Experimental Study on Influences of Self-Preservation Effects and Memory Effects on the Hydrate Decomposition Process

Natural gas hydrate, also known as hydrate, is a potential highly efficient and clean energy [1] mainly distributed in permafrost and deep-sea sediments. With high porosity, permafrost and seabed sediments are considered as pore systems. At present, most researches on hydrates worldwide are macroscale studies of porous media, while fewer studies are focused on the decomposition of hydrates at the pore scale. Previous studies about hydrate’s “self-preservation effect” and “memory effect” during its formation and decomposition stage are especially few. In the 1880s, Handa [2] found through experiments that gas hydrates showed an unusual stabilization in a nonequilibrium state below the freezing point. He named it “lattice stabilization.” In order to verify this phenomenon’s existence, in 1990, Gudmundsson [3] did a decomposition experiment by storing hydrates in storage tanks with standard thermal insulation performance. During decomposition, hydrates were in a metastable state and could be maintained for a long period. Yakushev [4] carried out a decomposition experiment of methane hydrate under atmospheric pressure. He proposed that an ice layer would form at the surface during the decomposition process, which could effectively prevent methane hydrate from decomposition and could maintain for months even years under atmospheric pressure and a specific subzero temperature condition. In addition, Stern et al. [5] completed a large number of decomposition experiments with pure methane hydrate at atmospheric pressure via a rapid decompression method and a temperature delay method, which also proved the self-preservation phenomenon of hydrate in a metastable state. Previous research believed that the hydrate’s “self-preservation effect” was caused by the ice layer’s shielding effect on gas molecules, which was on the surface of hydrate and produced from the decomposition process at the initial stage of hydrate dissolution. With the development of technology, scholars turned to explore the essence of hydrate’s self-preservation effect from a microscale perspective, which was a big progress compared to the initial observations, physical property tests, and experiments. For instance, in 2002, Takeya et al. [6] used X-ray diffraction technology to measure the decomposition rate of hydrate in different temperature zones and proposed that the “self- preservation” effect of hydrate mainly depended on the diffusion rate of methane on the ice surface. However, this idea failed to consider particle size changes in the process of hydrate decomposition. In 2008, Ebinuma et al. [7] monitored the complete decomposition process of hydrate sediments through X-ray computed tomography technology. On this basis, Guo et al. [8] used scanning confocal microscope to monitor the change of hydrate thickness during the decomposition process. This study suggested that the existence of the “self-preservation effect” would cause the thickness of hydrate ice to increase again. In 2011, Uchida et al. [9], through neutron diffraction technology, proposed that the gas composition, sample size, and the amount of pregenerated ice also impacted the “self-preservation effect.” Many studies have shown that various factors can affect the self-preservation effect of hydrates. However, the self-preservation effect indeed exists. Concerning the “memory effect” of hydrate formation, according to the current research, studies about the memory effect among different hydrates in the world are still in an initial stage. More experimental researches and theoretical analyses are required. In the 1980s, Makogon [10] proposed through experiments that after hydrate formation and decomposition, some microstructures of hydrate would remain in the aqueous phase to promote the growth of hydrate again. To verify this idea, Sloan et al. [11] measured the apparent viscosity of hydrate after decomposition. They believed that the residual cage-like water cluster structure was the possible cause of memory effect. However, due to technical restrictions, this study had a significant experimental limitation. Therefore, Nerheim [12] used laser light scattering technology to prove that the deeper the water structuration, the shorter the induction period. Different from the conclusion mentioned above, Rodger [13] proposed that the memory effect of the hydrate formation was not related to the residual hydrate structure in the solution, but in the high concentration of guest molecules in the solution. However, this hypothesis had some limitations that it could not explain the memory effect among different types of hydrates.

