Experimental study of microbial enhanced oil recovery in fractured porous media using the halophilic bacterium Haloferax mediterranei

TL;DR

This study explores using the halophilic bacterium Haloferax mediterranei to improve oil recovery in fractured reservoirs, finding that optimal biomass concentration enhances recovery while excessive amounts reduce it.

Contribution

The study introduces optimized biomass concentration for microbial enhanced oil recovery in fractured porous media using Haloferax mediterranei.

Findings

Optimal biomass concentration (5.07 g/L) increased oil recovery by 23% through selective plugging.

Higher biomass concentrations reduced oil recovery due to excessive polymer production blocking pathways.

Permeability reductions of 53.5%, 62.5%, and 73.5% were observed for different biomass concentrations.

Abstract

Fractured reservoirs, comprising over 20% of the world’s oil reserves, present significant challenges to enhanced oil recovery (EOR). This study investigates the potential of the halophilic bacterium Haloferax mediterranei in microbial EOR (MEOR) for improving oil recovery in fractured porous media. Experiments using a heterogeneous glass micromodel were conducted at biomass concentrations of 5.07, 6.74, and 10.14 g/L. Results showed that the optimal biomass concentration (5.07 g/L) increased oil recovery by 23% through selective plugging. Higher biomass concentrations caused excessive polymer production, blocking communication pathways between the matrix and fractures, reducing oil recovery to 11.7% and 7.8%. Permeability reductions were evaluated in two core samples, achieving average reductions of 53.5%, 62.5%, and 73.5% for the respective biomass concentrations. Using the optimal biomass concentration (5.07 g/L), oil recovery in two similar core samples increased by 14% and 12.6%. These findings underscore the importance of optimizing biomass concentration to balance permeability reduction and oil recovery efficiency. The study highlights MEOR’s potential as a cost-effective, sustainable strategy for enhancing oil recovery in fractured reservoirs. It also provides valuable insights for improving future EOR operations, contributing to global energy resource sustainability.

Figures

Stages of oil saturation: Progression of oil saturation within the micromodel.

Stages of water injection: Visual representation of oil displacement during water injection.

Stages of microbial injection: Impact of microbial injection on fracture closure and residual oil displacement.

Stages of water reinjection: Enhanced oil recovery during the water reinjection phase.

Measured contact angle between oil and water on a core surface after aging.

Visual representation of fractured core samples prepared for permeability and oil recovery experiments.

Oil production observed during core flooding experiments.

The growth curve of the bacterium Haloferax mediterranei.

Oil recovery curves based on dimensionless time in the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 5.07 g/L (B).

Oil recovery curves as a function of dimensionless time during the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 6.74 g/L (B).

SEM Image of Haloferax mediterranei Cells^[36](#CR36)^.

Oil recovery curves as a function of dimensionless time during the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 10.14 g/L (B).

Micromodel images illustrating the effect of increasing biomass concentration on oil displacement from the matrix space. (A: 5.07 g/L, B: 6.74 g/L, C: 10.14 g/L)

Oil recovery curve for core C as a function of dimensionless time following microbial and water flooding.

Oil recovery curve for core D as a function of dimensionless time following microbial and water flooding.

Chemical Structure of Polyhydroxybutyrate (PHB).

Comparison of microbial cultures in different media: 500-milliliter cultures prepared with the proposed culture medium of the Iranian Genetic Resources Center (A) and 1000-milliliter cultures prepared with the cost-effective culture medium desc.

Micromodel design created using Corel Draw.

The constructed micromodel.

Experimental setup for micromodel.

X-ray diffraction spectroscopy analysis of the cores.

Experimental setup of core flooding system.

Contact angle change: Transition of glass surface from water-wet (A) to oil-wet (B) following wettability alteration.

Full text

Introduction

With the steady decline in natural reservoir productivity and the growing global energy demand, enhancing oil recovery has become a critical priority. As the reserve replacement rate from new discoveries continues to fall, advanced recovery strategies are essential for maximizing output from mature fields and addressing future energy needs^1,2^. Oil recovery methods are categorized as primary, secondary, and tertiary. While primary and secondary techniques recover up to 65% of the original oil in place, a considerable fraction (35–55%) remains trapped as residual oil within the reservoir^3^. This untapped oil, estimated to amount to 2 to 4 trillion barrels globally, is the primary target of advanced tertiary recovery technologies^4^. Carbonate reservoirs are often highly heterogeneous and exhibit low porosity, posing challenges for efficient oil recovery. These complexities highlight the growing need for chemical EOR methods tailored to carbonate systems. As production from sandstone reservoirs declines, research has increasingly focused on optimizing CEOR strategies for carbonates^5^.

Chemically enhanced oil recovery (CEOR) offers a promising approach for extracting residual oil after primary and secondary recovery, utilizing formulations such as alkaline, surfactant, and polymer flooding, as well as their combinations to improve displacement efficiency^6^. Microbial enhanced oil recovery (MEOR), a promising subset of these techniques, harnesses the metabolic capabilities of microorganisms to facilitate oil mobilization and recovery^7^. MEOR First proposed by Beckman in 1926 and later developed by Zobell in 1947^8^, MEOR utilizes microbial metabolites such as biosurfactants, biopolymers, organic acids, and biogas to modify reservoir conditions and mobilize residual oil. Recent advances in microbial biotechnology have positioned MEOR as a cost-effective and environmentally sustainable alternative to conventional recovery methods, owing to its low toxicity, biodegradability, and operational flexibility under diverse reservoir conditions^9,10^. MEOR strategies are broadly classified into in-situ and ex-situ methods. In-situ techniques stimulate native microbial populations, whereas ex-situ methods inject cultured microbes and nutrients into the reservoir. Among these, selective plugging the microbial blockage of high permeability zones has emerged as an effective means to improve sweep efficiency by diverting flow toward previously unswept, low-permeability regions^11^.

In addition to the functional classifications of MEOR, recent advances in biofilm engineering have enabled the development of tailored biofilms through precise nutrient modulation, enhancing oil displacement efficiency. Furthermore, metabolic pathway optimization via advanced genetic engineering techniques, such as directed protein evolution, is essential for enhancing the production of beneficial microbial metabolites such as biosurfactants and biopolymers and for fine-tuning microbial performance under diverse reservoir conditions^12,13^.

Carbonate-fractured reservoirs, which often contain immobile oil trapped within their matrix, pose unique challenges for conventional EOR techniques. During water flooding, for instance, injected water bypasses the matrix and flows through fractures, leaving significant amounts of oil unrecovered. By employing biopolymers and microbial activity to selectively block fractures, water flow can be redirected toward the matrix, enabling greater oil recovery^14^. Bacterial selective plugging enhances oil recovery by promoting biomass accumulation in high-permeability zones, which redirects fluid flow toward low-permeability regions and improves access to previously unswept areas^15^.

Many microorganisms enhance MEOR efficiency through exopolymer production, which facilitates mobility control, biofilm formation, and pore plugging in high-permeability zones. These processes depend on multiple factors, including nutrient availability, fluid dynamics, and microbial physiology. While inactive or non-producing strains are less effective, bacterial polysaccharides improve surface adhesion and offer protective benefits. Additionally, microbial activity may induce CO₂ generation and calcite precipitation, further modifying reservoir flow paths^16^. Selective plugging using biopolymers and bacterial cells in fractured carbonate reservoirs redirects water flow from high- to low-permeability zones, thereby enhancing oil recovery. Biopolymers, due to their high molecular weight and bonding capabilities, form network structures that reduce permeability. Their efficiency in EOR depends on viscosity, shear resistance, and thermal stability. This process can involve activating native microbes or injecting external strains and nutrients, which accumulate in larger pores and block fluid pathways^17^.

Quantitative research has explored the potential of biopolymers to enhance microbial recovery and their efficacy in increasing oil recovery. Xiaosha Lin et al. studied the Enterobacter cloacae strain ZL-02, which produces exopolysaccharides (EPS). The results showed that the EPS produced by this strain, composed of proteins and polysaccharides, can achieve a maximum yield of 9.76 g per liter and increase the oil recovery coefficient by 26.1%^18^. Ehsan Sabooniha et al. investigated the selective plugging effect of hydrophobic bacterial cells on secondary oil recovery performance. They injected purified water and a bacterial solution containing Acinetobacter strain RAG-1 into a micromodel with a heterogeneous porous medium.

