CH₄–C₃H₈ mixed gas hydrate behavior in natural marine sediments: influence of sediment type and dissociation pathways

Abstract

The dissociation behavior of CH₄–C₃H₈ mixed gas hydrates in natural marine sediments is critical to global carbon storage and marine biogeochemical cycles, but the effects of sediment composition and dissociation pathways remain unclear. In this study, dissociation experiments were conducted using marine sediments from the South China Sea, including foraminifera-rich sands, mud-foraminifera sand mixtures, and muds, under controlled heating and depressurization. Hydrate dissociation dynamics and gas release were monitored using in situ and ex situ Raman spectroscopy, supplemented by microscopic observations. Our results show that dissociation of sII hydrate crystals in all sediments begins within the hydrate stability range. The morphological changes in hydrate crystal surfaces correlate with compositional shifts in sediments characterized by high heterogeneity and a broad particle-size distribution. In muddy sediments, dissociation behavior remained uniform regardless of the triggering mechanism, while hydrate crystals in foraminifera-rich sands exhibited distinct behaviors under heating compared to depressurization. Sediment composition influences gas release, although the L(CH₄)/S(CH₄) ratio remains nearly constant across all sediments. Specifically, coarse quartz particles enhance CH₄ and C₃H₈ release, while clay minerals have negligible effects. Foraminifera-rich sands preferentially facilitate CH₄ release under heating, whereas they promote CH₄ and C₃H₈ release under depressurization. These findings provide mechanistic and quantitative insights into sediment-carbon interactions in marine systems, with implications for sustainable carbon management and predicting ocean responses to anthropogenic and climate-driven perturbations.

Highlights

• Gas hydrate dissociation was investigated using in situ and ex situ Raman spectroscopy.

• Five natural marine sediments from the South China Sea were used under controlled heating and depressurization.

• Hydrate dissociation initiated within the stability zone, highlighting sediment-induced metastability.

• Morphology-composition coupling during hydrate dissociation in sediments with high heterogeneity and a broad particle size distribution.

• Foraminifera-rich sands exhibited distinct gas release behavior compared to muds under different dissociation pathways.

1 Introduction

Gas hydrates are crystalline substances with a defined structure, where water molecules form a three-dimensional lattice via hydrogen bonds. These formations create spaces of different sizes, effectively trapping guest molecules within them. In a natural environment, the guest gas molecules primarily consist of light hydrocarbon molecules, including methane (CH₄), ethane (C₂H₆), and propane (C₃H₈). The prospect of recovering CH₄ from the natural gas hydrate reserves has attracted significant interest, driving research into the formation and dissociation dynamics of CH₄ hydrates (Konno et al., 2012; Takeya et al., 2013; Song et al., 2015; Li et al., 2016a). Although CH₄ dominates natural gas hydrates, other hydrocarbons are also significant (Lu et al., 2007; Liu et al., 2015).

Mixed gas hydrates, including C₁-C₄ hydrocarbons, have been identified in various deposits worldwide, such as the Gulf of Mexico, the Caspian Sea, Lake Baikal, the Qilian Mount Permafrost of China, and the Pearl River Mouth Basin in the South China Sea (Klapp et al., 2010; Sun et al., 2018; Wei et al., 2018, 2021). Among mixed gas hydrates, CH₄–C₃H₈ mixed gas hydrate is significant in deep-sea sediments (Yang et al., 2017; Cai et al., 2022). The cage occupancy and gas composition of CH₄–C₃H₈ mixed gas hydrates during formation have been extensively studied (Aladko et al., 2002; Uchida et al., 2004a; Hester et al., 2007; Schicks and Luzi-Helbing, 2013; de Menezes et al., 2019; Klapproth et al., 2019). The results of these studies indicate that even a small amount of C₃H₈ mixed with CH₄ shifts the hydrate structure from sI to sII, significantly enhancing the stability range of the resulting hydrate phase, whereby the composition of the hydrate phase plays a crucial role when evaluating the dissociation process of CH₄–C₃H₈ mixed gas hydrates.

The dissociation of CH₄–C₃H₈ mixed gas hydrates can be triggered by, among others, ocean warming, submarine landslides, and depressurization or heating during production. It was observed that under depressurization, CH₄ is preferentially released over C₃H₈ due to cavity size effects (Sun et al., 2018), resulting in highly concentrated CH₄ which can be effectively separated (Truong-Lam et al., 2020). The dissociation rate follows CH₄ in 5¹² ≥ CH₄ in 5¹²6⁴ > C₃H₈ in 5¹²6⁴. Under heating, Tang et al. (2018) reported similar dissociation rates for small and large cages, while Pan et al. (2023) showed that CH₄ release is faster than C₃H₈, with preferential dissociation of small cages. Molecular dynamics simulations further revealed that CH₄ and C₃H₈ molecules remain in a quasi-liquid layer at the hydrate surface, slightly delaying further dissociation (Naeiji et al., 2024). Although these studies elucidate hydrate dissociation mechanisms, they mainly address simplified systems, leaving the role of natural marine sediments insufficiently understood.

Over the past decades, significant efforts have been devoted to investigating the dissociation of gas hydrates in porous media, involving various materials such as artificial porous media (e.g., silica gel, glass beads), quartz sand, and clayey silts (Yu et al., 2024; Wu et al., 2025). Handa and Stupin (1992) suggested that the dissociation pressure of CH₄–C₃H₈ hydrates in small pores of porous media was higher than those without pores. Riestenberg et al. (2003) and Prasad et al. (2012) found that CH₄ hydrate dissociation behavior in bentonite-clay was similar to that in silica. Uchida et al. (2004b) indicated the dissociation conditions of CH₄ hydrates are primarily influenced by pore sizes, with less influence from surface textures and mineral components. Liu et al. (2008) discovered that particle size impacts the ratio of large to small cages during CH₄ hydrate dissociation. Prasad et al. (2012) noted that the dissociation rate of CH₄ hydrates in fine beads was slower, while Zhao et al. (2015) suggested that the hydrate dissociation rate increases as the particles size decreases. These studies highlight the critical role of sediments in gas hydrate dissociation. However, the dissociation dynamics of CH₄–C₃H₈ mixed gas hydrates in natural marine sediments under controlled heating and depressurization remain insufficiently explored.

The South China Sea hosts extensive gas hydrate accumulations predominantly in clayey silts, which are fine-grained sediments often enriched with foraminiferal sand and typically exhibit high hydrate saturation levels (Li et al., 2016b). Our previous studies have highlighted the influence of muddy and foraminifera-rich sandy sediments from the South China Sea on the gas enclathration of CH₄–C₃H₈ mixed gas hydrates during formation (Mao et al., 2025), however, their dissociation behavior has not been investigated. In this study, a series of in situ and ex situ Raman spectroscopic measurements were conducted on the dissociation behavior of CH₄–C₃H₈ mixed gas hydrates within five natural sediments (namely, foraminifera-rich sands, mud-foraminifera sand mixtures, and muds), capturing molecular behaviors throughout the entire dissociation process. We primarily investigated how pressure-temperature (P-T) conditions and sediment types affect the dissociation of CH₄–C₃H₈ mixed gas hydrates. Accordingly, dissociation experiments were designed to simulate two key pathways: thermal stimulation and depressurization. We examined the morphology of hydrate crystals during dissociation, and assessed changes in hydrate structure, hydrate composition, and cage occupancy throughout the process. These findings not only deepen the understanding of thermodynamic and kinetic responses of CH₄–C₃H₈ mixed gas hydrates to environmental perturbations in natural sediments, but also provide new insights into sediment-carbon interactions within marine biogeochemical cycles.

2 Materials and methods

2.1 Experimental setups for in situ and ex situ Raman spectroscopic measurements

Raman spectroscopic measurements were conducted using a LabRAM HR Evolution dispersive Raman spectrometer from Horiba Scientific, coupled with an Olympus BX-FM microscope and a data acquisition system. For both in situ and ex situ Raman measurements, a 20× magnification microscope objective lens was utilized. A frequency-doubled Nd/YAG solid-state laser (λ = 532 nm) emitting at an output power of 100 mW served as the excitation source. The confocal pinhole was set to 100 μm during the measurements for both in situ and ex situ Raman measurements, providing optimal spatial resolution in z-direction. A silicon chip was employed as the reference standard for the calibration of the Raman system. In situ experiments were conducted within a custom-made high-pressure cell made of Hastelloy (Figure 1A), featuring an inner sample space volume of approximately 550 μl. This cell is capable of operating within a temperature range of 263 K to 295 K and a pressure range of 0.1 MPa to 10 MPa using a pressure controller (ER 3000, Tescom) regulating the pressure in the cell with a precision of 2% rel. Continuous comparison of the measured actual pressure with the target pressure ensures automatic pressure adjustment and thus constant pressure in the system during the dissociation process.

Figure 1

Ex situ Raman spectroscopic analysis was performed using a Linkam cooling stage (Figure 1B). This system maintained the gas hydrate samples at the target temperature of 173.15 K via liquid nitrogen cooling. Ex situ Raman observations are conducted after the gas hydrates formed in a high-pressure vessel, recovered and transported in liquid nitrogen to the Linkam cooling stage. The high-pressure vessel can be pressurized with gas up to 10 MPa. The internal pressure of the vessel is continuously monitored in real-time by a pressure sensor mounted on top of the vessel, using catmanEasy V4.2.2 software. Additional information concerning the Raman spectroscopy setup, the pressure cell, and data analysis can be found in the publication by Schicks et al. (2020) and Mao et al. (2025).

2.2 Sediments

Five natural marine unconsolidated sediments collected from the South China Sea by the Chinese Geological Survey were utilized: foraminifera-rich sands, mud-foraminifera sand mixtures, and muds. Detailed information about these sediments can be found in our previous publication related to the formation process of CH₄–C₃H₈ mixed gas hydrates in these sediments (Mao et al., 2025). We designate foraminifera-rich sands as sediment 1, mud-foraminifera sand mixtures as sediment 2, and the remaining three mud samples as sediment 3, 4, and 5, respectively. The particle size and mineral composition of these five sediments are shown in Figure 2. The particle size distributions are shown in Supplementary Figure S1 in the Supplementary Materials.

Figure 2

2.3 Experimental procedure and dissociation conditions

Before conducting in situ Raman measurements, sediments and deionized water were first loaded into the high-pressure cell. The weight and water-to-sediment ratio of samples are listed in Supplementary Table S1 in the Supplementary Materials. The cell was then sealed and pressurized to 5 MPa at 278 K with gas mixtures containing CH₄ and C₃H₈. A continuous gas flow of about 1 ml/min ensures a constant gas composition during the hydrate formation process. To avoid any temperature effects, the incoming gas was pre-cooled. This, coupled with the thorough cooling of the cell and the small sample volume, avoid any thermal gradients within the sample and facilitated rapid achievement of thermal equilibrium. We analyzed the feed gas composition using Raman spectroscopy. The Raman band at 2917 cm^-1 represents CH₄ molecules in the gas phase, while the Raman band at 869 cm^-1 represents C₃H₈ in the gas phase (Figure 3) (Pan et al., 2023). By calculating the integrated intensities of these two Raman bands, the feed gas composition was determined to be 95.4 mol% CH₄ and 4.6 mol% C₃H₈. For data acquisition and computational procedures, please refer to Mao et al. (2025). During the formation process, gas-hydrate crystals were continuously analyzed by Raman spectroscopy. The dissociation of CH₄–C₃H₈ mixed gas hydrates was induced after a stable hydrate composition and cage occupancy within the hydrate phase were maintained for 4~5 days.

Figure 3

Before conducting ex situ Raman measurements, five different sediment samples, each with a similar water-to-sediment weight ratio (Supplementary Table S1), were individually placed in high-pressure vessels. These vessels were pressurized to approximately 10 MPa using the same CH₄–C₃H₈ mixture as used for the in situ measurements. They were then placed in a cooling freezer where the temperature cycled between 272.15 K and 274.15 K. Throughout the procedure, changes in pressure and temperature within the reactor were monitored and recorded. A pressure decrease was observed during hydrate formation. The pressure in the vessels was monitored until it remained constant for at least 96 hours. This pressure stability indicated the complete transformation of water into gas hydrate. Following this, the synthesized hydrates were recovered from the high-pressure vessels and transferred to a pre-cooled Linkam cooling stage for ex situ Raman observations.

The dissociation of CH₄–C₃H₈ mixed gas hydrates was induced by increasing temperature or decreasing pressure to mimic climate change and its consequences such as global warming and the melting of ice sheets. According to calculations from CSMGem (Sloan and Koh, 2008), the crystals will begin to lose their original morphology near the equilibrium temperature T = 289.65 K at P = 5.0 MPa or the equilibrium pressure P = 1.23 MPa at T = 278.15 K, indicating that dissociation is activated (Figure 4). During the depressurization-induced dissociation process, the pressure was reduced stepwise from 2.5 MPa to 0.8 MPa at intervals of 0.1-0.2 MPa (T = 278.15 K), as shown in Supplementary Figure S2A and Supplementary Table S2 in the Supplementary Materials. The pressure was hold at each step for ~30 mins. During that time, in situ Raman spectrometric measurements were conducted.

Figure 4

During the heating-induced dissociation process, the temperature was increased stepwise from 278.15 K to 288.15 K at increments of 0.5-1.0 K, as detailed in Supplementary Figure S2B and Supplementary Table S2. It should be noted that during the heating-induced dissociation process the pressure maintained constant at P = 5.0 MPa). The temperature was held for 20–35 minutes, during which in situ Raman spectrometric measurements were conducted. Ex situ Raman analyses were initiated at 173.15 K and 0.1 MPa. The temperature was then increased stepwise to 268.15 K at increments of 5–10 K, as shown in Supplementary Figure S2C and Supplementary Table S2. The temperature at each step was maintained for at least 6 minutes during ex situ Raman spectrometric measurements, with the specific duration adapted to ensure optimal spectral quality at each temperature condition. Once significant hydrate dissociation was observed, the temperature was held constant. Only after no further morphological changes occurred the temperature was raised again. It is worth noting that ex situ Raman spectroscopic experiments only support heating-induced dissociation.

Selected hydrate crystals were continuously monitored and analyzed at each pressure/temperature step. During both in situ and ex situ Raman spectroscopic measurements, the laser beam was focused on the surface of different hydrate crystals formed in the natural sediments. For each run with a specific sediment, either three or seven hydrate crystals were selected for analysis. At each pressure/temperature step, Raman spectra were measured for each of the selected hydrate crystals. For the in situ measurements, seven crystals were selected and used to determine the hydrate composition and respective cage occupancy. During the ex situ Raman measurements, three different hydrate crystals were analyzed. Replicate experiments were conducted for each run to ensure reproducibility. For certain sediments, more than two repetitions were performed to guarantee the reliability and accuracy of the results. For instance, additional replicates were conducted for depressurization-induced dissociation in sediments 1 and 2 (in situ Raman measurements), heating-induced dissociation in sediments 2 and 3 (in situ Raman measurements), and heating-induced dissociation in sediment 1 and 5 (ex situ Raman measurements). In this study, however, only two representative sets of results are presented. We would like to point out here that when the Raman measurements on the gas hydrate crystals were carried out during their decomposition, not all selected crystals could be analyzed by the end of the experiment, as the decomposition process was completed earlier for some crystals than for others. The crystals were therefore not always analyzed in the same order.

2.4 Raman spectra data analysis

The Raman band observed at 876 cm^-1 corresponds to the C−C stretching vibrations of C₃H₈ encased into the larger 5¹²6⁴ cavities of sII hydrates (Figure 5). The Raman bands at 2902 cm^-1 and 2912 cm^-1 represent CH₄ within the larger 5¹²6⁴ and smaller 5¹² cavities of sII hydrates, respectively (Figure 5). In this study, the large-to-small cavity ratio was only accurately calculated using results from ex situ Raman measurements in the absence of a coexisting gas phase. This approach was chosen primarily to avoid an incorrect calculation of the integrated intensities for the Raman bands of CH₄ in 5¹² cages (2912 cm^-1) and CH₄ in the gas phase (2917 cm^-1) (Figure 5). This is because these two bands are very close to each other, making a clear assignment difficult. Especially with low intensities of the Raman bands, a partial overlapping of the bands could lead to considerable misinterpretations. Therefore, no calculation of the large-to-small cavity ratio was performed on the basis of the in situ Raman measurements.

Figure 5

The molar composition of the gas and hydrate phases was determined semi-quantitatively using Raman spectroscopy. The calculation method followed the approach outlined by Beeskow-Strauch et al. (2011). In a mixed system, the molar fraction of a specific component can be derived using the simplified Equation 1, which is based on Placzek’s ratio method (Placzek, 1934). This method is applicable because the integrated intensity of a component’s Raman band is directly proportional to the number of molecules present in the sample.

In this equation, X_a denotes the molar fraction of component a, A_a represents the integrated intensity of its Raman band, corresponds to the Raman scattering cross-section factor of component a, and accounts for instrumental efficiency. The index i and Σ signify the respective values for all species in the sample and their total sum. To improve accuracy, Raman band areas were adjusted using wavelength-independent scattering cross-section factors. These factors are assumed to remain constant regardless of pressure variations, cage structure, or phase composition (Schrötter and Klöckner, 1979; Schrader, 1995; Burke, 2001).

The large-to-small cavity ratio was determined from ex situ Raman measurements using Equation 2. This calculation assumes that CH₄ occupies both the small 5¹² and large 5¹²6⁴ cavities in sII hydrate, while C₃H₈ is exclusively confined to the large 5¹²6⁴ cavities.

Here, a represents the integrated intensity of a specific Raman band recorded in the spectrum. In this study, the relative Raman scattering cross-section factors applied were 8.63 for CH₄ and 1.60 for C₃H₈ (Schrader, 1995; Burke, 2001).

3 Results

3.1 Depressurization-induced dissociation process

The CH₄ content and C₃H₈ content of individual hydrate crystals from the five sediments, measured before depressurization-induced dissociation, vary slightly (Figure 6). The C₃H₈ content ranges from 20 mol% to 40 mol% in the hydrate phase, and the CH₄ concentration is between 60 mol% and 80 mol% for the hydrate crystals before the dissociation process starts. The average hydrate phase composition before dissociation is close to the composition calculated by CSMGem, approximately 66.5 mol%. CH₄ and 33.5 mol% C₃H₈. This significant enrichment of C₃H₈ into the hydrate phase is a result of its preferred incorporation into the large 5¹²6⁴ cages of the structure II hydrate. Due to their size, C₃H₈ molecules are preferentially enclosed in the large cages, on which they exert a superior stabilizing effect compared to CH₄ (Schicks and Luzi-Helbing, 2013; Medvedev et al., 2015; Du et al., 2023; Mao et al., 2025), leading to a higher stability of the resulting hydrate phase. For sediments 3 and 5, Runs 1 and 2 show a slight difference in the pressures at which a significant change in the morphology of the hydrate crystals was observed. The changes in the composition of the hydrate phase show a similar development in both runs, independently of this.

Figure 6

The mixed gas hydrates formed in sediments 1, 2 and 3, showed a strong increase in compositional variation after reaching the reduced pressure at which the significant change in hydrate crystal morphology occurred. In other words, the relative concentration of CH₄ in hydrate crystals exhibited different behaviors among sediments 1, 2, and 3, with values increasing in some dissociating crystals while decreasing in others (Figures 6A-C). Accordingly, the relative content of C₃H₈ also changes correspondingly. This indicates that in some hydrate crystals, the release rate of C₃H₈ molecules is faster than that of CH₄ molecules, while in others, the release rate of CH₄ molecules is faster than that of C₃H₈ molecules. It should be noted that due to the non-stoichiometric character of gas hydrates their composition varies between individual hydrate crystals but also within a hydrate crystal (Schicks et al., 2020). When the decomposition begins, it does not immediately affect the entire crystal, but starts at its surface. Due to the dissociation of the unit cells at the interface between the hydrate crystal and its environment, deeper layers of the crystal form the surface. In other words, the hydrate decomposition continues layer by layer as described in Pan et al. (2023). Since the composition of the hydrate crystals may also fluctuate from layer to layer, the compositional changes of the individual hydrate crystals as well as the entire hydrate phase do not show consistent changes in CH₄ or C₃H₈ concentrations, but – regardless of the absolute fluctuations in concentration – a general tendency regarding the compositional change.

The relative concentrations of CH₄ and C₃H₈ in hydrate crystals in sediments 4 and 5 remain almost unchanged during the entire process (Figures 6D, E), showing only a slight increase in variations, if any. This suggests that the release rates of both, C₃H₈ and CH₄ molecules from the hydrate cages remains in a fixed ratio for these sediments. Furthermore, no alternate transient structures were observed throughout the entire dissociation process, indicating that the collapse of gas hydrate cages occurs uniformly as a single entity.

The morphological changes on the surface of hydrate crystals in sediments 1, 2 and 3 correspond to the changes in hydrate composition. Taking a crystal from sediment 3 as an example (Figure 7A), there is a slight change in the appearance of the hydrate crystal morphology at pressure > 1.5 MPa, but it remains largely unchanged. When pressure decreases to ≤ 1.5 MPa, the hydrate crystals in the central region undergo a distinct morphological transition, coinciding with the pressure at which the hydrate composition begins to show a change. Finally, the crystal completely disappears at 1.2 MPa.

Figure 7

For the gas hydrates in sediments 4 and 5, the average composition remains almost unchanged, and the observed morphological changes on the surface of the hydrate crystals occur more gradually during the depressurization-induced dissociation process. Taking a crystal from sediment 5 as an example (Figure 7B), it appears as if the hydrate crystals are slowly melting without suddenly changing their morphology.

The equilibrium pressure of CH₄–C₃H₈ mixed gas hydrates at around 278.15 K is close to 1.23 MPa using CSMGem, with a feed gas composition of 95.4 mol% CH₄ and 4.6 mol% C₃H₈. The pressure at which the CH₄-C₃H₈ morphology in hydrate crystals begins to change significantly in different sediments is between 1.3 MPa and 1.7 MPa. The dissociation pressure of hydrate crystals in all sediments is higher than the equilibrium pressure.

3.2 Heating-induced dissociation process

3.2.1 In situ Raman measurements

The CH₄ content and C₃H₈ content of hydrate crystals before dissociation by heating in five sediments also vary slightly (Figure 8). The C₃H₈ content achieves 20~40 mol% in the hydrate phase, and the concentration of CH₄ reaches 60~80 mol% for the selected crystals before the dissociation process begins. This range is consistent with the results presented in Figure 6 for the gas hydrate crystals formed for the dissociation process induced by depressurization. In two dissociation experiments using the same sediments, the onset of obvious morphology changes occurs at slightly different temperatures in sediments 3, 4, and 5. Nevertheless, the changes in hydrate composition during the heating-induced dissociation process followed a similar development in both runs.

Figure 8

After the temperature reaches the level at which a significant change in hydrate crystal morphology is observed, the relative concentration of CH₄ decreases in some hydrate crystals in Runs 1 and 2 of sediment 1 (Figure 8A), indicating that CH₄ molecules are released from the hydrate cages faster than C₃H₈ molecules. In sediments 2 and 3, the relative concentration of CH₄ in the hydrate crystals observed in Runs 1 and 2 showed broader variation, with values increasing in some crystals and decreasing in others (Figures 8B, C). Accordingly, the relative content of C₃H₈ also changed. This indicates that in some hydrate crystals, the release rate of C₃H₈ molecules from the hydrate cages is faster than that of CH₄ molecules, while in others, the release rate of CH₄ molecules from the hydrate cages is faster than that of C₃H₈ molecules. The relative concentrations of CH₄ and C₃H₈ in hydrate crystals in sediments 4 and 5 remain almost unchanged during the dissociation process (Figures 8D, E), exhibiting only a slight increase in variations, if any. This indicates that the release rates of C₃H₈ and CH₄ molecules from the hydrate cages remain in a fixed ratio in the hydrate crystals. Additionally, no other transient structures were observed during the heating-induced dissociation process, and the gas hydrate cages appeared to collapse as an entity.

In sediment 1, noticeable surface morphological changes in hydrate crystals appear prior to any substantial alteration in their composition. For sediments 2 and 3, the pronounced surface morphological changes of hydrate crystals correspond to compositional variations. For instance, when considering a crystal from sediment 3 as an example (Figure 9A), only small changes in the hydrate crystal morphology are observed at temperatures < 286.15 K. However, after the temperature rises to 286.15 K, the hydrate crystals noticeably lose their smooth surface and sharp edges. Finally, the crystal disappeared at 287.15 K.

Figure 9

In sediment 4 and sediment 5, significant morphological changes on the surface of hydrate crystals gradually occur during the heating process, but they do not correspond to changes in hydrate composition. Taking a crystal from sediment 5 as an example (Figure 9B), it can be observed that the hydrate crystal becomes smaller and smaller and disappears at temperatures above 286.15 K.

The equilibrium temperature of CH₄–C₃H₈ mixed gas hydrates at 5 MPa is close to 289.65 K calculated using CSMGem with a feed gas composition of 95.4 mol% CH₄ and 4.6 mol% C₃H₈. In different sediments, significant morphological changes of CH₄–C₃H₈ hydrate crystals occur between 284.15 K and 287.15 K. Therefore, in all sediments, the dissociation temperature of hydrate crystals is lower than the equilibrium temperature calculated by CSMGem.

3.2.2 Ex situ Raman measurements

The results of ex situ Raman measurements are depicted in Figure 10, with repeat results shown in Supplementary Figure S4 in the Supplementary Materials.

Figure 10

While the gas hydrates in sediments 1 and 2 show a relatively homogeneous composition before decomposition, the gas hydrate crystals in sediments 3, 4 and 5 already show a significantly higher variance in their compositions (Figure 10). However, the average composition of the gas hydrate phase is similar to those obtained in the in situ measurements (Figures 6, 8). The initial ratio of large (L)-to-small (S) cavities filled with CH₄ (L(CH₄)/S(CH₄) ratio) in hydrate crystals across different sediments is generally less than 0.2. In contrast, the ratio when considering both CH₄ and C₃H₈ occupancy in large cages (i.e., L(CH₄+C₃H₈)/S(CH₄) ratio) ranges from approximately 0.4 to 0.8. Since structure II hydrates consist of 8 large and 16 small cavities, the ratio should be 0.5 provided that all cavities are filled with a CH₄ or C₃H₈ molecule. The ratios calculated from Raman measurements suggest, however, that not all small 5¹² cages and large 5¹²6⁴ cages in mixed gas hydrates were fully occupied by CH₄ and/or C₃H₈ (Mao et al., 2025). The gas hydrates in sediment 5 are an exception; here the values for the L(CH₄+C₃H₈)/S(CH₄) ratio vary considerably between 0.4 and 1.6 and are therefore not meaningful. The following discussion of the L(CH₄+C₃H₈)/S(CH₄) ratio therefore refers to the gas hydrates in sediments 1-4.

After the initiation of hydrate dissociation, the changes in hydrate composition from different sediments are consistent with the results obtained from in situ measurements. The L(CH₄)/S(CH₄) ratio in sediments 1–5 does not show a clear change with increasing temperature, suggesting that the breakdown speed of small cavities is similar to that of large cavities (Figure 10). The L(CH₄+C₃H₈)/S(CH₄) ratio in hydrate crystals in different sediments exhibits variations. In sediment 1, the L(CH₄+C₃H₈)/S(CH₄) ratio of hydrate crystals increases (Figure 10A). In sediments 2 and 3, the L(CH₄+C₃H₈)/S(CH₄) ratio of some hydrate crystals increases, while it decreases for others (Figures 10B, C). The distribution range of the L(CH₄+C₃H₈)/S(CH₄) ratio remains unchanged during the heating process in sediment 4 (Figure 10D).

The equilibrium temperature of CH₄–C₃H₈ mixed gas hydrates at one atmosphere is close to 235 K. In our study, CH₄–C₃H₈ mixed gas hydrates in different sediments began to undergo significant morphological changes between 218.15 K and 226.65 K, indicating that decomposition occurred in all sediments below the equilibrium temperature.

4 Discussion

4.1 Crystalline structure

No other transient structure was observed during dissociation by depressurization and heating methods, confirming the structural stability of pure sII hydrates. Although a clear determination of the hydrate structure by Raman spectroscopy alone is not possible, our Raman spectroscopic analyses show two important indications for the formation of a sII hydrate and distinguishing features from sI hydrates: (1) The positions of the Raman bands for CH₄ are at 2902 cm^-1 (large 5¹²6⁴ cages) and 2912 cm^-1 (small 5¹² cages) (Figure 11), distinct from sI hydrate signatures (2905 cm^-1 for 5¹²6⁴ and 2915 cm^-1 for 5¹² cages) (Pan et al., 2023; Naeiji et al., 2023); and (2) A markedly lower CH₄ occupancy in 5¹²6⁴ cages (2902 cm^-1) versus 5¹² cages (2912 cm^-1) (Figures 6, 8, 10), whereas sI hydrates show a characteristic 3:1 ratio (5¹²6⁴/5¹² cages) (Cai et al., 2022). These findings align with our prior work on sII CH₄–C₃H₈ mixed gas hydrate formation in sediments (Mao et al., 2025) and independent studies of pure systems (Tang et al., 2018; Truong-Lam et al., 2020; Pan et al., 2023). These results suggest that sediments have no influence on the structural type of CH₄–C₃H₈ mixed gas hydrates.

Figure 11

4.2 The impact of dissociation methods

Comparative analysis of different dissociation methods reveals two key findings regarding hydrate behavior in natural marine sediments: 1) The dissociation temperature of hydrate crystals is consistently lower than the calculated equilibrium temperature of the bulk hydrate phase at given pressure (Figure 8), while 2) the dissociation pressure in all sediment types exceeds the calculated equilibrium pressure at given temperature (Figure 6). In both cases, this means that the decomposition of the hydrates begins even though the temperature and pressure conditions are still within the stability range of the mixed gas hydrates. This observation is in general agreement with the work of Uchida et al. (2002), who investigated the direct influence of the pore size of porous media on the decomposition conditions of simple CH₄, CO₂ and C₃H₈ hydrates. They found that as the pore size decreased, the equilibrium temperature value tended to decrease at a given pressure. Assuming that hydrate crystals in small pores have different physical properties than a bulk hydrate phase, they explained the differing decomposition temperatures with the Gibbs-Thomson effect. These conclusions can generally be applied to our findings as well: the observed premature decomposition behavior may be attributed to sediment-specific properties and varying ratios of mineral components to hydrates affecting the microstructural environment and thus alter phase equilibrium conditions (Uchida et al., 2002).

The hydrate crystals in foraminifera-rich sands (sediment 1), however, exhibit distinct gas release behaviors under different dissociation methods. The results indicate that, during depressurization, the relative concentration of CH₄ in different hydrate crystals shows heterogeneous behavior: in some crystals it remains constant, in others it increases, and in yet others it decreases. Under heating, the relative concentration of CH₄ in some crystals remains unchanged, while in others it decreases. This differential behavior suggests that, in foraminifera-rich sands, heating provides more efficient CH₄ release compared to depressurization. The higher release rate of CH₄ molecules under heating may be explained by a higher porosity and interconnected pore structure of foraminifera sands, which promotes gas transport overall, allowing the smaller CH₄ molecules to be released particularly easily in this situation.

4.3 The impact of natural sediments on CH₄ and C₃H₈ release

The results obtained from ex situ Raman measurement were primarily used when discussing the influence of sediments on gas hydrate behaviors during the heating-induced dissociation process (Figure 10). In the studies reported by Truong-Lam et al. (2020) and Pan et al. (2023), it was observed that during the heating- or depressurization-induced dissociation process, the relative concentration of CH₄ in the hydrate crystals decreased, while that of C₃H₈ increased. Pan et al. (2023) indicated that CH₄ molecules exhibit higher diffusivity from the hydrate surface to the gas phase but have a lower capacity to stabilize hydrate cavities compared to C₃H₈. The results of this study indicate that, only the hydrate dissociation in sediment 1 during the heat-induced decomposition was consistent with the scenario observed in the studies mentioned above. Based on the results shown in Figures 8, 10 and the particle size and mineral composition presented in Figure 2, we conclude that the presence of foraminifera sands in sediment 1 facilitates CH₄ release. Sediments 3–5 are all muddy sediments, but the effect of sediment 3 on gas hydrate dissociation also exhibits differences compared to the other muddy sediments. This is primarily related to its high heterogeneity and wide particle size distribution, including coarse-grained particle content and muds. Similarly, sediment 2 contains a significant amount of coarse-grained sediment in addition to foraminifera-rich sands. It appears that the heterogeneity of the sediment leads to a different effect on the hydrate decomposition process. The release of CH₄ is promoted in some hydrate crystals, while in others, the release of C₃H₈ is promoted. This leads to a high variance in the composition of the hydrate phase during decomposition. The results of the experiments with sediments 4 and 5 indicate that clay has no significant effect on the relative release ratio of CH₄ to C₃H₈.

As shown in Figure 10, the L(CH₄)/S(CH₄) ratio of the hydrate in all sediments remains almost unchanged. This indicates that the sediment has no significant effect on the relative rate of CH₄ release from either large or small cages. The changes in the L(CH₄+C₃H₈)/S(CH₄) ratio in hydrate crystals in sediments 1–4 are consistent with the variation in C₃H₈ content within the hydrate crystals. The L(CH₄+C₃H₈)/S(CH₄) ratio in some hydrate crystals in foraminifera-rich sands (sediment 1) increases, consistent with the conclusions proposed by Pan et al. (2023). Above phenomena highlight the influence of sediment composition on gas hydrate dissociation dynamics.

It is important to acknowledge the limitations of our experimental approach. While the employment of a small-volume, constant-gas-supply system provides highly controlled conditions ideal for elucidating fundamental mechanisms of hydrate dissociation (e.g., dissociation begins even within the hydrate stability range), this setup undoubtedly simplifies the complex, multiphase environment of natural marine sediments. Natural systems involve dynamic biogeochemical processes, heterogeneous mineral surfaces, and fluctuating fluid flow, which are not fully captured here. Therefore, the kinetic-related quantitative findings of this study should be extrapolated to natural settings with caution.

5 Conclusion and implications

This study investigates the dissociation process of CH₄–C₃H₈ mixed gas hydrates in five different natural marine sediments from the South China Sea, including foraminifera-rich sands (sediment 1), a mixture of mud and foraminifera-rich sands (sediment 2) and muds (sediments 3-5), under heating and depressurization conditions. The results show that no transient structures were observed during dissociation by either heating or depressurization, reinforcing the structural stability of sII hydrates across different decomposition pathways. In sediments with high heterogeneity and broad particle size distribution, the observed morphological changes on the hydrate crystal surfaces correlate strongly with changes in their composition. This suggests that the dynamics of gas release may influence these morphological alterations.

The dissociation method did not significantly alter the development of composition change in sediments 2-5. In contrast, hydrate crystals in foraminifera-rich sands (sediment 1) exhibit distinct dissociation behaviors. Depressurization induces non-monotonic CH₄–C₃H₈ concentration variations, while heating yields predictable compositional shifts (CH₄ decrease and C₃H₈ increase with temperature). Sediment heterogeneity, specifically a wide range of particle sizes (e.g., coarse grains and mud), significantly influences gas hydrate dissociation. This leads to a high variance in the relative release of CH₄ versus C₃H₈, as observed in sediments 2 and 3. In contrast, clay-rich sediments (sediments 4 and 5) show no measurable influence on the relative release ratio of CH₄ to C₃H₈. The L(CH₄)/S(CH₄) ratio of the hydrate in all sediments remains almost unchanged. Regardless of the type of sediment, we observed that both pressure-induced and heat-induced dissociation of gas hydrates started under P-T conditions within the calculated stability range.

The aforementioned findings reveal that the dissociation kinetics of CH₄–C₃H₈ mixed gas hydrates vary under different conditions in sediments from the South China Sea, indicating that sediment composition and dissociation pathways jointly regulate gas release behavior. This understanding holds significant value for optimizing natural gas hydrate extraction technologies and elucidating the mechanisms of sediment-carbon interactions within the marine carbon cycle. Under climate warming conditions, CH₄–C₃H₈ mixed hydrate reservoirs in highly permeable sandy sediments (e.g., foraminifera-rich sands) may become hotspots for CH₄ release, triggering local microbial ecological shifts and seabed hypoxia. Furthermore, the observation that CH₄–C₃H₈ mixed hydrates in all sediment types dissociate within the theoretical stability zone challenges traditional phase equilibrium models. Sediment micro-scale effects may promote premature dissociation, suggesting that current models likely underestimate actual dissociation rates and carbon fluxes. Future carbon-climate predictions must incorporate sediment characteristics to improve accuracy.

References

• 1

  Aladko E. Y. Dyadin Y. A. Manakov A. Y. Zhurko F. V. Larionov E. G. (2002). Phase diagrams of the ternary gas hydrate forming systems at high pressures. Part 1. propane-methane-water system. J. Supramol. Chem.2, 369–376. doi: 10.1016/S1472-7862(03)00061-3

  • CrossRef
  • Google Scholar

• 2

  Beeskow-Strauch B. Schicks J. M. Spangenberg E. Erzinger J. (2011). The influence of SO₂ and NO₂ impurities on CO₂ gas hydrate formation and stability. Chem. Eur. J.17, 4376–4384. doi: 10.1002/chem.201003262

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  Burke E. A. J. (2001). Raman microspectrometry of fluid inclusions. Lithos, fluid Inclusions: Phase relationships-methods-applications. A special issue honour Jacques Touret55, 139–158. doi: 10.1016/S0024-4937(00)00043-8

  • CrossRef
  • Google Scholar

• 4

  Cai W. Zhan L. Huang X. Lu H. (2022). Raman micro-imaging of the coexistence of sI and sII hydrates formed from a mixed methane-propane gas in a confined space. Acta Geol. Sin. - Engl. Ed.96, 674–679. doi: 10.1111/1755-6724.14929

  • CrossRef
  • Google Scholar

• 5

  de Menezes D.É.S. Sum A. K. Desmedt A. Pessôa Filho P. de A. Robustillo Fuentes M. D. (2019). Coexistence of sI and sII in methane-propane hydrate former systems at high pressures. Chem. Eng. Sci.208, 115149. doi: 10.1016/j.ces.2019.08.007

  • CrossRef
  • Google Scholar

• 6

  Du S. Han X. Cai W. Zhu J. Ma X. Han S. et al . (2023). Formation of the structure-II gas hydrate from low-concentration propane mixed with methane. Chin. J. Chem. Eng.58, 306–314. doi: 10.1016/j.cjche.2022.10.014

  • CrossRef
  • Google Scholar

• 7

  Handa Y. P. Stupin D. Y. (1992). Thermodynamic properties and dissociation characteristics of methane and propane hydrates in 70-.ANG.-radius silica gel pores. J. Phys. Chem.96, 8599–8603. doi: 10.1021/j100200a071

  • CrossRef
  • Google Scholar

• 8

  Hester K. C. Dunk R. M. Walz P. M. Peltzer E. T. Sloan E. D. Brewer P. G. (2007). Direct measurements of multi-component hydrates on the seafloor: Pathways to growth. Fluid Phase Equilibria Properties Phase Equilibria Product Process Design261, 396–406. doi: 10.1016/j.fluid.2007.07.053

  • CrossRef
  • Google Scholar

• 9

  Klapp S. A. Murshed M. M. Pape T. Klein H. Bohrmann G. Brewer P. G. et al . (2010). Mixed gas hydrate structures at the Chapopote Knoll, southern Gulf of Mexico. Earth Planet. Sci. Lett.299, 207–217. doi: 10.1016/j.epsl.2010.09.001

  • CrossRef
  • Google Scholar

• 10

  Klapproth A. Piltz R. O. Kennedy S. J. