Experimental Study of the Microscopic Visualization of Gas Clogging during Groundwater Recharge

Abstract

Clogging is one of the most important factors that restricts the development and popularization of artificial groundwater recharge technology. Gas clogging is an important but often overlooked form of clogging. In this study, a high-speed image acquisition system was used to obtain high-resolution images of the migration of water and gas in the pore. The bypass flow, trapped bubbles in the H-shaped pore channel, blind end, and corner of the pore were directly observed and their clogging mechanisms were analyzed. The influences of the pore structure and gas content on the degree of gas clogging were quantified. The pore–throat size has a certain controlling effect on the movement of the gas and liquid phases. As the diameter of the pore–throat increases, the clogging effect of the gas decreases, and the relative permeability of the water (k_rw) increases. The pore–throat ratio exhibits a negative correlation with the relative permeability of the liquid phase, and the pore–throat sorting coefficient exhibits a positive correlation with k_rw. As the gas content increases, the degree of gas clogging increases, and the effect is more significant at low gas-to-liquid ratios (<1:2). These results provide theoretical support for the scientific quantitative evaluation and prediction of the occurrence of gas clogging in groundwater recharge projects.

1. Introduction

With climate change and accelerated urbanization, the human demand for water resources is increasing, which has led to the overexploitation of groundwater resources in many water-limited areas [1,2,3]. The Ministry of Water Resources of the People’s Republic of China carried out an evaluation of groundwater overexploitation areas nationwide during the evaluation period from 2001 to 2010. The results revealed that the current groundwater overexploitation area in the plain area is 287,100 km², and the overexploitation volume is 15.8 billion m³. Groundwater overexploitation is particularly serious in the northern part of China, accounting for 95% of the overexploitation area and 99% of the overexploitation volume in the country. Groundwater overexploitation can cause a series of problems, such as the depletion of groundwater resources, frequent environmental and geological hazards, and ecosystem degradation, which cause serious harm to sustainable economic and social development, people’s health, and the ecological environment. Artificial groundwater recharge is a proven method of mitigating groundwater overexploitation and achieving surplus water storage [4,5,6].

The clogging is frequently due to the presence of a variety of factors, such as the unsatisfactory water quality of the recharge water or the physical and chemical properties of the infiltration medium, resulting in the reduced permeability of the aquifer and, thus, the weakening of the recharge efficiency of the project. Beijing and Suzhou initiated recharge projects in the 1980s, and the percentage of recharge wells with a low recharge efficiency or even the need to be scrapped due to the fact of clogging was 80% and 54% in these areas, respectively, within 20 years [7]. Among the 207 rainwater infiltration systems built in Maryland in 1986, 33% of them were scrapped due to the fat of clogging within two years of operation, and the number of infiltration systems out of service due to the fact of clogging increased to 50% in 1990 [8]. Therefore, clogging is one of the most important factors restricting the development and promotion of artificial recharge technology [9].

