al., 2011a) and propagated under anaerobic conditions in the fractured carbonate rocks and enhanced the oil recovery by selective plugging (Al-Hattali et al., 2013). Thus, it was further selected in current study to investigate the effect of different carbohydrates in the minimal media and the potential of produced biosurfactant in biosurfactant based MEOR. The growth and production profile in the nine different production media are shown in Table 1. The better growth was observed in media 2, 5, and 8 within 24 h, whereas very little growth was observed in media 4. One possible reason for little growth in media 4 could be because of the acidic pH after 24 h, as compared to observed neutral to alkaline pH in rest of the media (Table 1), and this strain showed better growth in neutral to alkaline range (data not shown). The lowest ST and IFT values were observed in media 7 and 8, as 25.89–38.88 mN m⁻¹ and 4.48–18.88 mN m⁻¹ respectively. Therefore, media 7 (M7) and 8 (M8) were further selected for carbon source optimization studies. None of the media showed much increase in the viscosities even after 72 h, and the viscosities varied between 1.3 and 1.5 mPa s, for both abiotic controls and experimental samples. The viscosities of some of the biopolymers are reported to be in the range of 43–21535 cP (Joshi et al., 2016). Hence, it was concluded that this isolate B. licheniformis W16 produced only biosurfactant and not biopolymer, under tested environmental conditions in these nine minimal media.

As media M7 and M8 showed better growth and biosurfactant production, both media were selected for further screening of different types of carbohydrates as a carbon sources – glucose, sucrose, starch, cane moleasses, or date molasses. Better growth was observed in M8 media as compared to M7, in all five carbon sources, and highest growth (OD₆₆₀ – 2.3–2.4) was observed in M8 media containing cane or date molasses (Figure 1A). It was observed that in all media ST was reduced to <35 mN m⁻¹, from 62 to 70 mN m⁻¹, within 72 h (Figure 1B), except in starch containing Media M7, where higher ST was observed (45.42 ± 0.88 mN m⁻¹). Similar trend was observed for IFT also, where starch containing media M7 had higher IFT (10.27 ± 0.52) as compared to all other conditions (Figure 1C). The lower ST and IFT were observed in M8 media containing glucose, cane molasses or date molasses, and glucose containing media showed the lowest ST and IFT values of 24.33 ± 0.31 mN m⁻¹ and 2.47 ± 0.32 mN m⁻¹ respectively. These observed values of ST and IFT are in accordance with several other reports by researchers for biosurfactant production by Bacillus species (Joshi et al., 2008a; Al-Bahry et al., 2013; Chandankere et al., 2013; Pereira et al., 2013). In this study, it was observed that isolate B. licheniformis W16 produced a potent biosurfactant using either glucose or cane molasses as a carbon source. Glucose based media M8 was used for biosurfactant production and further studies, as it gave better results in ST and IFT reduction.

Any alteration at the reservoir ‘oil-water-rock’ interface, leads to change in the surface-wettability properties (oil-wet to water-wet and vice versa), which has also been proposed as one of the mechanism responsible for EOR (Kowalewski et al., 2006; Al-Sulaimani et al., 2012; Al-Wahaibi et al., 2014). The B. licheniformis W16 biosurfactant produced from glucose containing M8 medium, was studied for any effect on changes in the contact angle on a hydrophobic surface (provided with the instrument – DSA 100, KRÜSS, Germany). The contact angle was reduced from 55.67 ± 1.6° of un-inoculated abiotic control media to 19.54 ± 0.96° in cell-free biosurfactant broth (Figure 2A). Karimi et al. (2012) studied the effect of microbial solutions (using an Enterobacter cloacae strain) on 7–21 days aged glass surfaces, and have reported that it alters the wettability of hydrophobic glass surfaces toward more water-wet conditions. Al-Sulaimani et al. (2012) reported that biosurfactant produced by B. subtilis W19 changed the contact angle of distilled water from 70.6 ± 0.3° to 25.32 ± 0.06° at 0.25% (w/v) biosurfactant. Al-Wahaibi et al. (2014) also reported changes in wettability of hydrophobic surface from 58.7 ± 0.85° to 28.4 ± 1.03° and 27.2 ± 0.72° by biosurfactant produced by B. subtilis B30 in glucose or molasses based minimal media. In the current study, we observed that biosurfactant changed the wettability of hydrophobic surface toward more water-wet, which is beneficial during EOR applications. To the best of our knowledge this is the first report of wettability alteration using biosurfactant produced by B. licheniformis strain. Along with reduction in ST and IFT, wettability alteration by W16 biosurfactant could also play an important role in improving oil recovery at field scale applications. It is reported that the dilution at which the ST/IFT begins to increase is termed the CMD, which is actually the factor by which the effective biosurfactant concentration exceeds the critical micelle concentration (Ghurye et al., 1994; Al-Wahaibi et al., 2014). The ST and IFT values increased sharply after 4X dilution, where the ST and IFT values observed were 38.28 ± 1.36 mN m⁻¹ and 9.61 ± 0.65 mN m⁻¹ respectively (Figure 2B). Therefore, the observed CMD values for W16 biosurfactant was 4X. Makkar and Cameotra (1997) have reported the biosurfactant production by two B. subtilis strains which reduced the ST in the range of 30–40 dynes cm⁻¹ with 10 X–100 X CMD values after 96 h. Joshi and Desai (2013) reported CMD values of 75X–100X for biosurfactants mixtures produced by different Bacilli strains in carbohydrate based minimal media. Al-Wahaibi et al. (2014) reported CMD values of 8X for biosurfactant produced by B. subtilis B30. The CMD values observed for W16 biosurfactant was comparatively lower than biosurfactants produced by other Bacilli strains.