To sum up, the research on the “self-protection effect” and “memory effect” of hydrate only stays in the growing period of hydrate. However, the behavior mechanism of the “self-preservation effect” and “memory effect” is less discussed, and especially, the behavior characteristics are still in lack of explanations. Several critical scientific problems about the “self-preservation effect” and “memory effect” of hydrate in its decomposition period remain to be solved. Firstly, the behavior rule of hydrate’s “self-preservation effect” and “memory effect” in the decomposition period is unclear. The relationship between the two effects has not been studied in depth. Secondly, it still needs to be indicated whether a certain relationship exists between the energy consumption of “self-preservation effect” and gas production during the hydrate decomposition process and how to quantify the correlation. Thirdly, current researches on the secondary synthesis caused by the “memory effect” during the hydrate formation are systematic. However, research on the secondary synthesis caused by the “memory effect” in the decomposition period is less. This study speculates whether there are numerous times of syntheses during the hydrate decomposition period, and how these syntheses affect the hydrate decomposition period. Therefore, on the basis of summarizing the previous research results, this paper does a pore-scale microscopic hydrate decomposition experiment and analyzes the behavioral characteristics and the rule of the “self-preservation effect” and “memory effect” of hydrates in different stages in combination with theoretical and numerical methods. This paper was aimed at providing a theoretical basis for the optimization of the efficient natural gas hydrates exploitation method and the determination of the best exploitation timing.

A self-made experimental device for hydrate formation and decomposition was used in this experiment (Figure 1). The device mainly consists of four parts: the air induction system, the temperature control system, the simulation system for the hydrate formation and decomposition, and the monitoring system [14] (Table 1). Among them, the hydrate formation and decomposition simulation system is mainly composed of a visible reaction kettle with a built-in transparent photoetching glass and photoetching glass. The reactor is sealed at both ends, and the inner wall is threaded and fitted with an O-ring. The two ends of the reactor are stainless-steel cylinders sealed by hexagonal screws. The external size of the photoetched glass in the reactor is $76mm×76mm$⁠, the dissolution size is $45mm×45mm$⁠, and the pore size is ≥20 μm. Keep the oven dry before the experiment. The hydrate decomposition parameter table is shown in Table 1. An inlet valve and an outlet valve are installed on both sides of the reactor, and a pressure sensor is installed, respectively. The inlet valve is used to install a constant pressure when the hydrate is decomposed. At the same time, a high-precision gas flow meter is installed at the inlet and outlet of the reactor, with a minimum accuracy of 0.001 ml and a response time of ≦0.1 s, which is used to monitor the gas output and flow rate. The pipeline of the temperature control system is equipped with an error of ±1°C and PT100 temperature sensor, and the control range is controlled at -40-100°C. The reactor is equipped with an accuracy of ±0.01 MPa and a range of 0-40 MPa. A pressure sensor and a temperature sensor monitor the inlet and outlet pressure changes and temperature changes of the reactor, respectively. This experiment mainly collects the gas production rate and temperature, pressure, and hydrate decomposition situation of hydrate decomposition. The video collected by the microscope is output to the computer through the monitoring system. In order to minimize the heat exchange between the reactor and the external environment, a heat insulation layer is added to the outside of the reactor to achieve the purpose of isolating the experimental environment from the external environment as much as possible.

Part of the experimental materials was prepared in the laboratory. The gas used in this experiment was methane with a purity of over 99%, the liquid was high-purity deionized water, and the experimental materials and reagents are listed in Table 2.

After the hydrate synthesis is completed, adjust the pressure at the outlet end to the same as the other end with constant pressure to balance the integral pressure in the reactor while keeping the temperature in the thermostat constant. Turn on the video recorder, adjust the electron microscope to make the video clear, and have it capture the video of the reaction process. Open the gas outlet valve at the outlet end, reduce the outlet pressure slowly from 8.5 MPa to 1.0 MPa, and keep the pressure constant for about 0.5 h, to avoid the insufficiency of gas production caused by incomplete hydrate decomposition. Meanwhile, turn on the gas flowmeter and record the change of gas production during the decomposition process. When the reading of the gas flowmeter gradually stabilizes, close the gas outlet valve and record the gas production. In this experiment, the decomposition peak only lasts about 3 s, while the entire decomposition of hydrate lasts for quite a long time (more than 1 h), specific experimental steps are as follows.