There are four principal types of clogging: physical, chemical, biological, and gas. Physical clogging is caused mainly by the suspended solids in recharging water [10]; chemical clogging is caused mainly by chemical precipitation [11]. Bioclogging is due mainly to the presence of clogging problems caused by microorganism growth, production of extracellular polymeric substances (EPSs) and microbial activities [12]. Currently, studies on the effects and mechanisms of physical, chemical, and biological cloggings are relatively abundant and mature [9,13,14,15], but few studies have been conducted on gas clogging. Gas clogging is caused by a large amount of air bubbles and dissolved gases carried in the recharge water, which undergoes changes in pressure and temperature during injection, resulting in the release of the gas. Moreover, when the water flows downward in the well pipe due to the fact of gravity, the water flow is discontinuous and generates a negative pressure, leading to vaporization and, thus, the release of the gases dissolved in the water. There are also a large number of microorganisms in recharge water, which metabolize and produce gases under certain conditions [16,17]. These gases occupy the aquifer pores during transport, which reduces the permeability of the aquifer and decreases the recharge efficiency. Presently, studies on gas clogging are mainly divided into two categories. One is the use of column and tank experiments to observe the effect of gas clogging on the permeability coefficient. The research results show that the gas clogging mainly occurs in the shallow layer of the medium, with a depth of 30 cm, and there are two regions with a large clogging degree within the depth range of 0–30 cm. Gas clogging causes the permeability coefficient to decline exponentially, and it has a downward trend with time [18]. A column containing clay, sand, and gravel was used for an experiment, and it was found that the gas clogging was more obvious when the medium particle size was reduced [19]. The more gas contained in the reinjection water, the faster the gas clogging occurs in the medium, and the more serious the clogging [20]. Both the entrapped air and biogenic gases can occupy 7–20% of the entire pore space [21,22]. Once formed, gas bubbles tend to block the largest pore throats during water saturation, thus leading to lowered permeability and an elevated water level [23]. The other is the use of site tests to observe the effect of gas clogging on the groundwater recharge rate. According to the groundwater reinjection of the Sand Hollow Reservoir in southwestern Utah, the decline of the reinjection rate in 2002–2008 was caused by the occurrence of gas clogging and the reduction of the regional hydraulic gradient, and the reinjection rate increased from 2009 to 2010 due to the dissolution of the gases [24]. Compared with other clogging forms, the research on gas clogging is relatively less. Most previous studies on gas clogging explored some of the macroscopic laws, but there is a lack of intuitive studies on gas clogging from a microscopic perspective, such as the position and shape of bubbles when gas clogging occurs.

In this study, gas clogging in porous media was investigated from a microscopic perspective. The main aims of the current study were as follows: (1) to identify the morphology of bubbles in pores when gas clogging occurs; (2) to analysis of the mechanism of gas clogging from a microscopic perspective; (3) to quantify the degree of gas clogging; (4) to analyze the influence degree and mechanism of the pore structure and gas–liquid ratio on gas clogging. Addressing these aims will provide a theoretical basis for the quantitative evaluation and prevention of gas clogging in groundwater recharge projects.

2. Experimental Apparatus and Procedures

2.1. Experimental Materials

A microscopic visual model can be used to simulate and visualize the two-dimensional flow of water and gas in the process of groundwater recharge. The microscopic visual model was constructed by the Zhenjiang Huarui Chip Technology Co. (Zhengjiang, China). The model was made of special float glass produced in Germany. The type of glass was BOROFLOAT 33, which is a stable material and a particular chemically resistant glass. Its chemical composition is 75–85% silica, 10–20% boron oxide, 1–5% alumina, 1–5% sodium oxide, and 0–2% potassium oxide. Coated glass was used, and after uniformly coating the photoresist, the mask plate was covered to prevent exposure to ultraviolet (UV) light. The light-transmitting part of the protective photoresist layer was removed after development. Hydrofluoric acid with ammonium fluoride buffer solution was used to corrode the protective photoresist layer of the light-transmitting part of the wet method. The corrosion of the glass material was allowed to continue for approximately 2 h, achieving a corrosion depth of 50 μm. Then, the glass was removed, cleaned, heated in a constant temperature oven at 100 °C for 30 min, and etched to between 80 and 90 μm. The adhesive protective layer and coating layer were removed using a degumming cleaning solution. The cleaning agent was a heated solution consisting of sulfuric acid and hydrogen peroxide with a ratio of 9:1. Finally, the glass pieces were placed in a bonding machine to be vacuumed, heated, and bonded under stress. The fabrication process was the same for each microscopic visual model, except that the mask was changed according to the different pore structures.

Homogeneous and nonhomogeneous microscopic models were used in this experiment. The homogeneous models had a particle size of 1 mm and a porosity of 30% and a particle size of 2 mm and a porosity of 40%. The heterogeneous model was based on the real pore structure. The real pore structure was obtained by scanning compacted quartz sand with different particle sizes. This work was conducted in the Key Laboratory of the Ministry of Education for Efficient Mining and Safety of Metal Mines, University of Science and Technology, Beijing, China, using a Sanying nano 3502E high-resolution X-ray 3D scanning system (Sanying Precision Instrument Co., Tianjin, China). Then, the image segmentation process was conducted in Avizo (9.0.10) to classify the pores and rock particles and draw the model masks. The heterogeneous models had particle sizes of 1–2, 2–3, 3–4, and 5–8 mm.