While the injection of pure water displaced 41% of the oil, the bacterial solution achieved a significantly higher oil recovery of 59%^19^. Soodmand-Asli et al. evaluated the impact of Lactococcus misanthropes, a biopolymer-producing bacterium, in three micromodels with varying fracture angles. Their findings revealed that the production of an insoluble exopolymer called dextran resulted in the plugging of some communication pathways between the matrix and fractures. This selective plugging reduced permeability by 40%^20^. Gandler et al. investigated MEOR using Pseudomonas aeruginosa bacteria by injecting it into two types of porous media: glass beads and sandstone cores. Their permeability measurements showed reductions ranging from 20% to 54% in glass beads (initial permeability: 1375 millidarcies) and from 45% to 80% in sandstone cores (initial permeability: 13 millidarcies)^21^.

Suthar et al. utilized Bacillus licheniformis, an obligate anaerobic and heat-resistant bacterium, to produce polymer, biofilm, and biosurfactant. The biosurfactant effectively reduced interfacial tension from 72 to 34 milliNewtons per meter. Microscopic analysis indicated that biofilm formation in a sand-packed medium and the selective plugging of high-permeability pathways resulted in a 27.7% increase in oil recovery^22^ Jeong et al. developed a MEOR simulation using Lactobacillus mucosae and modeled fracture plugging via dextran production. Stoichiometric and kinetic equations described microbial growth and polymer formation. The model predicted a 5.61% increase in oil recovery over water flooding, with improvements reaching 7.36% under optimal dextran distribution^23^.

Kuto et al. utilized the bacterium Rhizobium viscosum (two strains, CECT 908) for biopolymer production. This bacterium exhibited superior rheological properties and higher viscosity compared to xanthan gum, withstanding temperatures up to 80 °C and salinities up to 200 g/l of NaCl. EOR experiments demonstrated that the biopolymers produced by Rhizobium viscosum outperformed xanthan gum, resulting in a 7.25 ± 0.5% improvement in the recovery coefficient. Notably, this bacterium produces 0.23 ± 0.11 g of biopolymer after 96 h of growth^24^. ElShafie et al. investigated the production of polyhydroxyalkanoate (PHA) by a native strain of Aeribacillus polyanthracis in Oman. Their findings showed a biopolymer production rate of 1.68 g per liter, which increased the viscosity of the fluid from 1 to 5 cp. to 26–59 cp. Applying this biopolymer in sandstone reservoirs led to a 9.4% improvement in oil recovery^25^.

Additionally, Al-Arimi et al. isolated a local strain of Aeribacillus and examined its morphological, molecular, and MEOR performance characteristics. Slurry flooding experiments conducted with this strain demonstrated its efficacy in recovering both light and heavy oils through the production of pullulan. Under optimal conditions, the strain produced up to 10 g per liter of pullulan. In MEOR experiments conducted after water injection, the application of pullulan facilitated the recovery of 34% of the remaining light oil. For heavy oil, the use of undiluted and diluted pullulan in saline water resulted in recoveries of 7.36% and 23.20%, respectively^26^.

Volpon et al. demonstrated that polymer-producing bacteria isolated from the Carmpolis oil field selectively penetrated high-permeability zones during slurry flooding. Reservoir pressure data confirmed microbial growth. The adhesive biopolymers produced by the B3 microbial system enhanced fluid mobility control, resulting in a 20% improvement in oil recovery^27^. Yongqiang et al. evaluated a synergistic MEOR strategy using Enterobacter cloacae (biopolymer producer) and Pseudomonas aeruginosa (biosurfactant producer). Laboratory studies confirmed compatibility between the strains and a reduction in interfacial tension from 70 to 30 mN/m. Core flooding tests showed that co-injection led to 17.4% oil recovery, outperforming individual strain injections. The combined application significantly enhanced oil recovery in heterogeneous reservoirs^28^.

Zhao et al. isolated Pseudomonas stutzeri XP1 from Shengli oil reservoirs and identified its ability to produce up to 16 g/L of adhesive dextran using corn starch and nitrate. Slurry flooding experiments showed a 13.56% increase in heavy oil recovery, confirming the strain’s effectiveness in MEOR^29^. Joshi et al. investigated the biopolymer schizophyllan, produced by Schizophyllum commune (ATCC 38548), for MEOR applications. Characterization studies and core flooding experiments at 45 °C revealed that schizophyllan significantly enhanced heavy oil recovery, achieving a 28% increase in oil production^30^. Huang et al. assessed Enterobacter cloacae (FY-07) for MEOR applications using glass micromodels. The bacterium demonstrated effective selective plugging, with rates ranging from 67% to 78% depending on permeability. Core flooding experiments showed a 10.54% increase in oil recovery^31^.

Elyasi Gomari et al. investigated the effect of MEOR techniques on oil recovery in oil-wet fractured carbonate rocks. By injecting a combination of xanthan gum (XG) as a biopolymer and RL as a biosurfactant, they achieved a recovery factor improvement of 7–9%. The biopolymer reduced the permeability of the fracture network by 18%, diverting fluids into the oil-wet matrix and enhancing water imbibition, while the biosurfactant altered the rock wettability to a water-wet state. SEM analysis confirmed the deposition of the biopolymer and biosurfactant on the rock surface^32^.

Xiaosha Lin et al. evaluated Enterobacter cloacae in MEOR, showing that microbial EPS production significantly reduced permeability. Simulated flooding tests achieved an RF of 2.5, an RRF of 5.45, and a plugging efficiency of 81.5%, demonstrating the strong potential of EPS for flow restriction and enhanced oil recovery^33^. The experiments focused on the metabolic capacity of Haloferax mediterranei to selectively close pores and enhance oil recovery. This halophilic archaeon is an aerobic, filamentous microorganism with exceptional tolerance to salt and temperature, thriving in environments with salinity levels ranging from 10% to 32.5% (~ 325,000 ppm) and temperatures between 32 °C and 52 °C, and it can also grow at temperatures below 32 °C within a pH range of 5.75–8.75. In addition to tolerating high salinity, H. mediterranei shows resilience to elevated concentrations of toxic metals such as nickel, lithium, cobalt, and arsenic, which inhibit the growth of many other microorganisms^34^.

Unlike conventional MEOR bacteria (e.g., Enterobacter, Pseudomonas, Bacillus spp.), which often fail to survive or maintain polymer performance under such extreme conditions, H. mediterranei sustains consistent growth and metabolite production in hypersaline environments. Importantly, it synthesizes polyhydroxybutyrate (PHB), a biodegradable thermoplastic biopolymer with remarkable thermal stability, retaining approximately 95% of its structural integrity up to 250 °C^34,35^.

Figure 1 shows the rod-shaped morphology of H. mediterranei (0.2–0.8 μm) as revealed by SEM analysis, while Fig. 2 presents the chemical structure of PHB, underscoring its relevance to microbial plugging and oil recovery^36^. In fractured carbonate reservoirs, conventional waterflooding often fails to displace oil trapped within the matrix due to the preferential flow of water through high-permeability fracture networks. MEOR emerges as a promising alternative, leveraging microbial activity and metabolite production in aqueous environments to redirect water flow toward the matrix through mechanisms such as selective plugging and biopolymer production, thereby enhancing oil displacement efficiency. Furthermore, the high cost of synthetic surfactants and polymers, coupled with their adsorption onto reservoir rock surfaces, raises concerns about the economic viability of conventional chemical enhanced oil recovery methods. Consequently, the application of natural surfactants and biopolymers has garnered increasing attention due to their biodegradability, reservoir compatibility, and cost-effectiveness^37,38^.

Fig. 1SEM Image of Haloferax mediterranei Cells^36^.

Fig. 2. Chemical Structure of Polyhydroxybutyrate (PHB).

Although MEOR has demonstrated significant potential for improving oil production, its application in fractured carbonate reservoirs remains largely underexplored. Most previous studies have concentrated on non-fractured systems, resulting in a substantial knowledge gap regarding the effectiveness of MEOR in the complex dynamics of fractured reservoirs. In particular, the role of microbial biomass concentration in simultaneously reducing fracture permeability while preserving matrix accessibility has not been thoroughly investigated.

