Importance of Clay Swelling on the Efficacy of Cyclic Steam Stimulation in the East Moldabek Formation in Kazakhstan

Abstract

Both steam and hot water flooding of high-viscosity oils in the presence of swelling clays are difficult methods for producing oil efficiently because of potential formation permeability reduction. This paper pertains to heavy oil recovery from the East Moldabek formation where the oil API gravity is about 22 and is inundated with swelling clays. To achieve this, we used the Intersect^TM reservoir simulator to compare oil recovery economics using both hot water and steam injection as a function of steam cycle duration, temperature, and steam dryness. We also studied clay swelling in the East Moldabek formation where clay poses a significant challenge due to its impact on permeability reduction. In this research, we developed an equation based on experimental data to establish a relationship between water mineralization and permeability in the East Moldabek formation. The equation provides valuable insight on how to mitigate clay swelling which is crucial for enhancing oil recovery efficiency—especially in sandstone reservoirs. Our modeling studies provide the recovery efficiencies for salinities of the hot water EOR versus cyclic steam EOR methods in a formation containing swelling clays. Specifically, the reduction in formation permeability as a function of the distilled water fraction is the controlling parameter in hot water or steam flooding—when the formation water mixture becomes less saline, oil recovery decreases. Our research shows that clay swelling can significantly impact cyclic steam stimulation outcomes, potentially reducing its effectiveness, while hot water flooding may offer a more cost-effective and operationally feasible solution in formations where clay swelling is a concern. Economic analysis reveals the potential for achieving an optimal favorable condition for hot water injection. Therefore, this paper provides a guideline on how to conduct thermal oil recovery for heavy oils in fields with high clay content such as the East Moldabek deposit.

1. Introduction

Fossil fuels make up over 80% of the global energy supply, with oil, natural gas, and coal accounting for 31%, 27%, and 23%, respectively [1]. Heavy oil, comprising a significant portion of global oil reserves, can help meet the rising energy demand [2,3]. Oil recovery from heavy oil reservoirs, characterized by low API gravity and high viscosity, is less than 20% due to low mobility [4,5]. Successful extraction of these resources requires the application of thermal enhanced oil recovery (thermal EOR) methods, primarily hot water injection (HWI), steam flooding (SF), and cyclic steam stimulation (CSS), to reduce oil viscosity [6,7].

Over the past four decades, thermal EOR methods have been widely employed in the oil and gas industry to enhance hydrocarbon recovery by mobilizing oil through the injection of heat via hot water or steam [8,9,10,11]. The primary focus of these methods has been on the thermal energy received by the reservoir, with little consideration given to the impact of injected water on the reservoir rock itself.

However, significant attention has been directed towards understanding the interactions between injected water/steam and reservoir rock. In some cases, the incompatibility of the injected fluid and rock/fluid in the reservoir results in formation damage, such as the migration of fine particles within the porous media and clay swelling.

Hence, the success of steam injection strongly depends on the quality of the water used to generate steam, as poor water quality can lead to formation damage [12]. To mitigate this risk, specific criteria are recommended for water quality, including total hardness less than 0.01 mg/L, oil content less than 0.5 mg/L, total suspended solids less than 0.5 mg/L, total iron concentration less than 0.5 mg/L, oxygen content less than 0.02 mg/L, and silica concentration less than 200 mg/L. By adhering to these guidelines, the effectiveness of steam-based EOR methods can be enhanced, reducing the potential for reservoir damage and improving overall recovery efficiency [13,14,15].

The salinity of the injected fluid is a critical determinant of the interactions between rock, oil, and brine during thermal Enhanced Oil Recovery (EOR). Variations in salinity influence the electrostatic forces between the rock matrix and fine particles, as well as the structural integrity of clay minerals within the reservoir. In scenarios where the injected steam or hot water exhibits low salinity, there is an increased risk of formation damage, particularly due to clay swelling. Consequently, a thorough understanding of water chemistry and rock/fluid interactions is essential. Over the past two decades, research has increasingly focused on optimizing water injection strategies by modulating salinity levels to enhance oil recovery. Empirical studies have demonstrated that precise control of the salinity in injected water can significantly improve the efficiency of reservoir flooding, while concurrently reducing operational costs and enhancing oil production. This advancing comprehension of the influence of water chemistry on thermal EOR processes marks a pivotal development in the field, providing new opportunities for maximizing recovery in complex reservoir environments [16,17,18,19].

Clay swelling, a possible formation damage mechanism, is a well-documented phenomenon that poses significant challenges during thermal EOR methods, particularly during SF and CSS [20,21]. Studies have shown that the injection of fluids, particularly freshwater used for the steam generator, can trigger the swelling of clay minerals present in oil-bearing sandstones [22,23]. This swelling leads to a reduction in reservoir permeability, hindering the flow of hydrocarbons and impeding oil recovery efforts [24,25]. Various research efforts have focused on elucidating the mechanisms of clay swelling and developing strategies to mitigate its detrimental effects on reservoir performance. Understanding this relationship highlights the importance of selecting injection fluids that minimize formation damage and enhance hydrocarbon recovery efficiency [26].

Yang et al. (2024) conducted a comprehensive review of significant research findings from the past half-century, focusing on the mechanisms by which thermal fluids impact the petrophysical properties of reservoir rocks. The study systematically classified and presented the various factors influencing these mechanisms, providing a detailed understanding of the underlying processes. Additionally, the study critically examined the current state of knowledge, highlighting both the advances and the limitations in the existing body of work. This review aims to contribute to the ongoing development of enhanced oil recovery techniques by shedding light on the complex interactions between thermal fluids and reservoir properties [27].