A model based on computed tomography (CT) scanning and microelectronic lithography has advantages compared with other traditional visualization models: (1) the sealing of the microscopic model is enhanced by the improved bonding technology, so it can be used to simulate gas flow; (2) the microstructural characteristics and distribution of the porous media are similar to those of real aquifers, which allows for the visual investigation of the water and gas flow characteristics and gas clogging during groundwater recharge; (3) the maximum permitted pressure is up to 8 MPa; thus, it can be applied to a wider range of formation depths.

2.2. Experimental Apparatus

A visual experimental investigation of the gas–liquid flow characteristics and clogged gas formation mechanism was conducted. The experimental setup mainly consisted of three systems: injection system, data acquisition system, and visualization system. A schematic diagram of the experimental setup connection is shown in Figure 1. The injection system included a double plunger pump (LP-10), nitrogen tank, and gas–liquid mixer. The gas–liquid mixer had a built-in fine screen to achieve consistent mixing of the water and gas phases. The data collection system included a high-speed image acquisition system, a pressure sensor, a differential pressure sensor, an electronic balance, a data collector, and a computer. The high-speed image acquisition system was used to capture the transport process of the water and gas phases within the microscopic pore space in real time. A pressure sensor and a differential pressure sensor were connected to the inlet end of the model and both ends of the model, respectively. The gas–liquid separation device was connected to the outlet end of the model, the balance was used to weigh the liquid flow rate, and the drainage method was used to measure the gas flow rate. All of the pressure data and flow rate data were automatically collected and recorded by the data collector. The visualization system included a microscopic visualization model and a model holder. The customized microscopic model gripper had injection and outflow ports, corresponding to the ports on the microscopic model, and O-rings were used on the gripper ports to enhance the gripper tightness.

The high-speed image acquisition system consisted of a high-speed camera (AcutEye-3M-540CXP) (RockeTech technology Co., Ltd., Changsha, China), microscope head, special industrial personal computer (IPC), and faceted light source. The high-speed camera was equipped with a high-sensitivity complementary metal oxide semiconductor (CMOS) image sensor, with a maximum shooting speed of 540 frames/s at a full resolution of 1696 × 1710, and a global shutter. The special IPC was capable of capturing, storing, and playing back the images recorded by the high-speed camera in real time, and it supported data processing functions, such as data playback and image processing. A faceted light source was placed underneath the microvisualization model to provide uniform and sufficient exposure. The system could clearly photograph and record the gas transport and accumulation in the pore space and allowed for the visual investigation of the gas-phase plugging process.

2.3. Main Experimental Procedure

Clean the microscopic visualization model. Inject a certain amount of the following solutions in sequence: ultrapure water, propanol, ultrapure water, SC-1 solution (deionized water, a 5:1:1 mixture of NH₄OH and H₂O₂), and ultrapure water. Connect the model to the experimental setup.

Set-up the liquid double plunger pump and a gas double plunger pump. Set the gas and liquid injection flow rates to 0.1, 0.25, 0.5, and 1 mL/min and 0.05, 0.125, 0.25, and 0.5 mL/min, respectively.

The high-speed image acquisition system records the transport process of both the water and gas in the pore at a shooting speed of 500 frames/s with a 1265 × 1094 resolution.

The pressure sensors monitor the inlet pressure and the differential pressure changes between the two ends of the model during the experiment. The electronic balance monitors the outflow of gas and liquid at the outlet end of the model.

Stop the gas injection after a certain duration of mixed water–gas injection. Continuously inject liquid according to the set flow rate to simulate the groundwater environment at the end of the actual groundwater recharge until the pressure difference between the two ends of the model becomes stable.

Repeat the above mentioned operation steps for the microscopic visualization model with different pore structures.

3. Results and Discussion

3.1. Formation Mechanism of Different Types of Gas Clogging

3.1.1. Gas Clogging Formed by Bypass Flow

When the flow velocity is low and the pressure differences are relatively low, the capillary force becomes the main driving force of the water–gas flow, and the bypass flow is the main reason for the formation of gas clogging. When the water enters two channels with different pore diameters, the water enters the smaller channel at a faster rate due to the capillary force [25]. In the larger channel, the water percolates at a slower rate due to the lower capillary force. When the water breaks through the small channel, the air bubbles are trapped in the large pore channel and form a gas clogging. The gas clogging formed in the large pore channel, in which the capillary force is the main driving force, is shown in Figure 2a.

When the flow rate is high and the pressure difference is relatively large, the inertial force becomes the main driving force of the water and gas movement. At this time, the water and gas two-phase flow mechanism is exactly the opposite of that when the capillary force is the driving force. When the water enters the two channels with different diameters, under the action of the inertial force, the water preferentially enters the larger channel. In the smaller channel, the flow velocity of the water is slower because of the large flow resistance. When the water breaks through the larger channel preferentially, the bubbles in the small channel are trapped, forming a gas clogging. The gas clogging formed in the small channel, in which the inertial force is the main driving force, is shown in Figure 2b.

3.1.2. Gas Clogging Formed in the H-Shaped Channel

Another main reason for the occurrence of gas clogging is the presence of an H-shaped channel. The main mechanism of gas clogging is that the water preferentially breaks through the two relatively parallel channels and flows forward. However, the gas is compressible, and when the water passes through both sides, due to the hydrophilicity of the pore medium, the water also enters the bridge of the H-shaped channel, which compresses the gas in the bridge into bubbles and forms a gas clogging. A gas clogging formed in an H-shaped channel is shown in Figure 3.

3.1.3. Gas Clogging Formed at the Blind End and Corner of a Pore

The blind end and the corner of a pore will always trap a certain amount of gas (Figure 4). Although the pore medium is hydrophilic, it is difficult for water to enter the blind end and the corner of a pore already occupied by gas, especially when the pressure in the flow channel is higher than the gas pressure in the blind end and corner of the pore. Even if the injection rate is increased or the pressure difference is increased, it is difficult to release, because increasing the differential pressure means that the trapped gas in the blind end and corner of the pore is subject to stronger compression, and the gas will be pushed to the depth of the hole and blind end, thus being compressed. Generally, the clogging gas can be removed by reducing the pressure.

3.1.4. Gas Clogging Formed by Cutoff

Due to the difference of the radius between the front and back of the pore channel, the Jamin effect will produce additional resistance to the passing gas when the gas and water flow through the narrow throat. At the same time, due to the hydrophilicity of the rock, water flows along the surface of the pore throat to form a water film, causing water lock damage and further increasing the flow resistance. At this time, the gas passes through the throat in a process of “contraction—deformation—re-expansion”, which will consume part of the energy, thus limiting the movement of gas. Due to the insufficient energy of the bubble, it will break at the exit of the throat, and the front of the bubble will separate to form a new micro bubble that will first pass through the narrow throat. The remaining part of the bubble is trapped in the narrow throat, waiting to collide with the subsequent bubble, and then pass again. The gas clogging formed by the cutoff stays in the center of the channel in the form of bead bubbles, as shown in Figure 5.

The main causes, formation mechanism, and mitigation measure of the four types of gas clogging are summarized in

Table 1. Main causes, formation mechanism, and mitigation measure of trapped gas.

Type of Gas Clogging	Causes	Mechanism	Mitigation Measures
Gas clogging formed by bypass flow	Capillary force and inertial force	Water reaches channels with different apertures. When capillary force is the main power, water preferentially enters channels with small apertures to form gas clogging in larger channels. When inertial force is the main power, water will preferentially enter the large channel and form gas clogging in the small channel.	Change pressure
Gas clogging formed in the H-shaped channel	Capillary force	Water preferentially breaks through the two relatively parallel channels and flows forward, compressing the gas in the bridge into bubbles and forming a gas clogging.	Reduce the pressure to make the gas on the “bridge” expand to the channel on one side or break the pressure balance between the two channels by increasing the pressure to make the gas on the “bridge” flow into the channel on one side.
Gas clogging formed at the blind end and corner of a pore	Disconnectivity of channels	No effective connecting channel.	Reduce pressure
Gas clogging formed by cutoff phenomenon	Jamin effect	Additional capillary resistance is produced by the Jamin effect in narrow or slender channels.	Increase the pressure and supply the energy to trapped gas

.

3.2. Quantitative Characterization of Visual Experiments

When multiphase fluids coexist, the ratio of the effective permeability of each phase of the fluid to the absolute permeability of the porous medium is the relative permeability. The relative permeability is an important parameter for describing multiphase flow in porous media. In this study, in addition to visualizing the location of the gas clogging via images, the relative permeability was also calculated to quantify the degree of clogging. The absolute permeability is only related to the geometric scale of the pores within the porous media [27] and is not affected by factors such as the fluid properties and flow velocity. For all of the microscopic models, a single water-phase seepage experiment was carried out. The absolute permeability of each microscopic model was calculated by recording the pressure difference and flow rate. The relative permeabilities of the water (k_rw) and gas phases were calculated accordingly.

During the water–gas two-phase seepage experiment, we recorded the pressure difference between the inlet and outlet of the microscopic model. The variation in the pressure difference at both ends of the model with time is shown in Figure 6. The pressure difference increased rapidly during the first few minutes of the experiment. Moreover, the higher the flow rate, the faster the pressure difference increased. The homogeneous model with a particle size of 1 mm and a porosity of 30% is presented as an example. In Figure 6, the slope of the blue curve (4.2 kPa/min) is larger than that of the green curve (1.8 kPa/min), i.e., the higher the flow rate, the faster the pressure increases.

For the experiment, the pressure difference exhibited a stable peak and valley band. When the pressure difference between the two ends of the model increased to a certain value, it broke through the bubble clog and reformed the gas-phase pathway. At this time, the resistance of the channel gradually decreased, and the pressure difference decreased. In a complete wave, the relative permeability of the water (k_rw) initially decreased and then increased. The higher the gas content, the stronger the clogging effect. By comparing the differential pressure curves (blue and orange) for the gas–liquid ratios of 1:1 and 1:2 and a liquid flow rate of 0.50 mL/min (Figure 5), it was found that the breakthrough pressure difference was significantly larger for the higher gas–liquid ratio. Moreover, the calculated k_rw values were 0.67 and 0.75, when the gas clogging was the most serious; that is, the higher gas content reduced k_rw. The high-speed images acquired during the experiments show the same pattern, i.e., the higher the gas content, the more pores were occupied by the gas when clogging occurred. The ratios of the gas volume to the pore volume were 5.33% and 4.38% for gas–liquid ratios of 1:1 and 1:2, respectively (Figure 7).

However, in the heterogeneous model experiments, the opposite phenomenon was observed. In the seepage experiment with a particle size of 2–3 mm, the k_rw values were 0.50 and 0.60 at a liquid flow rate of 0.1 mL/min and gas–liquid ratios of 1:1 and 1:2, respectively. The k_rw values were 0.63 and 0.83 at a liquid flow rate of 0.25 mL/min and gas–liquid ratios of 1:1 and 1:2, respectively. That is, the higher gas content decreased k_rw; this is consistent with the results for the homogeneous model. However, the high-speed images acquired during the experiment show that the percentage of the pores occupied by gas was lower when clogging occurred at the higher gas–liquid ratio (Figure 8). The gas clogs accounted for 15.55% and 19.20% of the total pore volume at a flow rate of 0.1 mL/min and gas–liquid ratios of 1:1 and 1:2, respectively. The gas clogs accounted for 5.84% and 11.01% of the total pore volume at a flow rate of 0.25 mL/min and gas–liquid ratios of 1:1 and 1:2, respectively. That is, when gas clogging occurred, k_rw corresponding to the larger volume of clogging gas was higher. The images show that the gas occupied more larger pores via the capillary force when clogging occurred. It is speculated that because gas occupied the large pores, the flow rate of the water in the small pores was accelerated, thereby increasing k_rw.

For the same gas–liquid ratio of 1:1, the higher the flow rate, the sooner the breakthrough time. The breakthrough times were 278 and 550 s at flow rates of 0.50 and 0.25 mL/min, respectively (blue and green curves in Figure 5). At the end of the experiment, after the air pump was stopped, the gas clogging in the pores gradually stabilized, and the differential pressure remained constant after a period of fluctuation.

3.3. Sensitivity Analysis

As was discussed in the previous sections, both the pore structure and gas–liquid ratio had some influence on the motion of the gas and water in the porous media.

3.3.1. Pore Structure

The pore structure directly affects the permeability of an aquifer and the distributions of the gas and liquid phases [28]. The throat is a narrow passage connecting two pore spaces. In the pore structure, the throat is an important factor affecting the relative permeability [29]. The pore–throat characteristic parameters generally include the pore–throat diameter, pore–throat ratio, and pore–throat sorting coefficient. The relationships between the abovementioned parameters and k_rw under the condition of gas clogging were established, and the influences of the pore–throat characteristics on the water flow under the condition of gas clogging were analyzed.

The pore–throat characteristics of the four heterogeneous models were obtained using the ImageJ software to analyze the pore structure of quartz sand with four different particle sizes, which were obtained from CT scans (

Table 2. Pore–throat characteristics of the heterogeneous models.

Model No.	Particle Size Range (mm)	Minimum Pore Throat Diameter (mm)	Maximum Pore Throat Diameter (mm)	Mean Pore Throat Diameter (mm)	Pore–Throat Sorting Coefficient
1	1–2	0.08	1.11	0.43	1.84
2	2–3	0.11	1.07	0.39	2.04
3	3–4	0.14	0.97	0.45	1.96
4	5–8	0.09	1.00	0.30	1.92

). The pore–throat diameter distribution is shown in Figure 9.

• Pore–Throat Diameter

The pore–throat size is measured by the diameter of the largest sphere that can pass through it, i.e., the pore–throat diameter (D_t). It has been shown that the pore–throat size has an important influence on the movement of the gas and liquid phases within a porous medium [30]. As can be seen from Figure 10, k_rw and the pore–throat diameter were correlated. As the pore–throat diameter increased, the connectivity of the pore space increased, the clogging effect of the gas decreased, the percolation resistance of the water decreased, and k_rw increased. Moreover, the larger the pore–throat diameter, the more significant the effect on k_rw. By comparing the results of the three groups of experiments with different flow rates and gas–liquid ratios, it was found that the pore–throat diameter had a certain controlling effect on k_rw, which was not affected by the flow rate and gas content (Figure 10).

2.

Pore–Throat Ratio

The pore–throat ratio is the ratio of the pore radius to the radius of the connected throat, which is a parameter reflecting the alternating characteristics of the pores and throats. The pore–throat ratio is usually used to measure the nonuniformity of the pore openings. In general, the larger the pore–throat ratio, the lower the value of k_rw (Figure 11). The larger the pore–throat ratio, the smaller the diameter of the throat connected to the pore. Thus, the variation in the capillary force between the pore and the throat is greater. Therefore, the gas is more likely to break when passing through a throat, after which the gas exists in the pores in the form of isolated bubbles, thus reducing the water seepage capacity.

3.

Pore–throat sorting coefficient

The pore–throat sorting coefficient is an important parameter used to characterize the pore–throat sorting characteristics of porous media. The sorting coefficient characterizes the uniformity of the pore–throat size distribution and is expressed as follows:

$$S_{\mathrm{p}}=\sqrt{\frac{{\displaystyle \sum _{i=1}^n{(r_i-d_m)}^2\Delta S_i}}{{\displaystyle \sum _{i=1}^n\Delta S_i}}}$$

where d_m is the mean pore–throat radius (mm); r_i is the radius of a certain pore–throat (mm); and ΔS_i is the saturation of a certain pore interval.