Although the concept of an optimal biomass concentration is well recognized in MEOR, no quantitative criteria have previously been reported for the halophilic archaeon Haloferax mediterranei. To the best of our knowledge, this study provides the first systematic investigation of biomass optimization of H. mediterranei in fractured carbonate systems. By producing the biodegradable biopolymer polyhydroxybutyrate (PHB), which remains stable under high salinity and temperature, H. mediterranei offers a promising, environmentally friendly strategy for improving oil recovery. A series of experiments using glass micromodels and fractured carbonate core samples were conducted to evaluate the effects of microbial injection on permeability reduction and oil displacement efficiency. This work establishes the first set of quantitative benchmarks for biomass optimization of this archaeon and presents a novel framework for assessing microbial performance in fractured reservoirs, thereby promoting sustainable and effective MEOR strategies for challenging carbonate reservoirs.

Materials and methods

Microorganisms

The halophilic archaeon Haloferax mediterranei was employed in this study, considering its compatibility with reservoir conditions and its growth characteristics. The archaeal strain was obtained from the Iranian Biological Resource Center (IBRC-M 10337, designated as R-4, type strain). The strain passport is available through IBRC and corresponds to international collections including DSM 1411, ATCC 33,500, BCRC 15,765, and CCM 3361, and it is classified as biosafety level 1. The strain was originally isolated as polymorphic pigmented rods, morphologically resembling Halobacterium volcanii, from a solar salt pond near Alicante, Spain^39^.

The application of H. mediterranei in enhanced oil recovery (EOR) is particularly significant due to its ability to synthesize polyhydroxybutyrate (PHB), which facilitates biofilm formation. Nevertheless, while this mechanism is considered primary, the potential roles of other microbial metabolites were not assessed in this study. As a limitation of this research, no dedicated experiments were performed to evaluate the effects of salinity, pH, or temperature on the growth of H. mediterranei, because the primary objective was to assess its performance in MEOR within fractured carbonate systems. Growth characteristics are therefore cited from established literature (e.g., Matarredona et al., 2021).

Growth condition

To ensure the viability and proliferation of the microorganism, the primary culture medium supplied by the Iranian Genetic Resources Center was initially used. A 200-milliliter microbial solution was prepared, comprising 160 milliliters of the primary culture medium and 40 milliliters of microbial inoculum. The preparation followed the recipe detailed in Table 1, with all components combined in a 500-milliliter Erlenmeyer flask. The medium was thoroughly mixed using a magnetic stirrer, and the pH was adjusted to the optimal value of 7.2 ± 0.2 using a one-molar solution of sodium hydroxide and sulfuric acid. To ensure reproducibility and scalability, microbial solutions were prepared in both 500-milliliter and 1000-milliliter Erlenmeyer flasks using two different culture media. The primary culture medium provided by the Iranian Genetic Resources Center was employed for the 500-milliliter cultures (Fig. 3A). Subsequently, after 5 days of growth under optimal conditions, the microbial culture was transferred to a more cost-effective medium, as described in Table 2, for 1000-milliliter cultures (Fig. 3B). This transition allowed for economical enhancement of polymer production while maintaining the microorganism’s viability and proliferation.

Furthermore, it is noteworthy that after pH adjustment, the flasks were covered with aluminum foil, sterilized with sterile gas, and sealed with cotton plugs. The culture medium was then sterilized by autoclaving at 121 °C for 20 min. Glucose was added to the medium after autoclaving to prevent degradation caused by its reaction with other substances during the autoclaving process. Following autoclaving, the flasks were cooled, and inoculation was performed within a sterile hood, which had been sterilized with UV light for 20 min, near an alcohol lamp to maintain a sterile environment. Each Erlenmeyer flask received 40 milliliters of microbial inoculum. To ensure adequate aeration, proper mixing of the culture medium, and to minimize evaporation over time, 40% of the flask’s capacity was reserved for the microbial solution.

After inoculation, the microbial solution was incubated in a rotating incubator at 37 °C under optimal growth conditions, which included a pH of 7.2 ± 0.2 adjusted with 1 M NaOH or H₂SO₄, a flask volume-to-culture ratio of 40% to ensure adequate aeration and minimize evaporation, and continuous agitation on a rotary shaker. These conditions were maintained for 5 days to allow active proliferation of the microorganism. Following this period, Following this period, about 20% of the biomass from the primary culture was inoculated into the secondary medium (Table 2) for further growth and polymer production experiments. This alternative medium was formulated to support enhanced polymer production while reducing cultivation expenses. The actively growing microorganism, previously cultured in one-liter Erlenmeyer flasks, was subsequently transferred to this medium for further experimentation and analysis^40,41^.

Fig. 3. Comparison of microbial cultures in different media: 500-milliliter cultures prepared with the proposed culture medium of the Iranian Genetic Resources Center (A) and 1000-milliliter cultures prepared with the cost-effective culture medium desc.

Table 1. Composition of the activation culture medium for haloferax mediterranei as proposed by the Iranian genetic resources Center^42^.SubstanceConcentrationNaCl184 g/lMgCl_2_.6H_2_O23 g/lMgSO_4_.7H2O26.83 g/lCaCl_2_5 mlNaBr0.61 g/lNaHCO_3_0.15 g/lVitamin B250 µlKCl7.76Yeast Extract1 g/lTryptone Broth1.7 g/lPeptone3.3 g/lAgar18 g/lTrace element SL-101.7 g/lThe SL-10 trace element solution contains the following components and concentrations: 10.00 ml of HCl (25%, 7.7 M), 1.50 g of FeCl_2_·4H_2_O, 70.00 mg of ZnCl₂, 100.00 mg of MnCl_2_·4H_2_O, 6.00 mg of H_3_BO_3_, 190.00 mg of CoCl_2_·6H_2_O, 2.00 mg of CuCl_2_·2H_2_O, 24.00 mg of NiCl_2_·6H_2_O, 36.00 mg of Na_2_MoO_2_·2H_2_O, and distilled water to a final volume of 990.00 ml.

Table 2. Composition of the optimized low-cost culture medium for Haloferax mediterranei designed to enhance polymer production^41^.SubstanceConcentration (g/l)NaCl195MgCl_2_34.6MgSO_4_44.9CaCl_2_0.92NaBr0.58NaHCO_3_0.17Yeast Extract5KCl0.5Glucose20

To obtain the different biomass concentrations required for subsequent experiments, two complementary approaches were employed. (i) Dilution: harvested microbial suspensions were diluted with sterile, cell-free MGM medium to lower the biomass concentration while maintaining identical chemical composition and osmotic pressure, thereby avoiding any osmotic or nutritional shock to the cells. (ii) Concentration: higher biomass levels were achieved either by extending the incubation time until the desired cell density was reached or, when very high concentrations were needed, by gentle centrifugation of the culture to remove part of the supernatant followed by resuspension of the cell pellet in a smaller volume of fresh sterile MGM medium. This combined strategy of controlled growth time and dilution/concentration enabled the preparation of three precisely defined biomass levels, each of which was verified by dry-weight measurements before use.

Fluids

Water flooding

For flooding experiments and the calculation of permeability and porosity, saline water was prepared with a concentration of 30,000 ppm using distilled water and sodium chloride. The density of this saline solution was determined to be 2.16 g per cubic centimeter, ensuring consistency and reliability for use in flooding tests.

Oil

The crude oil utilized in this study was obtained from the Marun oil field, located in Ramhormoz, Iran. This specific crude oil was selected due to its relevance to the region’s reservoir conditions and its chemical composition, which aligns with the objectives of this research. The key physical and chemical properties of the crude oil are provided in Table 3.

Table 3. Physical and chemical properties of crude oil used in the study.PropertiesQuantityAPI33/32Density at 60 degrees Fahrenheit857/7 kg/m^3^Specific gravity at 60 degrees Fahrenheit0/8585Kinematic viscosity at 10 degrees Celsius16/1 cStKinematic viscosity at 20 degrees Celsius10/5 cStKinematic viscosity at 40 degrees Celsius6/3 cStWater (volume percent)0/2Asphaltene (weight%)0/52Wax (weight%)3/8Hydrogen sulfide> 1ppmSulfur (weight%)1/6Total acid number0/14 mgKOH/G

The glass micromodel

To investigate the effects of MEOR in a porous medium and to observe fluid movement, a micromodel was constructed using 6-millimeter thick glass plates. This micromodel provides a visual platform for studying increased oil extraction in a controlled environment, replicating the complex conditions of porous and fractured reservoirs. The design of the micromodel was created using Corel Draw software and subsequently laser-engraved onto one of the glass plates^43^. The two plates were then bonded under carefully controlled temperature conditions to ensure structural integrity and functionality.