• (1)

  Clean the reactor wall and the inner and outer walls of plastic containers with deionized water to prevent impurities from interfering. Heat the incubator to 40°C, open the incubator and heat it for 30-40 min, to ensure that the incubator is dry

• (2)

  Open the temperature and pressure sensing system, after the reaction kettle is completely dry, open the video collector, adjust the electron microscope until the video is clear, and close the incubator to collect the reaction video

• (3)

  Open the injection valve, inject 4.5 ml deionized water into the reactor, and then inject 0.5 cm³ methane gas into the reactor, keeping the pressure constant at 8.5 MPa and the temperature at 2°C until the hydrate is completely synthesized (for more than 3 h)

• (4)

  After hydrate is completely synthesized, keep the system pressure constant at the inlet end, open the outlet valve at the outlet end, and slowly adjust the pressure to 1.0 MPa (for more than 1 h) to avoid incomplete hydrate decomposition, resulting in low gas production

• (5)

  At the same time as (4), the gas flowmeter was turned on, and the gas production change during decomposition was recorded. When the reading gradually stabilized, the gas outlet valve was closed and the gas production volume was recorded. In this experiment, the peak period of hydrate decomposition lasted approximately 3 s, but each group of hydrates took a long time (more than 1 h) before it completely decomposed

In the above formula, $n1$ is the substance amount of methane injection before the reaction, in mol

$n2$ is the substance amount of residual methane after the reaction, in mol

$nd$ is the methane gas volume decomposed by the hydrate, in mol

$P$ is the pressure in the lithography glass after the hydrate decomposition, in Pa

$vw$ is the volume of liquid water solution, in m³

$kp$ is the equilibrium constant

In the above formula, both $P$ and $vw$ can be measured through experiments. According to Reference [15], the equilibrium constant is calculated as $kp=e−18.561×e2133.89/T$

• (ii)

  After a comprehensive consideration of the relationship among the system temperature, pressure, and equilibrium constant $kp$⁠, this study establishes a relationship curve of the theoretical gas production of hydrate with time (Figure 2(a)). Two unusual phenomena are found through the comparison of the theoretical gas production curve with the actual gas production curve: firstly, the hydrate decomposition rate reaches a high peak within 500-2000 ms, but the actual gas production differs considerably from the theoretical gas production. Moreover, the actual gas production rate at this stage is well below the theoretical gas production rate. Therefore, it is speculated that hydrate’s “self-preservation” effect occurred in this stage [10]. Secondly, continuous oscillation appears in the hydrate decomposition’s temperature pressure curve during the decomposition period between 2000 ms and 3000 ms (Figure 2(b)). Meanwhile, the gas production rate decreases significantly. Cluster nucleation inside the hydrate is observed. It is presumed that at this stage, hydrate’s “memory effect” weakens multiple syntheses [10, 14]. Therefore, it is necessary to regard the “self-preservation” effect and the “memory effect” happened in the hydrate decomposition period as the critical points of this research and to analyze the performance of the hydrate’s gas production, so as to clarify the behavioral rule of the “self-preservation” effect and the “memory effect” during the decomposition period

In general, the decomposition process of methane hydrate is the solid hydrate decomposes into gaseous methane and liquid water, while the solid hydrate gradually disappears. Eventually, the experiment result can be defined as a complete decomposition in a macroscale sense. However, during the entire hydrate decomposition period, the self-preservation effect and memory effect often occur, which are caused by the intermolecular forces induced by phase transition. During the decomposition period, the two effects accompany with each other, which generally inhibits the hydrate decomposition process. However, the self-preservation effect and the memory effect are still different in essence.