The pore–throat sorting coefficient can also be obtained from the cumulative pore throat distribution curve, i.e., the ratio of the pore–throat radius (r₇₅) at a cumulative frequency of 75% to the pore–throat radius (r₂₅) at a cumulative frequency of 25%.

The range of the pore–throat sorting coefficients of the four heterogeneous models was 1.84–2.04, with a small variation range. Therefore, the experimental results of the mean model, which had an average pore throat diameter close to that of the heterogeneous model, were selected for discussion. The particle diameter of the homogeneous model was 2 mm, and the pore throat diameter was 0.44 mm. The existing research results show that the relative permeability of micropores is not related to porosity but is a function related to the distribution of water and gas phases in pores and their wetted perimeter [27]. Therefore, the distribution of water and gas affects the relative permeability. The smaller the sorting coefficient, the more irregular the pore shape. In the water–gas two-phase flow, the pore shape had an important influence on the water gas distribution. The analysis revealed that k_rw was positively correlated with the pore–throat sorting coefficient under the condition of gas clogging (Figure 12). The average pore–throat diameter was slightly negatively correlated with the sorting coefficient for all five models. The models with larger pore–throats had a higher degree of homogeneity, a larger proportion of large pore–throats, and a correspondingly higher k_rw. At the same time, with the decrease in the sorting coefficient, the pores become more complex and the corners increased. Water easily connects and clogs the gas in the center of the pore–throat. Therefore, the pore–throat sorting coefficients and k_rw exhibited a positive correlation under the condition of gas clogging.

3.3.2. Effect of the Gas–Liquid Ratio

The gas content of the recharge water is an important factor affecting gas clogging. Injection experiments with different gas–liquid ratios were carried out using a homogeneous model with a particle diameter of 2 mm and a porosity of 30%.

It can be seen from Figure 13 that at a certain flow rate, k_rw gradually decreased with increasing gas–liquid ratio (1:10, 1:5, 1:2, and 1:1), i.e., increasing the gas content exacerbated the degree of gas clogging. Moreover, when the gas–liquid ratio was low (<1:2), the change in the gas content had a more significant effect on the degree of clogging. When the gas–liquid ratio was high (>1:2), k_rw was less sensitive to the change in the gas content.

4. Conclusions

Clogging is one of the most important factors limiting the development and promotion of artificial groundwater recharge technology. Gas clogging is an important form of clogging. In this study, the location and degree of gas clogging were observed visually at the pore scale through water and gas seepage experiments using microscopic models. In addition, the factors influencing gas clogging were analyzed.

Using a high-speed image acquisition system, it was directly observed that the bypass flow, cutoff, H-shaped channel, and the blind end and corner of a pore can form gas clogging.

The results of the water–gas seepage experiments show that the pressure difference at the two ends of the model exhibited a stable peak–valley trend with time. Within a complete pressure difference peak, which included the process of gas accumulation, gas clogging, breakthrough, and new pathway formation within the pore space, k_rw initially decreased and then increased.

Both the pore structure and gas–liquid ratio affect the degree of gas clogging. As the pore–throat diameter increases, the clogging effect of the gas decreases and k_rw increases. The pore–throat size has a certain influence on the movement of gas and water, which is independent of the flow rate and gas content. The larger the pore–throat ratio, the more likely it is that gas clogging will occur, thus reducing k_rw. The pore–throat sorting coefficient is positively correlated with k_rw. The gas content of the water is also an important factor affecting gas clogging. An increase in the gas content will aggravate the degree of gas clogging, and this effect is more significant at low gas-to-liquid ratios (<1:2).

The next research plan is based on the limitations of the experiment, such as multiple pore structures and flow rates beyond the scope of the instrument. The numerical simulation method was used to build a dynamic model of water and gas coordinated migration under the condition of reinjection, and the results were compared with the microscopic experimental results to verify the accuracy and effectiveness of the model and further explain the results and mechanism of gas clogging under different conditions.