Designing the porous media

The porous medium was carefully engineered to replicate a matrix–fracture system analogous to that of naturally fractured reservoirs. Irregularly shaped matrix blocks were embedded within diagonally oriented fractures, with a measured fracture width of 1.8 mm. The matrix porosity, excluding fractures, was calculated at 29%, providing a realistic approximation of reservoir properties. The geometric configuration was developed using CorelDRAW software (Fig. 4), featuring a symmetrically designed inlet–outlet system on both sides of the micromodel to ensure uniform fluid distribution. This structural symmetry and spatial arrangement enhance fluid sweep efficiency and allow for a more accurate simulation of the complex flow dynamics characteristic of fractured reservoirs, thereby improving the evaluation of MEOR performance.

Fig. 4. Micromodel design created using Corel Draw.

Glass etching

The micromodel construction involved the use of two glass plates, each measuring 17 cm in length, 10 cm in width, and 6 mm in thickness. The porous section of the model measures 10 cm in length and 5 cm in width. The desired pattern of the porous environment and fractures was engraved onto one plate using a laser-engraving device, ensuring precise replication of the designed structure. The engraved sections were then machined to the required depth, and both plates were drilled at the inlet and outlet positions using a diamond drill bit to accommodate needle fittings.

Bonding glass sheets

To assemble the micromodel, the two glass plates were fused in a programmable furnace under strictly controlled thermal conditions to ensure reproducibility and prevent glass fracturing. The temperature was initially increased from room temperature to 300 °C at a rate of approximately 3 °C/min and held for 30 min to remove residual moisture. It was then raised to 520 °C at 2 °C/min and held for 20 min to relieve internal stresses. Subsequently, the furnace temperature was increased in increments of 73 °C up to a maximum bonding temperature of 660 °C, with sufficient resting periods at each stage, and maintained at 660 °C for 45 min to achieve complete glass fusion. Finally, the furnace was cooled to room temperature at a controlled rate of approximately 1.5 °C/min to avoid thermal shock and microcrack formation. Upon completing the bonding process, needles were attached to the inlet and outlet ports of the micromodel to facilitate fluid injection and withdrawal during experiments. The final constructed micromodel (Fig. 5) provides a robust and accurate representation of a fractured reservoir, making it an ideal platform for studying fluid flow and microbial recovery processes^44^.

Fig. 5. The constructed micromodel.

Micromodel experimental setup

To perform the micromodel experiments, essential equipment included a syringe pump for fluid injection, a light source to enhance image clarity, and a digital camera (SAMSUNG NX2000) for high-resolution image capture. The experimental setup is illustrated in Fig. 6, highlighting the arrangement of these components. The syringe pump ensured consistent and precise fluid injection into the micromodel, while the light source was positioned beneath the micromodel to illuminate the porous environment and enhance visibility of fluid movement. The camera was placed above the micromodel and connected to a computer via the Samsung Smart Camera App. Image capture was automated using Clicker Automatic software, which triggered photography at specified time intervals throughout the flooding process. The recorded images were systematically analyzed using IMAGE J software, allowing for quantification of saturation changes and evaluation of ultimate oil recovery. By analyzing the colored areas occupied by oil within the micromodel, the software provided detailed insights into the efficiency of oil displacement and microbial-enhanced recovery. The uniform depth and precise design of the micromodel ensured the accuracy and repeatability of the experiment.

Fig. 6. Experimental setup for micromodel.

Cores

The cores used in this research were composed primarily of carbonate and dolomite. To confirm the dolomitic composition of the cores, an X-ray diffraction (XRD) test was performed. The XRD results (Fig. 7) validated the dolomitic nature of the cores, showing characteristic diffraction peaks that indicate dolomite’s crystal structure. The diagram illustrates the intensity of the reflected X-rays as a function of the diffraction angle. In XRD analysis, X-rays are directed at the sample at various angles, and the resulting diffraction pattern is recorded and analyzed. Since each material has a unique diffraction pattern, this method is highly effective for identifying the atomic structure, crystal phase, grain size, crystal morphology, and interlayer spacing of materials. This ensures that the cores used in the experiments were representative of dolomitic reservoirs. The physical properties of the cores used in this study are presented in Table 4. These properties, including core dimensions, dry and wet weights, void volume, and porosity, were carefully measured and analyzed to ensure accurate experimental results. Porosity values ranged from 15% to 24%, highlighting the variability of the pore structures within the cores, which closely resemble natural reservoir conditions. To prepare the cores for experimentation, they were cleaned using the Soxhlet extraction method with toluene and methanol as solvents. Toluene was used to remove non-polar compounds like oil, while methanol was employed to extract polar compounds such as water and salts. The cleaning process continued until no further color changes were observed in the solvents, indicating the complete removal of contaminants. This process took two to three weeks for toluene and three to seven days for methanol. After cleaning, the cores were dried in an oven at 100 °C for 24 h to eliminate residual solvents and moisture, ensuring their suitability for subsequent experiments with consistent and reliable results.

Fig. 7X-ray diffraction spectroscopy analysis of the cores.

Table 4. Physical properties of the cores.No.Core nameLength (cm)Diameter (cm)Dry weight (gr)Wet weight (gr)Void volume (cm³)Porosity (%)1A7.13.75162.56178.81616202B6.653.75170.38182.0211.4153C6.63.75153.45171.3917.66244D7.53.75195.10207.7012.415

Core flooding

The core flooding experiments were conducted using the PetroAzma-S14 model apparatus, specifically designed for EOR studies. This apparatus consists of several key components, including transfer vessels for holding fluids such as water, oil, and gas, a precision injection pump, a core holder, a manual hydraulic pump for adjusting overburden pressure, and integrated sensors for temperature and pressure measurement. The core holder and fluid storage cylinders are constructed from stainless steel to ensure durability and compatibility with various fluids under high-pressure and high-temperature conditions. The apparatus is equipped with a touchscreen user interface for monitoring and recording real-time pressure and temperature data. Additionally, the system can be connected to a computer for data transfer, enabling the storage and analysis of pressure-time graphs in Excel format. This feature facilitates accurate tracking of core flooding experiments and ensures the reproducibility of results. The experimental setup, schematically depicted in Fig. 8, is designed to simulate reservoir conditions by controlling pressure, temperature, and fluid flow precisely. The transfer vessels enable seamless switching between fluids, and the core holder securely houses the core sample while maintaining uniform pressure and temperature. The system’s nitrogen inlet is used for pressurizing the confining vent and maintaining overburden pressure, which is critical for replicating in-situ reservoir conditions during the flooding process. The PetroAzma-S14 apparatus offers reliable functionality for conducting detailed flooding experiments, providing essential insights into fluid flow dynamics, permeability changes, and recovery efficiency under controlled laboratory conditions.

Fig. 8. Experimental setup of core flooding system.

The experimental procedure

Preparation of fluids

To evaluate the efficiency of MEOR and to simulate reservoir conditions, a comprehensive experimental procedure was designed. This procedure involved the preparation of fluids, the measurement of biomass, and a series of micromodel experiments, including oil saturation, water injection, microbial injection, and water reinjection. The experimental setup ensured precise replication of real reservoir conditions and provided detailed insights into the interaction between fluids, microbes, and the porous medium. For fluid preparation, formation water was prepared by dissolving 3 g of 99% pure sodium chloride in 100 milliliters of distilled water and homogenizing the solution with a magnetic stirrer. This resulted in a saltwater solution with a calculated concentration of 30,000 ppm and a density of 1.016 g per milliliter. Crude oil, used for flooding experiments, was centrifuged at 9000 rpm for 30 min to separate wax and asphaltene compounds, thereby preventing clogging in the oil flooding path. The viscosity of the centrifuged oil was measured using a rotary viscometer, yielding a value of 13 centipoises, which was used for permeability-related calculations.