Naghizadeh et al. (2023) and Kudrashou et al. (2020) also studied formation damage in thermal-enhanced oil recovery processes, providing a crucial foundation for understanding the types of formation damage caused by thermal operations in subsurface reservoirs. Naghizadeh et al. (2023) underscored the importance of recognizing the potential impacts of such damage, aiming to enhance the awareness of reservoir engineers, geologists, and geophysicists [24]. Kudrashou et al. (2020) synthesized recent experimental findings, including steam treatment studies on rock mixtures with varying mineral compositions. Based on the findings, significant changes in rock permeability were observed across various rock types. In calcite-cemented sandstone, permeability decreased by 5%, primarily due to silica dissolution. Sandy samples containing quartz, kaolinite, and carbonates exhibited a more substantial reduction in permeability, ranging from 11% to 22%. This decline was linked to the migration of kaolinite particles and the formation of swelling smectite clay. The most pronounced reduction in permeability, ranging from 77% to 84%, was observed in mixtures of quartz and montmorillonite, which was attributed to the intense swelling of the clay. These findings highlight the critical importance of monitoring mineral composition and rock properties to mitigate risks associated with permeability alterations during hydrocarbon extraction [9].

Temperature is another critical factor influencing these processes, especially in thermal Enhanced Oil Recovery (EOR) methods. Research by Musharova et al. (2012) and You et al. (2019) demonstrated that temperature affects zeta potential and promotes various mineral reactions ([28,29]). Specifically, Musharova et al. (2012) noted that permeability decreases more rapidly at higher temperatures. Similarly, Wang et al. (2021) found that injecting low-salinity fluid into the sandstone at elevated temperatures resulted in higher concentrations of calcium and magnesium in the effluent, indicating that mineral dissolution becomes the dominant mechanism, which may reduce the effectiveness of sweep enhancement due to permeability damage [30]. However, their study primarily considered dolomite dissolution and did not account for other mineral reactions such as ion exchange. Rosenbrand et al. (2015) also reported more severe permeability decreases in sandstone at higher temperatures [31].

Real-world examples from oilfields worldwide provide insights into the detrimental effects of clay swelling on reservoir permeability and oil recovery efficiency. For instance, the Kern River Field in California experienced decreased permeability due to clay swelling, leading to diminished oil recovery [32]. Similarly, operations in the Wilmington Field in California encountered challenges attributed to clay swelling [33], necessitating adjustments in operational strategies to mitigate formation damage [34]. After water injection, clay mineral content in the low-permeability sandstone reservoir of the oilfield decreases by almost 50%. Kaolinite content decreases by 32.75% due to its large grain size hindering passage through pore-throat channels. This migration of kaolinites obstructs pore-throats, leading to formation damage, particularly in argillaceous siltstones and siltstones. While water injection causes slight changes in porosity, permeability undergoes significant reduction [23,35]. These case studies underscore the necessity of implementing effective reservoir management practices and operational strategies to mitigate formation damage caused by clay swelling and optimize the success of thermal EOR projects.

The East Moldabek field in Kazakhstan faces similar challenges, with its geological reserves concentrated in the Cretaceous deposit. This deposit is characterized by heavy oil with high viscosity and low extraction efficiency, resulting in a reduced recovery factor. Thermal methods, especially steam injection, have been identified as effective methods for increasing production in such fields; however, studies have revealed that the presence of clay minerals in the reservoir causes swelling when low-mineralized water interacts with clay-rich rocks, thereby reducing permeability [36]. Thus, further examination and the analysis of steam injection efficiency at the Moldabek field are warranted.

In this study, the effect of clay swelling on the performance of different thermal EOR methods at the field scale was studied for the Moldabek field. Based on extensive laboratory analysis, a model was developed to correlate permeability reduction at different field locations to parameters such as water salinity and temperature. Through this novel approach, it is possible to evaluate and select the best thermal EOR method in such challenging reservoirs. A comprehensive simulation and an economic analysis were also developed to compare thermal EOR options in this field.

Traditionally, clay swelling has been an overlooked factor in the evaluation of thermal EOR methods, leading to potential inaccuracies in predicting reservoir performance. This study introduces a novel model that accurately simulates clay swelling and its influence on fluid flow and oil recovery during thermal EOR processes. By incorporating the swelling behavior of clays into the simulation, this model provides a more realistic and detailed understanding of reservoir dynamics, particularly in high-clay-content formations.

2. Materials and Methods

2.1. Background

The study area comprises two primary geological complexes: both a salt-bearing and a supra-salt complex. The salt-bearing complex, formed during the Early Permian period, is dominated by a salt dome composed of rock salt and anhydrite. The supra-salt complex, which lies above, consists of Triassic, Jurassic, and Cretaceous formations, unconformably overlain by Neogene and Quaternary sediments. Stratigraphic delineation is based on lithological and paleontological data, providing a detailed understanding of the subsurface architecture.

The Cretaceous deposits, which contain the productive oil-bearing horizons, are composed of sands, clays, and marls. This geological structure introduces specific challenges for heavy oil recovery, particularly in the supra-salt formations. The presence of clay minerals and their interaction with injected fluids, along with variable reservoir permeability, significantly affects recovery efficiency. Additionally, the high viscosity of the oil further complicates extraction, particularly when using thermal enhanced oil recovery (EOR) techniques. Understanding these geological factors is essential for optimizing recovery strategies and improving production performance.