• (1)

  Hydrate’s self-preservation effect occurs in the entire hydrate decomposition period. When the experiment begins, the pressure drop promotes the decomposition of the solid hydrate on the surface to begin. The decomposition starts from the phase transition of the solid hydrate on the surface, and then, a dense water film is gradually formed in the surface medium, which increases the density of the surface hydrate. The dense water film is wrapped on the outer edge of the surface hydrate, which blocks the internal hydrate’s pressure to change drastically to some extent, thereby inhibiting the decomposition of methane hydrate. As the solid hydrate decomposes, the surface temperature rapidly decreases. The wall of the surface water film gradually undergoes a phase change, forming a metastable “quasiliquid film,” which further maintains the internal phase equilibrium conditions for the hydrate. Therefore, the mechanism of the hydrate’s self-preservation effect during its decomposition is to change the phase state of the surface water film to suppress the methane hydrate decomposition. Its essence is a complex phase transition from the liquid phase to the critical state of the metastable “quasiliquid film.” This is one of the key issues of this research

• (2)

  Hydrate’s memory effect refers to the random phenomenon that the CH₄ produced by hydrate decomposition resynthesizes into hydrates with water. It occurs during the entire period of hydrate decomposition. Unlike the self-preservation effect, its essence is a phase transition from the gas-liquid mixed phase to a solid phase induced by residual methane. The pressure curve shows a brief decrement with fluctuation. The memory effect also inhibits the hydrate decomposition process. It is another key research point targeted by this study

In this experiment, the etched glass matrix’s surface is hydrophilic, so the water molecules can disperse on the surface of porous media, which expands the contact area of the surface gas-liquid. Therefore, the phase change of the surface gas-liquid separation can be directly observed during the methane hydrate decomposition. From the separation process of the methane gas and aqueous solution at the surface during the methane hydrate decomposition, the rapid separation stage of the surface gas-liquid can be divided into two stages: methane gas separation at the surface and the formation of “quasiliquid film.” The separation of methane gas at the surface is the main reason which leads the hydrate’s self-preservation effect to begin. This stage begins with the decrease of the hydrate system’s pressure and ends with the random distribution of microbubbles on the hydrate’s surface. The criteria are the methane gas distributes in the form of bubbles on the inner surface of the etched glass matrix and is insoluble in the aqueous solution. For instance, it is observed that methane gas generates in the form of small bubbles in the surface free water at 250 ms after the experiment begins, and the amount of methane gas bubbles increases gradually (Figure 3(c)). This stage lasts about 250 ms with about 0.1 mol of gas production (Table 3).

As the system pressure decreasing continuously, microbubbles accumulate on the surface of the etched glass matrix, gradually break through the barrier of aqueous solution, and form a gas stream along the surface of the glass matrix. At this time, the local temperature of the hydrate surface drops rapidly, and the surface free water distributes on the hydrates surface in the form of a film. Meanwhile, the surface water film transits to a “quasiliquid film,” which indicates that the self-preservation effect begins to act. Besides, methane gas exists in the form of microscale gas flow (Figure 4(b)). When the microscale methane flows on the hydrate’s surface stagnates, and a crystal-like water solution shell different from solid hydrate appears, the separation stage of the surface gas-liquid ends, and the hydrate “quasiliquid film” is eventually finalized. For instance, at 1080 ms after the experiment begins, a crystal-like film on the hydrate’s surface can be observed clearly (Figure 4(c)). This stage lasts for 830 ms, with about 1.5 mol of gas production (Table 3).

The existence of “quasiliquid film” increases the concentration difference between inside and outside the hydrate surface, resulting in the “forced convection” on the hydrate surface and breaking the phase equilibrium condition of the local hydrate. Therefore, in order to better explain the “quasiliquid film” effect on the hydrate decomposition, this study further analyzes the action mechanism of “quasiliquid film” on hydrate decomposition. The essence of self-preservation effect during the hydrate decomposition is the transformation of the free water decomposed by hydrate into a “quasiliquid film,” which wraps around hydrate particles. This film can reduce the conductivity of the hydrate’s temperature and pressure as well as increase the mass transfer resistance. With the continuous decrease of system pressure, the inner hydrate gradually decomposes and shrinks, and the external “quasiliquid film” grows inward and thickens. This phenomenon is known as “core shrinking” (Figure 5), and the ice-like film formed by which inhibits the further decomposition of methane hydrate [16]. In general, temperature and pressure are the main factors affecting the formation of “quasiliquid film.” For instance, when the system temperature falls to the critical freezing point, water molecules in the “quasiliquid film” are more crystal-like, the mass transfer resistance of the “quasiliquid film” is more significant, and the self-preservation effect is more obvious. On the contrary, when the system temperature is higher than the critical freezing point, water molecules in the “quasiliquid film” are more liquid-like, the mass transfer resistance of the film decreases significantly, and the self-preservation effect is inhibited. Similarly, the pressure change can cause the temperature change of the front edge of the hydrate decomposition. Therefore, the influence of the pressure on the “quasiliquid film” cannot be neglected.