Measurement of dry biomass weight

The dry weight of the biomass was determined through a series of precise steps to ensure accurate measurement. Initially, two dry Falcon tubes were weighed individually to establish their baseline weights. Subsequently, 40 milliliters of microbial solution were transferred into each Falcon tube. The tubes were then centrifuged at 3000 rpm for 20 min to separate the biomass from the supernatant. After the initial centrifugation, the supernatant was carefully removed, leaving the biomass at the bottom of each tube. To further purify the biomass, it was washed with a 10% saline solution (by weight) and centrifuged again under the same conditions. Following this second centrifugation, the supernatant was discarded once more, and the tubes were prepared for drying. The Falcon tubes containing the biomass were placed in an oven set to a temperature of 35 °C to ensure complete drying. Once dried, the final weight of each Falcon tube was recorded. The dry weight of the biomass was then calculated as the difference between the weight of the tube containing the dried biomass and the weight of the empty tube. In this experiment, the dry biomass weights recorded for the two Falcon tubes were 5.27 g and 4.87 g, respectively. The average biomass weight was calculated as 5.07 g per liter, which served as the baseline concentration for subsequent experimental procedures.

Micromodel experiments

The micromodel experiments were conducted to simulate key processes in MEOR and to investigate the impact of microbial activity on oil displacement and fracture closure under controlled conditions. All experiments were performed in three independent replicates. Each step of the experiment was carefully designed and executed to ensure accurate and reproducible results. In this study, microbial injection was designed in an ex-situ manner, where the active biomass and primary metabolites of Haloferax mediterranei were grown and prepared in the optimized culture medium (37 °C, high salinity) prior to injection. The main objective was to evaluate the real-time effect of the prepared microbial solution on permeability reduction and oil displacement during flow.

1. Measurement of micromodel depth: The depth of the micromodel, critical for understanding fluid flow dynamics, was determined using a roughness test. This test employed a needle-like device that traversed the micromodel surface, measuring vertical deviations. The maximum depth was recorded as 165 microns, reflecting the uniformity and precision of the micromodel’s construction.
2. Measurement of pore volume: The pore volume of the micromodel was calculated to quantify the volume available for fluid flow, a crucial parameter for simulating the porous environment of fractured reservoirs. This measurement involved determining the weight of the glass plate in both its intact and cut states. The difference in weight was then used in conjunction with the density of the glass to calculate the pore volume, using Eq. (1):

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:PV=\frac{(\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\:\mathrm{o}\mathrm{f}\:\mathrm{t}\mathrm{h}\mathrm{e}\:\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{r}\mathrm{o}\mathrm{m}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{l}\:\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}\:\mathrm{b}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{g}\:\mathrm{c}\mathrm{u}\mathrm{t}\:-\:\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\:\mathrm{o}\mathrm{f}\:\mathrm{t}\mathrm{h}\mathrm{e}\:\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{r}\mathrm{o}\mathrm{m}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{l}\:\mathrm{a}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{r}\:\mathrm{b}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{g}\:\mathrm{c}\mathrm{u}\mathrm{t})\:}{\mathrm{D}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{y}\:\mathrm{o}\mathrm{f}\:\mathrm{t}\mathrm{h}\mathrm{e}\:\mathrm{g}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s}}$$\end{document}

Based on the described specifications, the total pore volume of the micromodel was estimated to be 0.4 cm^3^.

• 3.3. Alteration of wettability To replicate the conditions of fractured carbonate reservoirs, the naturally water-wet glass surface of the micromodel was chemically treated to create an oil-wet surface. The wettability alteration process included:

• Acid wash: The micromodel was washed with a 20% hydrochloric acid solution to clean the surface. (Omitting this step would result in neutral wettability.)

• Rinsing and drying: The micromodel was rinsed with distilled water and dried in an oven at 100 °C for two hours.

• Silane treatment: The glass surface was treated with a solution containing 2% trichloromethylsilane in 98% toluene for 15 min. This reaction formed a methylpolysiloxane layer on the glass, rendering the surface hydrophobic.

• Final rinse and drying: The micromodel was rinsed with methanol and dried in an oven to complete the wettability alteration process. The effectiveness of the wettability change was verified by contact angle measurements, as shown in Fig. 9, demonstrating the transition from water-wet to oil-wet conditions^45^.

Fig. 9. Contact angle change: Transition of glass surface from water-wet (A) to oil-wet (B) following wettability alteration.

• 4.Oil Saturation Oil saturation of the micromodel was performed by injecting oil at a controlled flow rate of 0.1 milliliters per minute until the entire porous network was fully saturated. The process was systematically documented through high-resolution photographs taken every 20 s to capture the progression of oil distribution. These stages are illustrated in Fig. 10.

Fig. 10. Stages of oil saturation: Progression of oil saturation within the micromodel.

• 5.Water injection Following oil saturation, water was injected into the micromodel at a calibrated flow rate of 0.009 milliliters per minute to displace the oil within the fracture and surrounding porous medium. The injection process was monitored through images captured at 10-minute intervals, providing a detailed visual representation of the oil displacement process. The progression of water injection and oil recovery is depicted in Fig. 11.

Fig. 11. Stages of water injection: Visual representation of oil displacement during water injection.

6. Microbial injection To facilitate fracture closure and enhance oil recovery, microbes were injected into the micromodel at a flow rate of 0.009 milliliters per minute, occupying 25% of the porous space. The microbial activity aimed to seal the fractures, preventing further water loss during subsequent reinjection phases. Images captured at 10-minute intervals showcased the progression of microbial injection and its effectiveness in fracture closure, as seen in Fig. 12.

Fig. 12. Stages of microbial injection: Impact of microbial injection on fracture closure and residual oil displacement.

7. Reinjection of water To evaluate the impact of microbial activity on enhancing oil recovery, water was reinjected into the micromodel at the same flow rate of 0.009 milliliters per minute. Unlike the initial water injection phase, this stage allowed water to penetrate the matrix, displacing additional residual oil. Systematic photographs captured at 10-minute intervals were used to analyze the displacement process and to construct a detailed oil recovery curve. The stages of water reinjection are shown in Fig. 13.

Fig. 13. Stages of water reinjection: Enhanced oil recovery during the water reinjection phase.

Each biomass concentration (5.07, 6.74, and 10.14 g/L) was evaluated in an independent micromodel run following the same protocol. Stage-1 denotes the baseline waterflood without microbes. Minor differences in Stage-1 recovery between runs arise from small variations in initial water saturation (S_wi_​), pore-scale heterogeneity, and measurement uncertainty.

Core flooding experiments

The core flooding experiments were meticulously designed to simulate subsurface reservoir conditions and evaluate MEOR efficiency. All experiments were performed in three independent replicates. The methodology includes precise measurements, preparation of core samples, and systematic flooding procedures.

Saturation of samples with formation water and measurement of porosity

Porosity was determined from dry and brine-saturated weights after vacuum saturation with formation brine (30,000 ppm NaCl). Each oven-dried core was evacuated in a desiccator for ~ 5 h, brine was then admitted to the evacuated chamber and the core was submerged for 72 h to ensure full saturation; afterwards, the core was flooded with excess brine (> 1 pore volume) to verify saturation. The dry ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Wdry$$\end{document} ) and saturated ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Wwet$$\end{document} ) weights were recorded and pore volume was calculated using Eq. (2). The bulk volume was determined from core dimensions [Eq. (4)], porosity from Eq. (3), and rock density from Eq. (5). W_dry_ and W_wet_ are the dry and wet weights of the core, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\uprho\:}\mathrm{b}\mathrm{r}\mathrm{i}\mathrm{n}\mathrm{e}$$\end{document} is the density of the formation water, and V_b_ represents the bulk volume of the core^46^.

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{v}_{p}=\:\frac{Wdry-Wwet\:}{{\uprho\:}\mathrm{b}\mathrm{r}\mathrm{i}\mathrm{n}\mathrm{e}}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\varphi\:=\frac{Vp}{Vb}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{V}}_{b}=\frac{{\pi\:\:d}^{2}}{4}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\uprho\:}}_{Rock}=\frac{Wdry}{Vb-Vp}$$\end{document}

Determining core absolute permeability

Absolute permeability was measured following Darcy’s law. The core was saturated with formation water, and flow rates were varied across four levels while pressure drops were recorded. The relationship between flow rate and pressure drop was plotted, and the permeability (K) was calculated using Darcy’s equation (Eq. (6)):

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathrm{q}=\frac{\mathrm{K}\mathrm{A}}{{\upmu\:}\mathrm{L}}{\Delta\:}\mathrm{P}$$\end{document}

Here, q is the flow rate, A is the cross-sectional area of the core, µ is the fluid viscosity, L is the core length, and ΔP is the pressure drop. The slope of the flow rate vs. pressure drop graph provided K, the permeability.