2.2. Experiments

A comprehensive simulation was developed to assess the impact of clay swelling on thermal EOR, focusing on key operational parameters. Accurate reservoir modeling requires detailed measurements of rock and fluid properties, such as how oil viscosity changes with temperature, and the effect of injected brine on clay minerals [37].

The experiments aimed to collect data on the composition and behavior of rocks and fluids, including thermal properties, to simulate heavy oil reservoirs. Samples from the East Moldabek field were used, which provided the basis for developing models reflecting the complexities of oil extraction in this environment.

2.2.1. Oil

This study focuses on the comprehensive analysis of the M-I object within the Moldabek oil field, with specific emphasis on three distinguished horizons labeled as A, B, and V. The reservoir oil is characterized by low sulfur, low resin, and low paraffin content. It is classified as heavy and bituminous.

In-depth examination of the oil characteristics within this reservoir reveals density variations within the range of 858.9 to 899.9 kg/m³ under reservoir conditions at a temperature of 23 °C. At the surface level, the density fluctuates between 909.3 and 926.4 kg/m³, with an average of 918.2 kg/m³. Furthermore, the dynamic viscosity of the oil ranges from 242.0 to 443.2 mPa·s, with an average of 377.6 mPa·s. The kinematic viscosity at 20 °C is 783.2 mm²/s, while at 50 °C, it is 107.4 mm²/s. Under reservoir conditions, the oil’s volume coefficient has been determined to be 1.0235. The sulfur content is 0.39%, and the paraffin content is 0.995%. Oil saturation pressure is 1.29 MPa, and the gas/oil ratio was measured to be 5.2 m³/t. The oil is classified as low sulfur.

Experimental temperature–viscosity profile measurement was conducted across a temperature range from 20 °C to 100 °C, as shown in Figure 1.

Using an analytical method, we derived a mathematical relationship from the viscosity/temperature experimental data, as shown in Equation (1):

µ0 (T) = 3260.2 × T^−2.374,

(1)

Therefore, experimental temperature–viscosity profile measurements from 20 °C to 100 °C allowed us to derive a mathematical relationship, extending the dependence value up to 300 °C (Figure 2) for both reservoir (live) and degassed (dead) oil for comparison purposes. The red dots in the figure represent data obtained from the laboratory experiments.

Since the degassed oil data were not significantly different from the live oil data, only reservoir oil data were used for the modeling.

Crude oil’s thermal conductivity (λ₀) is equal to 11.95 KJ/(m·d·K).

2.2.2. Steam and Formation Brine

The groundwater of the East Moldabek deposit is classified as moderately saline. The salinity of the water ranged from 87.928 to 125.965 mg/L. The total hardness ranged from 55 to 245 mg-eq/L, indicating it to be very hard water. The obtained pH values ranged from 6.65 to 6.92, characterizing the water samples as slightly acidic. According to the Sulin classification, the water type is chloride–calcium.

Table 1. Physicochemical properties and compositions of the formation water.

SN	Parameter	Unit	Sample 1	Sample 2
1	Hydrocarbon ion HCO3−	mg/L	220	207
2	Chloride ion Cl−	mg/L	850.939	85.939
3	Sulfate ion SO42−	mg/L	33	32
4	(Sodium + potassium) ion Na+ + K+	mg/L	50.336	50.331
5	Calcium ion Ca2+	mg/L	2.705	2.505
6	Magnesium ion Mg2+	mg/L	1.277	1.398
7	Total mineralization	mg/L	140.510	140.412
8	Oxidized iron Fe3+	mg/L	0.98	3.36
9	Reduced iron Fe2+	mg/L	16.66	15.68
10	Total hardness	mg/L	240.00	240.00
11	Density at 20 °C	mg/L	1.0972	1.0970
12	Salinity	mg/L	13.00	13.00
13	pH	pH units	6.65	6.58
14	Water type by Sulin		chloride–calcium
15	Dissolved oxygen	mg/L	1.13	4.33

shows the physicochemical properties and compositions of the formation water.

Water’s thermal conductivity (λw), which is a function of its salinity, is equal to 52.1 KJ/(m·d·K).

2.2.3. Rock

Oil-bearing sandstones are crucial reservoirs in the petroleum industry, known for their complex composition and dynamic behavior. Among the myriad components found within these formations, clay minerals play a significant role in shaping the properties and behavior of the rock matrix [38]. Understanding the interaction between clay minerals and reservoir fluids and reservoir rock is essential for predicting reservoir performance and optimizing hydrocarbon recovery strategies [39].

Clay minerals are ubiquitous in oil-bearing sandstones, serving as key constituents that influence reservoir permeability and stability [24,25]. However, their presence introduces complexities, particularly regarding their response to external fluid such as injected water. The process of water flooding, especially when the water composition is different from the formation brine, may lead to swelling and subsequent destruction of reservoir clays [22,23]. This phenomenon directly impacts reservoir permeability, potentially impeding the flow of hydrocarbons through the rock matrix.

The physicochemical interaction between clay minerals and water is multifaceted, dependent on various factors including the type of clay minerals present, the composition of absorbed bases, and the chemical composition of the water itself. Different clay minerals exhibit distinct behaviors in response to hydration, forming a hierarchy based on their ability to absorb water. This hierarchy, with the montmorillonite group exhibiting the highest hydration potential followed by hydrosilicates and kaolinite, underscores the complexity of clay–water interactions and their implications for reservoir behavior [35,39].

Table 2. X-ray structural analysis results.