To determine the yield of biosurfactant, it was extracted by acid-precipitation and crude-powder was collected by spray-drying. The yield of partially purified biosurfactant was 0.52 ± 0.1 g l⁻¹, which was used for structural identification by FTIR. The IR spectra of biosurfactants produced by B. licheniformis W16 is shown in Figure 3. The IR spectrum in KBr showed bands characteristic of peptides at 3300–3400 cm⁻¹ (N-H stretching mode) and at 1650–1700 cm⁻¹ (stretching mode of the CO–N bond). The bands at 1200–1400 cm⁻¹ reflected the aliphatic chains (–CH₃, –CH₂–) of the isolated fraction (Al-Sulaimani et al., 2011a; Al-Wahaibi et al., 2014; Joshi et al., 2015). These results imply the presence of aliphatic groups as well as a peptide-like moiety in the biosurfactant. For comparison of the biosurfactant, surfactin (Sigma chemicals, USA) was also analyzed by FTIR (Figure 3). The IR spectrum of surfactin and W16 biosurfactant was quite similar, which suggested the similarity in the structure of the biosurfactant produced by B. licheniformis W16 with lipopeptide surfactin. Structurally, similar lipopeptide biosurfactants – surfactins and lichenysins are reported as excellent surface active agents with potential in EOR applications (Jean-Marc et al., 2003).

Out of four solvent systems used for HPTLC, SS4 gave better separation (Figure 4A) as compared to SS1, SS2, and SS3; hence, we used solvent system – SS4 for further characterization of biosurfactant. The TLC or HPTLC are reported as quite useful tool for initial qualitative or quantitative analysis of different types of biosurfactants (Mukherjee et al., 2009b; Ismail et al., 2013; Al-Wahaibi et al., 2014). We further utilized TLC–MS interface for extraction and elution of separated biosurfactant bands and analyzed them directly by ESI – MS under positive and negative modes. Total five major bands were scrapped, eluted and analyzed by ESI–MS. The mass spectra (Figure 4B) revealed the major group of peaks at m/z values between 1000 and 1090 (The mass spectra of remaining bands are as Supplementary Figure S1). This group could be attributed to the different variants of surfactins or lichenysins, as previously described (Pereira et al., 2013; Al-Wahaibi et al., 2014; Joshi et al., 2015). The HPTLC–ESI–MS is quite easy and quick technique to identify the biosurfactants. To further confirm the identity of different biosurfactant isoforms, biosurfactant was analyzed using MALDI-TOF-MS.

MALDI-TOF mass spectrum of the biosurfactant produced by B. licheniformis W16 is shown in Figure 5. The observed main peaks were similar to the molecular masses of known lipopeptides – lichensyin and their homologs (Horowitz and Griffin, 1991; Mikkola et al., 2000; Li et al., 2008; Nerurkar, 2010; Madslien et al., 2013; Pereira et al., 2013; Zhang et al., 2014). There were a total of 12 main peaks with the m/z value as 1015.5–1087.5. Considering the molecular mass of different homologs of lichenysin (C₁₂–C₁₆), these peaks were identified as protonated ions [M + H] ⁺ m/z 1049.1; sodium adduct ions [M + Na] ⁺ with m/z of 1015.5, 1029.5, 1043.5, and 1057.6; sodium adducts [M-H + 2Na]⁺ with m/z of 1051.5, 1065.5, and 1079.5; potassium adducts [M + K] ⁺ with m/z of 1087.5. MALDI-TOF analysis of W16 biosurfactant showed similarity with lichenysin-A, produced by B. licheniformis strains (Grangemard et al., 1999; Mikkola et al., 2000; Li et al., 2008; Zhang et al., 2014; Joshi et al., 2015). They have reported that lichenysin – A is a cyclic heptalipopeptide having a small peptide (Gln, Leu, Leu, Val, Asp, Leu, and Ile) linked to 3-hydroxy fatty acid residue with amide (Gln) and lactone (Ile) bonds forming a cyclic structure, with main fatty acids are 3-hydroxylated tri, tetra, penta, and hexadecanoic acids.

The results obtained with ¹H NMR (Figure 6A) further confirmed the lipopeptide nature of the molecule. Seven amide (-NH-) groups were observed in the region from 8.4 to 7.4 ppm, and the alpha-hydrogen (Hαs) of the amino acids showed resonance from 4.7 to 3.8 ppm. A doublet at δ = 0.904 ppm was observed, which indicated a terminal branching in the fatty acyl chain [-(CH₃)₂-CH-]. Other multiplets in the upfield region arise as a result of the side chain protons of the amino acids, and remaining spectra confirmed the presence of β-hydroxy fatty acid. The ¹³C-NMR spectrum showed strong signals at 10.744–17.527, 18.638–59.817, and 171.450–176.880 ppm from methyl, methylene, and carboxyl group, respectively (Figure 6B). Therefore, the biosurfactant produced by B. licheniformis W16 was characterized as lipopeptide, similar to lichenysin-A (Horowitz and Griffin, 1991; Lin et al., 1994; Makkar and Cameotra, 1999; Saimmai et al., 2013; Zhang et al., 2014; Joshi et al., 2015).

Any chemical agents to be used for EOR needs to be stable within a range of high temperatures (>40°C), wide range of pH and salinities. Therefore, the biosurfactant produced by B. licheniformis W16 was studied for its stability at wide range of harsh environmental conditions normally encountered in oil reservoirs. The biosurfactant was quite stable over a wide range of temperatures from 40 to 100°C (Figure 7A), and also was stable at temperatures higher than 100°C (121°C when subjected to autoclaving conditions; and at 100–160°C during drying-extraction using spray-dryer). The W16 biosurfactant was also stable in pH range of 6–12 and salt concentration up to 4% NaCl (Figures 7B, C). Under highly acidic pH (pH ≤ 4.0) biosurfactant was precipitated and thus higher ST and IFT values were observed. Several researchers also reported the stability of biosurfactants under high temperatures, at different pH values and salinities (Gudina et al., 2010; Khopade et al., 2012; Vaz et al., 2012; Saimmai et al., 2013; Al-Wahaibi et al., 2014; Jha et al., 2016).

Several mechanisms are reported for MEOR such as – selective plugging of highly porous rocks, reducing the viscosity of heavy oil by biodegradation, reduction of ST and IFT, wettability alteration etc. Previously, Al-Hattali et al. (2013) reported the potential of B. licheniformis W16 in enhancing the oil recovery in fractured carbonate rocks by growth under anaerobic conditions and selectively plugging the highly porous rock after 11 h incubation under reservoir conditions. They reported 27–30% additional oil recovery over residual oil saturation by microbial permeability profile modification (MPPM). In this study, the ex situ produced biosurfactant by B. licheniformis W16 was used for core-flooding studies using Berea sandstone core-plugs under reservoir conditions.

The salinity of the formation water was between 7 and 9%, pH ∼ 6.9, TDS ∼ 71000–72000 ppm, and major ions were (kg m⁻³): Sodium, 20.2–20.7; Calcium, 3.96–4.52; Magnesium, 0.95–1.02; Iron, 5; Chloride, 41.8–43.2; Sulfate, 0.33–0.35, Bromide, 0.30–0.31 (concentration of other ions and complete water chemistry can be found in Supplementary Table S1).

For core-flood experiments, cell-free biosurfactant broth was used. The pore volumes of the core-plugs were 17–19 cm³. Initial oil saturation (So_i) in cores after oil flooding was around 73–77%, and after 7 PV of brine injection, nearly 80–85% (≅11.5 ml) of oil was produced and residual oil saturation (So_r) was about 15–20%. Extra oil recovery was observed after injecting 4–5 PV of biosurfactant solution, where 24–26% (≅0.5 ml) of So_r was produced (Figure 8). Other researchers have reported 20–37% additional oil recovery over the residual oil saturation using cell-free biosurfactant injection in the core-plugs or sand-pack columns (Yakimov et al., 1997; Almeida et al., 2004; Al-Sulaimani et al., 2011a; Darvishi et al., 2011; Joshi and Desai, 2013; Al-Wahaibi et al., 2014; Arora et al., 2014; Elshafie et al., 2015; Joshi et al., 2015; Jha et al., 2016). In present study, the additional oil recovered could be due to mechanisms like reduction in ST/IFT and/or due to wettability alteration at the rock-oil-water interface, as observed during the course of study.

To isolate B. licheniformis W16 produced a potent lipopeptide biosurfactant, within 48–72 h in carbohydrate based minimal media. It reduced ST, IFT and also altered the wettability thus changing the hydrophobic surface to more water-wet. To the best of our knowledge this is the first report of wettability alteration using biosurfactant produced by B. licheniformis strain. The biosurfactant produced by B. licheniformis W16 was characterized as lipopeptide, similar to lichenysin-A. It was quite stable under wide range of harsh reservoir conditions like high temperatures, pH range and salinities. It also successfully produced around 24–26% of residual oil (S_or) from Berea sandstone core-plugs at 60°C. Further scale-up studies are recommended to better understand the economics of the biosurfactant applications at pilot-scale.

Conceived and designed the experiments: SJ, YA-W, SA-B, AE, ASA-B. Provided all the resources for performing experiments: YA-W, SA-B, AE, ASA-B, MA-M. Performed the experiments: SJ, AA-B. Analyzed the data: SJ, YA-W, SA-B, AE, ASA-B, AA-B. Drafting the work, revising it critically for important intellectual content, and final approval of the version to be published: all authors.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.