Crude oil preparation for core flooding

To avoid clogging caused by wax and asphaltene, crude oil was centrifuged at 9000 rpm for 30 min. The viscosity of the centrifuged oil was measured using a rotating viscometer, yielding 13 cp. This viscosity value was critical for accurate permeability calculations.

Irreducible water saturation

To achieve irreducible water saturation, crude oil was injected into the core until no water was produced from the outlet. The injection rate was calculated based on the fluid velocity (1 ft/day) and the core’s cross-sectional area, set at 0.2 mL/min. Irreducible water saturation was calculated using Eq. (7):

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathrm{I}\mathrm{r}\mathrm{r}\mathrm{u}\mathrm{d}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{b}\mathrm{l}\mathrm{e}\:\mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r}\:\mathrm{s}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}=\frac{\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{e}\:\mathrm{v}\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\:\mathrm{s}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{e}-\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{e}\mathrm{d}\:\mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r}}{\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{e}\:\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\:\mathrm{s}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{e}}\:$$\end{document}

Aging and alteration of core wettability process

To render the core oil-wet, samples were immersed in crude oil at 75 °C for three weeks. This process facilitated the absorption of polar components, altering the wettability. Wettability alteration was verified by measuring the contact angle between oil and water, which was found to be 154 degrees using IMAGE J software (Fig. 14).

Fig. 14. Measured contact angle between oil and water on a core surface after aging.

Cutting and fracturing of core samples

Core samples were carefully fractured using a diamond drill bit to ensure precision and minimize material waste during the process. The fractured cores were then systematically categorized into two distinct groups based on their intended experimental use. The first group, consisting of Cores A and B, was designated for permeability experiments without prior aging, focusing on understanding the permeability characteristics of freshly fractured samples under standard conditions. The second group, including Cores C and D, was reserved for oil recovery experiments conducted post-aging. These cores underwent a detailed aging process, involving immersion in crude oil at elevated temperatures to alter their wettability to an oil-wet state, simulating real reservoir conditions. The fracturing process significantly enhanced the permeability of the cores, as evident from permeability measurements and visual observations. This increase in permeability reflects the structural changes induced by the fracturing process, facilitating fluid flow through the samples. Figure 15 provides a visual representation of the fractured cores within the core holder, showcasing the precision of the fracturing process and its impact on the core structure. This careful categorization and preparation of the fractured cores ensured that each experimental group was tailored to meet the specific objectives of permeability assessment and oil recovery evaluation, providing robust and reliable data for further analysis.

Fig. 15. Visual representation of fractured core samples prepared for permeability and oil recovery experiments.

Permeability measurement in fractured samples

Absolute permeability in fractured cores was measured by injecting formation water at four different flow rates, recording corresponding pressure drops, and plotting the flow rate vs. pressure drop graph. Permeability was recalculated using Darcy’s law, revealing a significant increase due to fracturing, particularly at a confining pressure of 500 psi.

Investigating permeability changes after microbial flooding

To evaluate microbial-induced permeability reduction, fractured cores were injected with microbes at a rate of 0.2 milliliters per minute under 500 PSI confining pressure. Microbial injection corresponded to the void space volume in the samples. Water injection followed microbial flooding at varying flow rates, with pressure drops recorded to assess permeability changes. This experiment, conducted at biomass concentrations of 5.07, 6.74, and 10.14 g per liter, demonstrated significant permeability reduction and fracture closure.

Microbial flooding experiment for oil recovery assessment

Two fractured cores with similar petrophysical properties were used to assess MEOR efficiency. Initially, water was injected into the cores at 0.2 milliliters per minute until oil production ceased. Microbes were then introduced into the core, followed by a second water flooding phase. Oil production during this phase indicated successful microbial-induced fracture closure and enhanced recovery. Oil output was measured using graduated cylinders, with readings taken at 15-minute intervals. The oil recovery curve was plotted based on the cumulative data. When oil output contained emulsified water, centrifugation was employed to separate the two phases. Figure 16 illustrates oil production during the flooding process.

Fig. 16. Oil production observed during core flooding experiments.

Definitions and reporting conventions

To ensure clarity, all recovery values in this study are reported on an oil originally in place (OOIP) basis. Stage-1 denotes the baseline waterflood with no microbes present, and its value is reported as cumulative recovery after completion of the waterflood, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{R}_{S1,cum}$$\end{document} = \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\frac{oil\:produced\:during\:Stage\:1}{OOIP}\times\:100$$\end{document} (% OOIP). Stage-2 and stage-3 denote the MEOR stages and are reported as incremental percentage-point gains relative to OOIP, measured from the end of the preceding stage: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{R}_{Sk,inc}=\frac{oil\:produced\:during\:Stage\:k}{OOIP}\times\:100$$\end{document} . Throughout the manuscript, the term enhanced oil recovery factor in stage-2/3 refers to these incremental gains (percentage-points of OOIP), whereas stage-1 values are cumulative. Each biomass concentration (5.07, 6.74, and 10.14 g/L) was evaluated in an independent micromodel run under an identical flooding protocol; minor differences observed in stage-1 reflect run-to-run variability (small variations in initial water saturation, and measurement uncertainty) and are not microbial effects.

Results and discussion

Growth curve

The growth behavior of Haloferax mediterranei was monitored by measuring the optical density (OD₆₀₀) of the culture at regular time intervals using a UV-1601 spectrophotometer (RayLeigh, China). The measurements were performed at a wavelength of 600 nm. Sterile culture medium of the same formulation as the samples was used as the blank control. Each time point was measured in three independent replicates (n = 3) to ensure data consistency and reliability.Sampling and OD measurements were performed every 4 h throughout the growth period until the culture reached the stationary phase. As shown in Fig. 17, the biomass concentration increased steadily and reached its maximum after approximately five days of incubation at 37 °C, after which growth stabilized. This observation was used to determine the optimal harvesting time for transferring the actively growing culture to the secondary medium.

Fig. 17. The growth curve of the bacterium Haloferax mediterranei.

Insights from micromodel experiments: evaluation of microbial recovery efficiency

In this study, the glass micromodel experiments were conducted to investigate the efficacy of MEOR under varying biomass concentrations. Initially, distilled water was injected into the micromodel at a flow rate of 0.009 mL/min to simulate the first stage of water flooding. During this stage, oil recovery occurred only in the inlet, outlet, and near the fractures, as the water failed to penetrate the matrix due to its preferential flow through the fractures. The recovery rates during this phase were recorded as 23% for biomass concentrations of 5.07 g/L and 6.74 g/L, and 20.7% for 10.14 g/L. Once oil production ceased, microbial solutions with biomass concentrations of 5.07, 6.74, and 10.14 g/L were injected at the same flow rate, introducing biofilm-forming microbes into the fracture pathways. The biofilms effectively reduced fracture permeability, redirecting water flow into the matrix and enabling displacement of the trapped oil.

However, the extent of oil recovery was highly dependent on the biomass concentration of the microbial solution. At 5.07 g/L, the biofilm achieved an optimal balance by sufficiently blocking the fractures while maintaining access to the matrix entry points, allowing water to efficiently displace oil and maintain the recovery rate at 23%. At a concentration of 6.74 g/L, while the biofilm increased in density, partial blockage of matrix entry points occurred, leading to a reduced oil recovery rate of 11.7%. At the highest biomass concentration of 10.14 g/L, excessive biofilm formation not only blocked the fractures but also sealed many matrix entry points, thereby reducing oil recovery to 7.8%. This trend highlights the importance of optimizing microbial biomass concentration in MEOR applications, as excessive microbial activity can hinder oil displacement by obstructing flow paths. The recovery curves presented in Figs. 18, 19 and 20 illustrate the performance of microbial flooding at different biomass concentrations, emphasizing the direct correlation between biofilm formation and oil recovery. The recovery values presented in Table 5 were calculated using image analysis performed in ImageJ software. These values represent the percentage of oil recovered relative to the remaining oil at each stage. Stage-1 corresponds to the baseline waterflood (no microbes) and is reported as cumulative % OOIP. Stage-2 values are incremental percentage-point gains relative to OOIP, measured from the end of Stage-1 (additional oil recovered during microbial injection). Each concentration represents a separate micromodel run; minor Stage-1 differences reflect run-to-run variability.