Mineralogical Composition, %
No	Quartz	Illite	Kaolinite
1	21.8	78.0	-
2	30.6	69.0	-
3	41.6	58.4	-
4	42.3	58.0	-
5	54.9	45.0	-
6	66.0	34.0	-
7	15.8	-	84.0
8	18.6	-	81.0
9	23.8	-	76.0
10	40.2	-	60.0
11	51.1	-	49.1
12	51.0	-	49.0

presents the average content of various clay minerals in the Eastern Moldabek deposit, measured by X-ray mineralogical analysis, shedding light on the mineral composition of the reservoir.

X-ray diffraction (XRD) studies of the Eastern Moldabek deposit indicate the presence of mixed-layer minerals (illite, kaolinite, plagioclase), with swelling layers constituting 40–50%. The presence of these clay minerals underlines the importance of studying clay swelling in the Eastern Moldabek reservoir.

Rock thermal properties are required for calculating thermophysical processes in the reservoir. Samples from the field were analyzed resulting in measures for heat capacity and thermal conductivity. Average heat capacity (c) was equal to 1950 KJ/(m³·K), and thermal conductivity equaled 116.64 KJ/(m·d·K). The thermal conductivity of formations is dependent on temperature, pressure, porosity, composition, and properties of liquids and gases. For poorly consolidated rocks, such as the field under study, increasing the stress substantially increases the thermal conductivity.

2.2.4. Core Flooding to Evaluate Clay Swelling

Displacement tests were conducted using Eastern Moldabek sandstone cores. The experiment involved approximately 8 samples, all of which were obtained from a single large fragment of exposure aligned with the bedding planes. The cores were standardized to identical dimensions: a diameter of 3.81 cm and a length of 6.31 cm. Before each test, the samples were dried for 48 h in an oven set to 80 °C. This temperature was selected to ensure complete dehydration without reaching levels that might deactivate clay minerals. A fresh core sample was used for each test.

The experimental setup used for filtration tests is depicted in Figure 3 [40]. The flow system was composed of a network of pipes and valves, through which oil and water were injected into the core sample. The experiments utilized two accumulator multipliers, each consisting of a piston and cylinder, containing injection fluids such as formation water and distilled water.

The core holder employed was a Hassler sleeve type, situated inside the oven. The main cylinder had a length of about 10 inches and a diameter of 4 inches, accommodating a Hassler sleeve measuring 5 inches in length and 2.25 inches in width. This sleeve could securely hold a core sample up to 4 inches long and 1.5 inches in diameter. The heating system included an oven, a temperature controller, a thermometer, and a fan for air circulation.

A back pressure system managed the fluid flow through the core sample, enabling control of the output fluid flow rate. A sensor system tracked drops in pressure across the core, recording the differences between the inlet and outlet flows. These measurements were displayed on a digital sensor indicator and were also transmitted to a recorder to maintain a continuous archive of the data.

The porosity of cores was determined using a helium porosimeter. Also, the permeabilities were determined using an air permeameter. To determine the effect of clay swelling on permeability, a series of flooding tests was conducted. Initially, the formation brine was injected, followed by diluted brines. Diluted samples were prepared by mixing the formation brine with the freshwater in proportions of 75–25%, 50–50%, and 25–75%. At the final stage, distilled water was pumped. Permeability was measured at each stage to evaluate the extent of clay swelling and its effect on the reduction in permeability.

To evaluate the changes in core samples after the experiments, scanning electron microscopy (SEM) was conducted. Small sections of each rock sample were carefully dried and coated with a thin layer of a gold–palladium alloy to enhance conductivity for SEM imaging. Figure 4 shows an example of the clays which are presented in the rock structure.

The SEM images in Figure 4 provide detailed insights into the microstructural characteristics of the clay minerals. Specifically, Figure 4a shows kaolinite (E6) at a magnification of 9500×, while Figure 4b presents it at an even higher magnification of 10,000×, allowing for a closer examination of its structural details. Additionally, Figure 4c displays the smectite–illite cement at a magnification of 1300×, and Figure 4d captures the same material at 4500×.

The SEM analysis revealed that the clay material in the rock samples primarily consists of mixed-layer minerals such as illite–smectite and smectite–illite appearing as irregular scaly plates, while kaolinite minerals with a layered structure were rarely detected and only in trace amounts under high magnification.

Table 3. Elemental composition of the clay material in the rock samples based on the SEM.

Elemental Composition, %
Spectrum No.	O	C	Al	Si	K	Na	Fe	Ti	Ca	Mg	Cl	Total
1	54.2	20.8	4.3	9.2	5.0	2.7	3.7	-	-	0.1	-	100
2	51.8	4.2	4.3	13.5	7.4	3.1	9.9	1.2	2.7	0.1	1.7	100
3	52.2	6.0	3.5	18.3	8.2	4.2	5.4	-	2.0	0.2	-	100

summarizes the elemental composition of the clays present in the rock.

X-ray diffraction (XRD) analysis was also conducted on the core samples, targeting both the bulk material and the fine fractions, which were obtained by crushing the samples and sieving them through a 230-mesh sieve to evaluate the mineral composition of the core samples. The samples were collected from wells at different locations of the East Moldabek formation. Examples of the XRD results are shown in Figure 5,

Table 4. XRD analysis results of the samples.

Mineralogical Composition, %
	Quartz	Albite	Paragonite	Plagioclase	Illite	Kaolinite
Sample 1	36.0	13.1	1.9	48.9	-	-
Sample 2	21.8	-	-	-	78.0	-
Sample 3	41.6	-	-	-	58.4	-

.

The XRD analysis reveals the following dominant minerals: plagioclase in Sample 1 (48.9%), and illite in Samples 2 (78.0%) and 3 (58.4%). Quartz is present in all samples. The high illite content, especially in Samples 2 and 3, suggests significant potential for clay swelling in the formation. These findings highlight the varied mineralogical compositions across different locations in the East Moldabek formation.

3. Simulation Study

3.1. Background

Simulation studies are essential tools for modeling complex real-world systems, processes, or phenomena. They offer researchers a systematic approach to investigating systems where real-world experimentation may be impractical due to logistical, financial, or physical constraints [41]. This study uses Intersect™ software version 2020.4 to model thermal recovery processes, particularly effective for handling complex reservoir conditions such as those encountered in thermal recovery scenarios.

The simulation focuses on the East Moldabek formation, utilizing a single-porosity model to represent fluid flow and heat transfer within the porous medium. This methodology integrates crucial factors, including fluid phase behavior, phase equilibrium, energy conservation, and component balance, offering a comprehensive framework for analyzing reservoir dynamics under thermal recovery conditions. It is based on the fundamental principles of mass, momentum, and energy conservation, alongside Darcy’s law for fluid flow in porous media.

The key equations employed in the simulation include the component balance, phase equilibrium, and energy conservation:

• Component Balance

The component balance governs the rate of change for each component in the simulation grid, accounting for intercell flows and external sources, such as wells or aquifers. For each component, the balance is expressed by Equation (2) [42]:

$${\displaystyle \frac{\partial M_i^j}{\partial t}}+\Sigma F_i^{\left\{jk\right\}}+\Sigma Q_i^{\left\{jw\right\}}=0,$$

(2)

where the following apply:

${\displaystyle \frac{\partial M_i^j}{\partial t}}$—the rate of change of the number of moles of component i in cell j.

$\Sigma F_i^{\left\{jk\right\}}$—the sum of inter-cell flows of component i into cell j from connected cells k.

$\Sigma Q_i^{\left\{jw\right\}}$—the sum of flows of component i into cell j from external sources (e.g., wells, aquifers).

2.

Phase Equilibrium

Phase equilibrium ensures that the fugacity (chemical potential) of each component remains consistent across phases (oil, gas, water) in thermodynamic equilibrium.

3.

Energy Conservation (Thermal Balance)

Energy conservation is critical for modeling heat transfer during thermal recovery processes like steam stimulation and hot water injection. The energy balance equation for thermal simulations can be written as in Equation (3) [43]:

$$\begin{gathered} {\displaystyle \frac{\partial }{\partial t}}\left[\rho \left(e+{\displaystyle \frac{V^2}{2}}\right)\right]+\nabla \cdot \left[\rho \left(e+{\displaystyle \frac{V^2}{2}}\overrightarrow{V}\right)\right] \\ =\rho \dot{q}+{\displaystyle \frac{\partial }{\partial x}}\left(k{\displaystyle \frac{\partial T}{\partial x}}\right)+{\displaystyle \frac{\partial }{\partial y}}\left(k{\displaystyle \frac{\partial T}{\partial y}}\right)+{\displaystyle \frac{\partial }{\partial z}}\left(k{\displaystyle \frac{\partial T}{\partial z}}\right)-{\displaystyle \frac{\partial \left(up\right)}{\partial x}}-{\displaystyle \frac{\partial \left(vp\right)}{\partial y}} \\ -{\displaystyle \frac{\partial \left(wp\right)}{\partial z}}+{\displaystyle \frac{\partial \left(u{\tau }_{xx}\right)}{\partial x}}+{\displaystyle \frac{\partial \left(u{\tau }_{xy}\right)}{\partial y}}+{\displaystyle \frac{\partial \left(u{\tau }_{xz}\right)}{\partial z}}+{\displaystyle \frac{\partial \left(v{\tau }_{yx}\right)}{\partial x}}+{\displaystyle \frac{\partial \left(v{\tau }_{yy}\right)}{\partial y}} \\ +{\displaystyle \frac{\partial \left(v{\tau }_{zy}\right)}{\partial z}}+{\displaystyle \frac{\partial \left(w{\tau }_{xz}\right)}{\partial x}}+{\displaystyle \frac{\partial \left(w{\tau }_{yz}\right)}{\partial y}}+{\displaystyle \frac{\partial \left(w{\tau }_{zz}\right)}{\partial z}}+\rho \overrightarrow{f}\cdot \overrightarrow{V}, \end{gathered}$$

(3)

where the following apply:

${\displaystyle \frac{\partial }{\partial t}}\left[\rho \left(e+{\displaystyle \frac{V^2}{2}}\right)\right]$ represents the rate of change of the total energy of the fluid with respect to time $t$.

$\rho$ is the fluid density, kg/m³.

$e$ is the internal energy per unit mass, m²/s².

$\nabla \cdot$ is the divergence operator indicating how the energy flux changes within the control volume.

$V$ is the velocity of the fluid, m/s.

${\displaystyle \frac{\partial }{\partial x}}\left(k{\displaystyle \frac{\partial T}{\partial x}}\right)$ is the conduction of heat within the fluid in the x direction.

$k$ is the thermal conductivity, J/s·m·K.

$T$ is the temperature, K.

$p$ is the pressure, N/m².

${\displaystyle \frac{\partial \left(u{\tau }_{xx}\right)}{\partial x}}$ is the term indicating how the x-direction viscous stress affects the flow’s energy.

The governing equations are discretized using a fully implicit time stepping scheme. At each step, based on the temperature and salinity, the permeability at each cell is updated by the model we developed to show the effect of clay swelling.

3.2. Simulation Methodology

The simulation methodology involves the development of a 3D sector model for the East Moldabek reservoir. This model consists of 3,650,361 total cells, with 2,476,283 active cells, enabling a detailed resolution of fluid flow within the reservoir. The model’s key parameters are outlined in

Table 5. Fundamental parameters of sector model.

Parameter	Model
Size	143 × 127 × 201
Total Cells	3,650,361
Active Cells	2,476,283
Cell Size, m	30 × 30 × 0.5 Model
Type	Thermal
Initial Oil, m3	12 million

.

The simulation model incorporates a Pressure–Volume–Temperature (PVT) framework to predict fluid behavior under various reservoir conditions. K-values, which define the relationship between fluid phases and temperature, were added through field and laboratory data tables. The thermal compositional model was employed to capture temperature-dependent fluid properties [44].

Initial and boundary conditions were established based on three distinct geological horizons, capturing the stratigraphic variability in the East Moldabek formation. The fully implicit time discretization scheme was chosen to enhance both the stability and convergence of the simulation, particularly under complex thermal recovery scenarios.

Verification and validation were essential to ensure the reliability of the model. History matching was conducted to calibrate the simulation against field data, ensuring consistency between observed and predicted results. Additionally, benchmark comparisons were performed using laboratory data and field measurements to further validate the outputs.

The simulation was executed using Intersect™ software, which is well regarded for its thermal modeling capabilities. Furthermore, custom Python scripts were utilized for pre-processing, post-processing, and to implement the effect of clay swelling on permeability changes, enhancing the accuracy of the simulation.

3.3. Simulation Cases

To achieve the predefined objectives using the simulation model, a systematic approach was adopted, encompassing the following steps:

• Model Initialization: The simulation model was initialized to establish the baseline conditions for subsequent analyses.
• History Matching and Quality Control: Calibration of the model parameters was performed through history matching to ensure consistency between simulated and observed data.
• Sensitivity Analysis of Basic Thermal Parameters: Sensitivity analysis was conducted on key thermal parameters, including injection period, production period, soak period, temperature, and steam quality, to assess their impact on reservoir performance.
• Simulation Cases Execution: Multiple simulation cases were executed to evaluate the effectiveness of different injection strategies, including CSS and HWI.
• Evaluation of Economic Effects: The economic viability of both cyclic steam injection and hot water injection techniques was assessed to determine their potential for enhancing oil recovery while considering operational costs and benefits.

For CSS cases, thorough sensitivity analysis was crucial in determining optimal operational parameters. Different scenarios were formulated and calculated to identify the effects of the injection period, period of production, soak period, and steam quality on the steam cyclic injection performance. Subsequently, the permeability changes dependency function, developed by our experimental data, was considered in the simulation.

During the injection period, parameters such as maximum bottomhole pressure (80 bar), injection rate (30 m³/d), and temperature (250 °C) were determined based on sensitivity analysis. Similarly, during the production period, the bottomhole pressure was set to be maintained at saturation pressure (14.4 bar), production period (39 days), and shut-in period (3 days) determined through sensitivity analysis. An additional economic constraint was implemented for producing wells, wherein perforations would be shut off if the water cut exceeded 98%.

Hot water injection on the studied asset is planned with temperatures ranging from 60 °C to 80 °C. Thus, during hot water injection, the surface pressure will be 8.5 MPa, facilitating the production of high-viscosity oil.

4. Results and Discussion

4.1. History Matching and Model Calibration

The primary objectives of history matching include refining the geological and physical properties of the reservoir model to enable realistic predictions for both Cyclic Steam Stimulation (CSS) and Hot Water Injection (HWI). The numerical model includes specific limitations related to field management and grid structuring. During the historical control phase, production wells were regulated by liquid production rates, while water injectors were managed by water injection rates. Cell activity settings were carefully defined, with a minimum pore volume of 0.0001, a minimum rock volume of 1 × 10⁻⁶, and a pinch Z-value of 0.1. The grid structure consisted of 143 cells in the I-direction, 127 cells in the J-direction, and 201 cells in the K-direction. For field management, bottomhole pressure (BHP) constraints were set for production wells operating at potential, with a BHP of 1 bar for all forecast cases. Injection wells maintained a BHP of 100 bar.

Additionally, the simulation results were validated using core flooding tests conducted in the laboratory. These tests involved detailed core analysis and experimental data, ensuring that the simulation closely matched experimental conditions, thereby improving its accuracy and relevance.

The history matching process was particularly challenging due to the significant viscosity contrast of the fluids. Consequently, variations in water cut significantly influenced the calculated bottomhole pressure. Changes in permeability also alter the amount of mobile water, leading to substantial variations in water cut. The results of the history matching process are presented in Figure 6. History matching was simulated by Intersect^TM software.

The history match profiles indicate a close alignment between the simulated and historical data, suggesting that the model accurately captures the key dynamics and behaviors of the reservoir. This high level of agreement confirms the model’s reliability in predicting future reservoir performance under similar conditions. The concordance between the two datasets implies that the model parameters are well calibrated and that the overall model structure is robust, instilling confidence in the simulation outcomes. Over 80% of the wells exhibit good alignment with the actual production history, as demonstrated in Figure 7.

For model initialization, the transition zones in different wells and the initial water saturation distribution were estimated using log data, as shown in Figure 8.

Saturation endpoints and gas relative permeability were estimated as average statistical values derived from measured data. The critical gas saturation was set at 0.05, while gas relative permeability at residual oil saturation and irreducible water saturation was assumed to be 0.8.

The results of the initial saturation distribution are illustrated in Figure 9.