The inner gas-liquid separation stage refers to the following process. After the finalization of the “quasiliquid film” (Figure 6(c)), as the system pressure decreases, the inner hydrates gradually decompose, and methane gas diffuses through the “quasiliquid film” to the surface free water (Figure 6(d)), which promotes the “quasiliquid film” to accumulate inward continuously. Unlike the separation stage of the surface gas-liquid, the inner gas-liquid separation stage is the peak period of gas production during the entire decomposition. Therefore, a large amount of gas stream appears in the surface free water in this stage (Figure 7(b)). The secondary flow of the methane gas stream marks the beginning of the gas-liquid separation inside the hydrate. As the inner hydrates decompose, the temperature of the inner edge of the “quasiliquid film” begins to decrease. Meanwhile, a large amount of free water is produced at the same time. Affected by the decrease of temperature, a dense aqueous film is formed on the inner edge of the “quasiliquid film,” which makes the film thicken (Figure 6(e)). As the “quasiliquid film” thickens, the driving force for the hydrate decomposition gradually decreases, while the mass transfer resistance gradually increases. When the mass transfer resistance is equal to or greater than the decomposition driving force, the diffusion of methane gas is suppressed, and a large number of tiny methane bubbles remain on the inner edge of the “quasiliquid film”. At this time, a floccule aggregated by tiny methane bubbles appears inside the hydrate, marking the winding down of the gas-liquid separation (Figures 6(f) and 7(c)). For instance, the thickening of the “quasiliquid film” is obvious at 2000 ms after the experiment begins. Meanwhile, a large number of flocculent crystals appear locally. This phenomenon is maintained for about 920 ms with about 2.5 mol of gas production (Table 3).

To better explain the hydrate’s self-preservation effect, it is necessary to analyze the hydrate decomposition mechanism. Firstly, steady-state hydrate is attached to the etched glass matrix in a solid state. As the decompression by depressurization happens, a large number of liquid-solid and liquid-liquid phase interfaces are formed in the system. At the same time, a smooth gas-liquid interface exists between the gas and the liquid. Hydrates on the complex internal phase interface own low free energy. On the contrary, the gas-liquid interface on the surface is affected by the driving force of decomposition. Thus, the free energy is relatively higher, which provides proper conditions for the start of the self-preservation effect during the methane hydrate decomposition. Secondly, hydrate decomposition is an energy-consuming process that requires the a large amount of heat, which causes the system temperature to decrease significantly. Therefore, an ice-like liquid film is likely to form on the hydrate’s surface, which reduces the pressure transition and increases the mass transfer resistance. Under this decomposition pattern, the hydrate’s preservation effect is more likely to come into play than the decomposition in a single medium.

As mentioned before, hydrate decomposition is an energy-consuming process, which leads to the decrease of the system temperature. Besides, the hydrate’s memory effect often induces multiple hydrate syntheses during the hydrate decomposition process. The microscopic experiment proves that the induction of multiple syntheses by memory effect indeed exists during the hydrate decomposition, which not only has a great influence on hydrate decomposition but also provides a good environment for the hydrate’s decomposition and resynthesis. This study finds out that the memory effect exists throughout the whole decomposition period. For instance, during the memory-effect-induced secondary synthesis process, the memory effect comes into play when the system pressure starts to decrease, forms in the inner gas-liquid separation stage, and ends when the phase equilibrium is destroyed in the methane hydrate system. The floccules generated by the self-preservation effect act as a carrier of the memory effect and gradually form tiny crystal particles. These particles are the crystal nuclei for the hydrate’s secondary synthesis. In the early growth stage of the crystal nuclei, the methane-water phase interface is almost unchanged. The crystal nuclei only grow in the mixed phase of methane gas and hydrate. As the crystal nuclei continue to grow, they slowly extend into the interior of the hydrate in spiny shapes. Hydrates in the front end gradually grow toward the water phase and laterally along the methane-water interface until the hydrate nuclei fill the methane-water interface. It marks the end of the secondary synthesis induced by the hydrate memory effect (Figure 8).

To better understand the memory effect during the hydrate decomposition and the multiple syntheses induced by which, it is necessary to analyze the mechanism of the memory effect. When the methane hydrate begins to decompose, its phase state changes from ordered to disordered. After the surface hydrate has decomposed, the guest molecules become disordered again, promoting the cage-like structure of the water molecule to break. As the hydrate continues to decompose, the internal guest molecules are detached from the cage-like structure, and the system temperature decreases. When the mass transfer resistance is greater than the driving force of the hydrate decomposition, the guest molecules are resealed up in the hydrate’s residual cage-like structure. When the internal hydrate reaches the required temperature and pressure again, the gas molecules will be quickly dissolved by the aqueous solution, forming a high-concentration methane gas miscible region, which promotes the resynthesis of the hydrate. Since the cage-like structure has not been completely destroyed, the reaction skips the nucleus formation stage and directly forms the crystal nuclei of the hydrate. After the new guest molecules enter the cage-like structure, they reconstruct into new crystal structures and finally form new solid-phase hydrates.

• (1)

  Unlike the memory effect in the hydrate synthesis period, the memory effect in the decomposition period exists throughout the entire decomposition period. Besides, the memory effect not only causes the hydrate’s secondary synthesis, but also tertiary, quaternary, even $n$ times of synthesis. The continuous pressure drop is a key reason for the self-preservation effect and the memory effect to stop. It can be seen from the temperature and pressure curve of the hydrate decomposition (Figure 9(a)) that the trend in the late decomposition period shows a continuous oscillation. Due to the low accuracy of the curve, the periodic and frequency changes are indistinct, so this study applies the coupled method of scattering curve to process the curve in the late decomposition period precisely. At the terminal stage, the curve goes stabilized (Figure 9(b)). On the whole, the pressure curve in the hydrate decomposition period is characterized by periodic attenuation. The fluctuation period of the pressure curve can be considered as quantitatively equivalent to be $n$⁠, the number of hydrate synthesis times caused by the memory effect. The fluctuation frequency of the pressure curve can be regarded as the rate of the memory-effect-induced hydrate synthesis. The oscillation characteristic of the curve caused by the memory effect gradually decays then disappears. The probability of hydrate synthesis gradually decreases. When the curve becomes stable, the hydrate’s memory effect disappears, and the hydrate-decomposed gas production reaches its peak

The synthesis induced by the hydrate’s memory effect is a periodic attenuation process. Its synthesis times, $n$⁠, and synthesis rate are important factors which affect the hydrate’s decomposition rate

In the above formula, $i$ represents monoatomic gas, diatomic gas, or triatomic gas, whose values are 3, 5, and 6 generally. $n$ represents the amount of substance. $R$ is the perfect gas constant. $∆T$ is the thermodynamic temperature

When $n≥2$⁠, the $n$time-syntheses of hydrates happen, which are induced by the memory effect during the hydrate decomposition period

It can be seen that the value of $n$ in the memory effect during the decomposition period mainly depends on the energy change in the system. Under ideal conditions, assuming energy loss and pressure are constant, $n$ heads for infinity. When the external energy is sufficient enough, the value of $n$ increases. On the contrary, when the external energy fails to meet the phase equilibrium conditions, the memory effect during the hydrate decomposition period is difficult to act

• (2)

  The self-preservation effect not only reduces the gas production rate during the hydrate decomposition but also affects the gas production volume. The reason is that the energy consumption of the self-preservation effect cannot be neglected. The self-preservation effect directly affects the hydrate decomposition efficiency. The thicker the “quasiliquid film” formed by the self-preservation effect, the greater the energy consumption and the deviation of the gas production amount are. According to the laws of mass and energy conservation, considering the attenuated characteristic of the hydrate decomposition period, this study establishes a mathematical model of the self-preservation’s energy consumption and gas production volume of methane hydrate’s depressurizing production

In the above formula, $K$ is the hydrate’s permeability, $Krg$ is the relative permeability, $μg$ is viscosity, $p$ is the pressure, and $g$ is the gravity acceleration

In the above formula, $C$ is the heat capacity, $T$ is temperature, $e$ is internal energy, and $h$ is the hydrate phase

In the above formula, $M$ is the molecular mass, $ko$ is the decomposition constant of the hydrate, $∆Ea$ is the activation energy of the hydrate, $R$ is the gas constant, $FA$ is the contact area factor of the methane hydrate decomposition, $A$ is the specific area, and $f$ is methane hydrate’s decomposition fugacity

In the above formula, $W$ is the molar molecular weight, $ΔHO$ is the heat produced by per unit volume of hydrate during the decomposition, $h$ is the energy, and $e−λt$ is the attenuation index

Formula (15) is the gas production volume during the hydrate decomposition period

Solve Formula (15) through calculus method. The main parameters involved in the solution are obtained from laboratory tests, which are listed as follows: $ϕ=0.3$⁠, $Sg=0.1$⁠, $∆Ea=8×104J/mol$⁠, $p=920Pa$⁠, $A=0.02$⁠, $ΔHO=51856J/kg×k$⁠. The calculation results are as follows:

• (1)

  The hydrate decomposition period is often accompanied by the occurrence of a self-preservation effect and a memory effect under the effect of intermolecular forces induced by phase transition. In general, these two effects inhibit the hydrate decomposition process. However, they are quite different essentially. The self-preservation effect refers to the change of the phase state of surface water film to inhibit the decomposition of methane hydrate, which is essentially a complex phase transition process from the liquid phase to the metastable “quasiliquid film,” The memory effect is the transition of residual methane gas from the gas-liquid mixed phase to the solid phase, which also inhibits the hydrate decomposition process

• (2)

  The self-preservation effect during the decomposition period is caused by the combined action of the free energy and the decomposition driving force on the complex internal phase interface. The self-preservation effect in the hydrate decomposition period is divided into a “crashing” separation stage of the surface gas-liquid and an “extruding” separation stage of the inner gas-liquid. The surface gas-liquid separation stage is divided into two substages: the surface methane gas separation stage and the formation stage of “quasiliquid film.” Considerable differences are shown in the hydrate decomposition’s specificity in different periods and stages

• (3)

  The memory effect in the hydrate decomposition period is different from that in the hydrate generation period. The memory effect in the decomposition period exists throughout the entire decomposition process. Besides, the hydrates caused by the memory effect not only show the phenomenon of secondary synthesis, which is tertiary, quaternary, and even $n$ times of synthesis. The multiple syntheses induced by the memory effect are periodic attenuation processes. The number of syntheses times is an essential factor affecting the gas production and the hydrate decomposition rate

• (4)

  The self-preservation effect and memory effect are essential behavioral characteristics during the hydrate decomposition period. They are associated with each other rather than independent. The self-preservation effect is an abnormal behavior of the hydrate, which refers to the transition from the solid phase to the gas-liquid mixed phase. The memory effect is another abnormal behavior, which refers to the transition from the gas-liquid mixed phase to the solid phase. In different stages, the two effects play the dominant role alternately. The sustaining pressure drop is the critical reason for the disappearance of the two effects

The data do come from actual experimental simulation, but the data are mainly in the manuscript table and do not need to be listed separately.

The authors declare that they have no conflicts of interest.

The authors acknowledge the support from the State Key Laboratory of Natural Gas Hydrates Key Research and Development Projects (CCL2020RCPS0215ZQN), Heilongjiang Province Key Research and Development Project (GZ20210015), and Heilongjiang Province Joint Guide Fund Project (LH2021E017).