Fig. 18. Oil recovery curves based on dimensionless time in the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 5.07 g/L (B).

Fig. 19. Oil recovery curves as a function of dimensionless time during the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 6.74 g/L (B).

Fig. 20. Oil recovery curves as a function of dimensionless time during the first stage of water injection (A) and the second stage of water injection after injecting a microbial solution with a biomass concentration of 10.14 g/L (B).

Table 5. Stage-1 cumulative recovery and Stage-2 incremental MEOR gains at different microbial concentrations (% OOIP, mean ± SD, n = 3).Concentration (g/L)5.076.7410.14Stage-1 recovery (cumulative, % OOIP)23.0 ± 1.223.0 ± 1.120.7 ± 1.3Stage-2 recovery (incremental gain, percentage points of OOIP)23.0 ± 1.411.7 ± 0.97.8 ± 1.3

Furthermore, Fig. 21 visually demonstrates the effects of biomass concentration on oil displacement in the micromodel, where water (blue) and oil are shown alongside the biofilm presence (colorless regions). At higher biomass concentrations, the visual evidence underscores how biofilm encroachment on matrix entry points significantly impedes oil displacement. During the first stage of water injection, oil recovery was limited to the regions around the fracture, yielding recovery percentages between 20.7% and 23% across all tested concentrations (5.07, 6.74, and 10.14 g/L). However, during the second stage, where microbial solutions were injected, the impact of biomass concentration became evident. While microbial activity effectively reduced fracture permeability and redirected water flow toward the matrix, higher biomass concentrations led to diminished oil recovery. For instance, doubling the biomass concentration from 5.07 g/L to 10.14 g/L caused oil recovery to drop significantly from 23% to 7.8%. Analysis of recorded images revealed that higher biomass concentrations not only enhanced biofilm formation within the fractures but also led to the blockage of matrix entry points. This phenomenon inhibited water from displacing oil within the matrix, resulting in lower recovery rates. These observations emphasize the necessity of optimizing biomass concentrations in MEOR to avoid counterproductive effects on oil displacement. Figure 20 visually depicts the effects of microbial solution injection during the second stage of water flooding. The micromodel, containing water (blue), oil, and microbes (colorless), demonstrates how excessive biofilm accumulation at higher biomass concentrations (10.14 g/L) obstructs both the fracture pathway and the matrix entry points, reducing the effectiveness of water injection in displacing oil^47^. Thus, while microbial flooding proved effective in closing fractures and redirecting water flow to extract matrix oil, the results underscore the critical balance required in MEOR operations. Careful calibration of biomass concentration is essential to optimize oil recovery while avoiding excessive biofilm formation, which can hinder fluid dynamics and lower the overall efficiency of the process. These findings highlight the intricate interplay between microbial activity, fracture permeability, and matrix oil displacement, providing valuable insights for improving MEOR strategies in fractured reservoir systems^48^.

Fig. 21. Micromodel images illustrating the effect of increasing biomass concentration on oil displacement from the matrix space. (A: 5.07 g/L, B: 6.74 g/L, C: 10.14 g/L)

Findings from core flooding experiments: analyzing permeability and recovery factors

Influence of microbial solution on fracture permeability reduction

The permeability of fractured core samples A and B was measured to investigate the impact of microbial injection on sealing fractures and reducing permeability. Microbial solutions were prepared at three biomass concentrations: 5.07 g/L, 6.74 g/L, and 10.14 g/L. The experiments aimed to evaluate the relationship between microbial activity and permeability reduction, simulating conditions relevant to enhanced oil recovery in fractured reservoirs. The permeability of the cores was measured using the relationship k = \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\:\frac{\:m\mu\:L}{A}$$\end{document} , derived from the slope of the flow rate versus pressure drop graphs. The experimental results presented in Tables 6, 7 and 8 clearly demonstrate a reduction in permeability for all biomass concentrations (5.07 g/L, 6.74 g/L, and 10.14 g/L) in both cores (A and B). This reduction becomes more pronounced as the biomass concentration increases. At a concentration of 5.07 g/L, permeability decreased moderately by 56% and 51% for cores A and B, respectively. As the biomass concentration was increased to 6.74 g/L, permeability reductions were more substantial, reaching 63% and 62% for cores A and B, respectively. The maximum reduction was observed at a concentration of 10.14 g/L, with permeability reductions of 76% for core A and 71% for core B. This trend can be attributed to the increased amount of polymer and bacterial cells in higher biomass concentrations, which facilitate biofilm formation and effectively seal the fractures. As the microbial biofilm accumulates, it closes more pores along the fracture path, restricting fluid flow and reducing permeability. This phenomenon demonstrates the potential of microbial injection as an effective method for managing fracture permeability in enhanced oil recovery processes. However, the results also suggest that while higher biomass concentrations are effective in reducing permeability, excessive closure of fractures could impede fluid flow, which is critical for optimal oil recovery. Therefore, an optimal biomass concentration must be determined to balance permeability reduction with the efficient displacement of oil. Future studies should explore the long-term stability of biofilms under varying reservoir conditions, as well as the interactions between microbial activity and reservoir fluids, to further optimize this technique for practical applications. The permeability reduction percentages for each condition provide a quantitative perspective, illustrating the efficiency of microbial solutions in altering fracture characteristics to enhance oil recovery. This analysis highlights the practical implications of using microbial-enhanced oil recovery techniques in fractured reservoirs.

Table 6. Reduction in permeability after microbe injection (5.07 g/L, mean ± SD, n = 3).Core nameNormal state permeability (md)Fractured state permeability (md)Permeability after microbe injection (md)Permeability reduction percentage (%)A29 ± 1.0170 ± 683 ± 456 ± 3B30 ± 1.2240 ± 8109 ± 551 ± 2

Table 7. Reduction in permeability after microbe injection (6.74 g/L, mean ± SD, n = 3).Core nameNormal state permeability (md)Fractured state permeability (md)Permeability after microbe injection (md)Permeability reduction percentage (%)A29 ± 1.0170 ± 663 ± 363 ± 3B30 ± 1.2240 ± 894 ± 462 ± 2

Table 8. Reduction in permeability after microbe injection (10.14 g/L, mean ± SD, n = 3).Core nameNormal state permeability (md)Fractured state permeability (md)Permeability after microbe injection (md)Permeability reduction percentage (%)A29 ± 1.0170 ± 640 ± 376 ± 3B30 ± 1.2240 ± 874 ± 471 ± 3

Investigating the impact of microbial solution on enhancing oil recovery

The investigation into the effectiveness of microbial flooding for enhancing oil recovery was carried out in three distinct stages: initial water flooding, microbial solution flooding, and a subsequent water flooding stage. This process aimed to evaluate the impact of microbial action on increasing oil displacement within the matrix spaces of fractured cores C and D. In the first stage, water flooding was performed at a controlled flow rate of 0.2 milliliters per minute, injecting water up to three times the pore volume. The water exited directly through the fractures without infiltrating the matrix space, leaving the majority of residual oil untouched. This was primarily due to the high permeability of the fractures, which allowed the injected water to bypass the oil trapped in the matrix, demonstrating the limitations of conventional water flooding in such systems. The second stage involved injecting a microbial solution with a biomass concentration of 5.07 g per liter at the same flow rate of 0.2 milliliters per minute. The injected volume was equivalent to the pore volume of the cores, ensuring complete saturation of the fracture space with the microbial solution. Using an ex-situ method, the microbial solution was introduced at peak bacterial growth, maximizing the production of microbial metabolites such as biosurfactants and biopolymers. These metabolites played a crucial role in reducing fracture permeability by forming biofilms on the fracture surfaces, transitioning the fracture sides to an adhesive state. This redirection of water flow from the fracture to the matrix enabled the displacement of residual oil trapped within the matrix spaces, which was previously inaccessible during water flooding. Following microbial solution flooding, a second stage of water flooding was performed to evaluate the long-term effectiveness of the microbial treatment. As expected, the biofilm formed during microbial flooding reduced fracture permeability, forcing the injected water into the matrix space. This facilitated the displacement of residual oil within the matrix, leading to a notable increase in oil recovery. During this stage, an oil-water emulsion was observed due to the microbial action, and the separated oil was collected using a centrifuge method. The results demonstrated a significant enhancement in oil recovery due to microbial flooding. Core C achieved a final recovery of 14%, while Core D exhibited a recovery of 12.6%. Figures 22 and 23 illustrate the oil recovery curves for cores C and D, respectively. For Core C, water flooding continued at 0.2 milliliters per minute for 240 min, during which a volume equivalent to 2.8 times the pore volume was injected. Similarly, in Core D, water flooding persisted for 250 min, with an injected volume equivalent to four times the pore volume. These findings highlight the effectiveness of microbial flooding in redirecting water flow into the matrix and improving overall oil recovery.

Fig. 22. Oil recovery curve for core C as a function of dimensionless time following microbial and water flooding.

Fig. 23. Oil recovery curve for core D as a function of dimensionless time following microbial and water flooding.

In this study, the oil recovery factor obtained from micromodel experiments (23% at a biomass concentration of 5.07 g L⁻¹) was markedly higher than the values observed in fractured-core flooding tests (12–14%). This discrepancy can be attributed to fundamental differences between the two experimental systems. Micromodels exhibit smooth fracture geometries and relatively homogeneous flow paths, which facilitate uniform biomass growth and PHB deposition, thereby enhancing the efficiency of selective plugging. In contrast, fractured carbonate cores are inherently heterogeneous, characterized by rough fracture surfaces, variable fracture apertures, and complex secondary porosity. These features create preferential flow channels that cannot be uniformly blocked by microbial activity, resulting in lower oil recovery. In addition, matrix–fracture interactions, ion exchange, and the high salinity of the rock environment impose further constraints on microbial transport, adhesion, and PHB accumulation. Consequently, while the micromodel results reflect the maximum potential of H. mediterranei for permeability reduction and oil displacement under controlled conditions, the core-flooding results represent more realistic limitations in fractured reservoirs. Comparison with previous studies further shows that the recovery factors and permeability reductions achieved by H. mediterranei fall within the range reported for conventional bacterial strains. For example, Lin et al. reported a 26.1% increase in recovery using Enterobacter cloacae, and Joshi et al. achieved about 28%, whereas permeability reductions of 20–54% and 45–80% were recorded by Gandler et al., and approximately 67–78% by Huang et al. Although these investigations were conducted mainly in porous rather than fractured media, the quantitative comparison indicates that H. mediterranei can deliver comparable performance. The key distinction, however, is that this archaeon not only produces a robust biopolymer but can also continue to grow and synthesize thermally stable PHB at salinities exceeding 300,000 ppm—conditions under which many common MEOR bacteria such as Enterobacter, Pseudomonas, and Bacillus spp. fail to survive. These combined traits position H. mediterranei as a uniquely promising candidate for application in fractured carbonate reservoirs with high salinity and temperature.

Taken together, our results are most consistent with a selective plugging mechanism. First, the microbial stages produced large permeability reductions (~ 50–75%) in fractured samples, indicating effective blockage at hydraulically dominant pathways. Second, micromodel observations show biofilm bridging across fracture constrictions and at fracture–matrix junctions, a geometry that naturally diverts flow and promotes additional displacement from the matrix. Third, the pattern in recovery is diagnostic: a moderate biomass yields the largest incremental gain in recovery, whereas higher biomass produces diminishing returns, consistent with over-plugging that increasingly restricts fracture–matrix communication. We therefore interpret the recovery enhancement primarily as a consequence of biofilm-mediated selective plugging and controlled flow diversion.

Conclusion

This study demonstrates that microbial solution injection can substantially reduce fracture permeability and enhance oil recovery in fractured systems. Across the tested biomass concentrations, fracture permeability was reduced by approximately 50–75%, confirming the effectiveness of biofilm-driven fracture sealing. At the same time, increasing biomass beyond a moderate level led to diminishing returns in oil recovery, as thicker biofilms increasingly impeded fracture–matrix communication and limited displacement efficiency. Micromodel observations corroborated this mechanism: higher biomass promoted extensive biofilm development that not only bridged fractures but also restricted pathways between fractures and matrix.

Overall, the results underscore the need to optimize biomass concentration to balance fracture sealing with sustained access to the matrix. A moderate biomass level (e.g., 5.07 g/L) provided the most favorable trade-off, delivering strong permeability control together with the largest incremental improvements in recovery. These findings provide the first quantitative benchmarks for H. mediterranei in fractured systems and indicate performance comparable to common MEOR strains, while uniquely maintaining growth and PHB production under high salinity and elevated temperatures, where many conventional MEOR bacteria fail to remain viable or preserve polymer integrity.

From an application standpoint, the results suggest practical guidelines for MEOR design in fractured media: (i) prioritize moderate biomass slugs that ensure effective fracture bridging without choking fracture–matrix pathways; (ii) couple microbial injection with flow-back or staged waterflooding to capitalize on restored connectivity; and (iii) implement operational controls (nutrient dosing, shut-in timing, and brine chemistry) to tune biofilm strength and spatial placement. The demonstrated tolerance of H. mediterranei to harsh conditions broadens the operating envelope for saline and thermally stressed reservoirs, offering opportunities for MEOR where polymer stability or microbial survivability is typically limiting.

Finally, while micromodels capture key pore-scale mechanisms, field deployment will benefit from scale-up validation in cores and pilots that examine long-term stability, injectivity impacts, and selective placement in heterogeneous fracture networks. Future work should quantify biofilm reversibility and mobility control over extended cycles, assess compatibility with reservoir brines and additives, and explore adaptive strategies (e.g., pulsed or zonal injections) that preserve fracture–matrix connectivity while sustaining permeability management. Together, these insights chart a clear path toward robust, tunable MEOR in fractured reservoirs.

Recommendations

• Optimal biomass concentration The study demonstrates that moderate biomass concentrations, such as 5.07 g/L, are most effective for balancing fracture permeability reduction and oil recovery. Higher concentrations, while useful for sealing fractures, can lead to excessive biofilm formation, blocking matrix entry points and reducing oil displacement efficiency. Therefore, careful control of microbial solution concentration is essential to maximize recovery while preventing unnecessary obstruction of flow paths. Future research should focus on determining the optimal biomass concentration for different reservoir conditions and fluid properties.
• Controlled injection methods The results suggest that alternating between microbial flooding and water flooding could enhance oil recovery. Microbial flooding helps reduce fracture permeability and redirects water flow into the matrix, but excessive microbial growth can hinder fluid flow. A controlled injection method, where microbial solutions are injected in phases with adequate rest periods for biofilm stabilization, may be more effective in maximizing recovery without blocking vital pathways.
• Long-term monitoring and stability of biofilms Future studies should explore the long-term stability of microbial biofilms under reservoir conditions, including factors such as temperature, salinity, and pressure. Monitoring the persistence and effectiveness of biofilms in sealing fractures over extended periods can provide valuable insights into the sustainability of microbial flooding techniques. This would help in understanding how long the microbial solution can maintain fracture closure and enhance oil recovery.
• Microbial solution formulation and delivery The composition of the microbial solution plays a crucial role in its efficiency. Research should focus on optimizing the microbial strains used, along with their metabolites (e.g., biosurfactants, biopolymers), to ensure maximum biofilm formation without excessive clogging. Additionally, the development of advanced delivery systems for microbial solutions, such as controlled-release technologies or targeted injection methods, could improve the effectiveness of MEOR in heterogeneous reservoirs.
• Combination with other EOR techniques Microbial flooding could be further optimized by integrating it with other EOR methods such as surfactant injection or gas flooding. Hybrid EOR techniques that combine microbial solutions with traditional methods may provide synergistic effects, improving oil displacement and recovery rates. Research into these combined approaches could lead to more robust and cost-effective solutions for oil recovery in challenging reservoir environments.
• Field-scale application and pilot testing While the laboratory results are promising, the scalability of microbial enhanced oil recovery techniques must be validated through field-scale applications and pilot testing. It is crucial to test microbial flooding methods in real-world reservoir conditions to assess their performance under varying temperature, pressure, and fluid chemistry. Pilot studies will help refine microbial injection protocols, optimize biomass concentrations, and validate the economic feasibility of MEOR for large-scale oil production.