4.2. Sensitivity Analysis

Sensitivity analysis, by systematically varying input parameters and observing their impact on output variables, provides deeper insights into the behavior of this complex reservoir system. This study conducted a comprehensive sensitivity analysis to evaluate the effectiveness of CSS technology in maximizing oil recovery within the analyzed horizon. Key operational parameters were assessed, including steam dryness, soaking period, injection period, and production period.

The sensitivity analysis for steam dryness considered two forecast scenarios. In the baseline scenario, steam dryness was set at 0.9, representing high-quality steam. In contrast, the second scenario employed a steam dryness value of 0.5. All other parameters, including injection, soaking, and production periods, were kept constant in both cases. The findings indicated that reducing steam dryness from 0.9 to 0.5 resulted in an approximate 10% decrease in cumulative oil production over a 5-year period, suggesting a relatively minor impact of steam dryness on overall results.

For the soaking period, simulations explored various durations ranging from 3 to 7 days. The results indicated that the highest cumulative oil production was achieved when the soaking period was approximately 3 days. In addition, several scenarios were also assessed for the injection period, which varied between 7 and 28 days. The highest cumulative production occurred with an injection period of about 14 days. Finally, the production period was varied from 25 to 123 days, with optimal cumulative production observed at around 39 days. These optimized parameters were subsequently employed to evaluate the effect of clay swelling on CSS performance.

4.3. Experimental Evaluation of Clay Swelling

Laboratory experiments revealed significant changes in permeability due to variations in brine salinity during injection.

Table 6. Effect of water salinity on permeability in laboratory studies (FW: formation water, DW: distilled water).

Sample Model No.	№1	№2	№3	№4	№5	№6	№7	№8
Gas Permeability, mD	518.0	572.0	232.0	2060.0	1350.0	1580.0	3630.0	5050.0
Permeability after injection of FW (mD)	323.2	112.5	49.8	407.1	332.5	396.1	643.7	712.9
Permeability after the first dilution (75% FW-25% DW) (mD)	277.1	109.3	44.0	443.2	190.7	288.1	573.9	719.1
Permeability after the second dilution (50% FW-50% DW) (mD)	228.3	105.2	36.4	396.3	242.3	247.2	581.7	620.7
Permeability after the third dilution (25% FW-75% DW) (mD)	197.0	96.0	27.1	285.4	214.0	207.0	470.4	579.5
Permeability after injection of DW (mD)	11.2	10.2	2.2	8.2	184.9	61.7	485.6	256.7

presents the results of eight flooding tests conducted on core samples from various wells. Marked reductions in permeability were observed following the injection of diluted brines, indicating the occurrence of clay swelling.

Figure 10 illustrates the variations in absolute permeability of rock formations due to the dilution of injected brine, as observed in a series of experiments. The dots in the figure represent data obtained from the laboratory experiments. The y-axis represents the change in rock permeability following the introduction of diluted brine solutions. These permeability changes are plotted against the x-axis, showing the freshwater proportion in the brine mixture. Absolute permeability is defined as the intrinsic ability of a rock to transmit fluids through its interconnected pore spaces, underscoring its critical role in the dynamics of fluid flow within geological formations.

The points on the curve show the experimental data, while the dashed line represents the model matched to the data. The relationship between permeability alteration and brine salinity can be expressed by the following in Equation (4):

$${\displaystyle \frac{∆\kappa }{\kappa }}=-2.16D^3+1.86D^2-0.6D+1.01$$

(4)

where D represents the fraction of distilled water in the brine (dimensionless, ranging from 0 to 1).

The equation provides a fitted curve based on experimental data, quantifying the permeability change (∆κ/κ) as the brine composition shifts from 100% formation water to 100% distilled water. Specifically, D = 1 indicates 100% distilled water, while D = 0 represents the initial brine composition without any dilution.

A static clay swelling test also confirmed the sensitivity of clays in this formation to low salinity and high temperature. Figure 11 shows clays in contact with high salinity and low salinity brines at 100 degrees.

The swelling of clays is obvious at low salinity conditions.

4.4. Selection of Thermal EOR Approach in the Presence of Clay Swelling

In thermal Enhanced Oil Recovery (EOR), various mechanisms contribute to enhancing oil detachment and movement within the porous media. Among these, the reduction in viscosity and alteration of wettability are the most effective mechanisms in this study [45]. With increasing temperature, changes in permeability and residual oil saturation occur, which must be considered for accurate reservoir simulation. Three flooding experiments were conducted to measure residual oil at different temperatures using core samples from the East Moldabek field (Figure 12).

At 20 °C, which is close to the reservoir temperature of 23 °C, the residual oil saturation ranged from 0.36 to 0.41. The average reduction in residual oil saturation, as temperature increased to 130 °C, was approximately 22%. The relationship between residual oil saturation and temperature, as determined from these experiments, was modeled and subsequently applied in the simulation.

Clay swelling, a phenomenon particularly relevant during thermal EOR operations, significantly affects oil production by obstructing pore spaces and diverting fluid flow. In this study, a 10-year forecast analysis of Cyclic Steam Stimulation (CSS) was conducted, explicitly considering the impact of clay swelling. This process is initiated by the absorption of water by water-sensitive clay minerals in the reservoir rock, which leads to their expansion and the subsequent blockage of pores, thereby reducing permeability. Therefore, this process is crucial during CSS, as the injected steam reduces the formation water salinities at various points within the reservoir.

To accurately capture the clay swelling effect in our numerical model, we adopted a meticulous approach. Freshwater sourced from the generator is introduced into the model cell, intricately mingling with the reservoir water, a process quantified by laboratory equations. The resulting mixture induces alterations in permeability, meticulously simulated through Python scripting supported by the simulator. The alteration in permeability is estimated by the developed model from experiments, as explained earlier. Figure 13 visually illustrates an example of the dynamic variations in the salinity of the brine and the permeability of the matrix at each cell in response to the injection of freshwater into the reservoir. This approach was used to simulate the interplay between clay swelling and reservoir behavior during thermal EOR processes.

To assess the impact of clay swelling on oil production, two simulations were run: one accounting for the effects of clay swelling and the other without. In both scenarios, the optimum operational parameters were applied, as discussed in the sensitivity analysis section. These included a steam temperature of 250 °C, a soak period of 3 days, an injection period of 14 days, and a production period of 39 days. Figure 14 presents the additional oil production from CSS under two different scenarios with the left y-axis showing the normalized additional oil production (baseline set to 1 in the first year) and the right y-axis displaying the absolute additional oil production in thousands of tons. The x-axis represents time in years.

The results demonstrate that neglecting the effects of clay swelling results in a significant overestimation of oil production, highlighting the critical importance of accounting for this phenomenon in reservoir simulations.

To assess the financial viability of the project, an economic analysis was conducted based on the evaluated oil production. Figure 15 presents the economic evaluation, including the corresponding Internal Rate of Return (IRR) of the project over the 10-year period for both scenarios.

Economic calculations without considering clay swelling effects exhibit a steady ascent, reaching a noteworthy 50% IRR by the tenth year. Conversely, factoring in clay swelling yields a stark contrast, with a gradual improvement resulting in a 2% IRR by the tenth year. This underscores the substantial impact of clay swelling on project economics, leading to significantly lower additional oil production.

Our studies reveal that the freshwater concentration’s impact on reservoir response, particularly in cases where clay swelling is significant, can greatly alter the effectiveness of the Enhanced Oil Recovery (EOR) strategy. For example, increasing the freshwater ratio up to 75% reduces water permeability by 10% to 40%. However, as more water is injected, permeability continues to decline, suggesting that higher freshwater proportions may consistently reduce permeability, making steam injection a less favorable option.

The technical and economic efficiency of hot water injection, conducted at a temperature of 60 °C with an injection rate of 50 m³/day, was also evaluated as a thermal EOR option. Figure 16 compares oil production over a 10-year period from hot water injection versus CSS, showing both absolute and normalized values relative to the first year. Simulation results indicate that hot water injection produced an additional 16 thousand tons of oil, which is 14% less than the output achieved with CSS. However, the capital investment required for hot water injection is approximately three times lower than for CSS.

The implementation of hot water injection resulted in a steady increase in additional oil production over the forecast period. The economic evaluation revealed that, while revenues gradually increased, expenses were primarily due to operational costs and initial investments. Notably, revenues exceeded expenses in most years, indicating potential profitability. The IRR in Figure 17 for the hot water injection project starts at −62% and improves to 25% by the end of the forecast period, signifying increasing financial attractiveness and enhanced viability over time. Therefore, in this field, hot water flooding technology shows significant promise over CSS, mainly due to its cost-effectiveness. While steam flooding offers the potential for enhanced oil recovery, it faces challenges, such as the need for specialized steam boilers.

Both hot water injection and cyclic steam simulation demonstrate negative IRR values initially.

The initial internal rate of return (IRR) for hot water injection begins at −62%, whereas cyclic steam stimulation (CSS) starts at a significantly lower IRR of −88%. This contrast underscores the greater initial economic risks associated with CSS in the context of clay swelling effects. However, over time, it demonstrates improvement, though it continues to remain below zero. A significant observation is the transition from negative to positive IRR values: hot water injection reaches economic viability after 3 years, whereas CSS with clay swelling effects takes 9 years to do so. This suggests that hot water injection achieves economic viability significantly earlier than CSS when clay swelling effects are present.

These observations highlight the necessity of considering geological complexities and rock/injected brine interactions in economic evaluations to accurately assess the feasibility and economic viability of reservoir management strategies. It also indicates that CSS is not a viable option in this case. Hot water flooding emerges as a promising alternative, with the produced water being utilized for injection, thereby minimizing the impact on formation water salinity.

5. Conclusions

This study demonstrates the critical role clay swelling plays in the performance of enhanced oil recovery (EOR) methods, particularly in the East Moldabek formation, a site characterized by high-viscosity oil and the presence of swelling clays. Our findings show that clay swelling has a detrimental effect on reservoir permeability and fluid flow, which in turn reduces the effectiveness of cyclic steam stimulation (CSS).

Laboratory experiments provided further evidence of substantial permeability reductions caused by variations in brine salinity. Severe permeability reductions were observed following the injection of increasingly diluted brines, clearly indicating clay swelling. As presented in the experimental evaluation section, permeability consistently declined as the fraction of distilled water increased. For example, permeability dropped from 518 mD to just 11.2 mD after the injection of distilled water in one sample, illustrating the dramatic impact of brine dilution on reservoir permeability.

We found that, although CSS can yield higher total oil recovery, the presence of swelling clays significantly undermines its economic viability. Under the conditions studied, CSS achieved only a 2% internal rate of return (IRR), whereas hot water injection, which mitigates the effects of clay swelling, produced a far superior IRR of 25%. Hence, the results challenge the traditional reliance on steam injection for oil recovery in formations containing swelling clays. Hot water injection not only reduces the negative impact of clay swelling but also offers a more cost-effective and sustainable solution for oil recovery. This insight underscores the need to tailor EOR methods to the specific geological features of each reservoir, especially in formations prone to clay swelling, to optimize recovery and minimize costs.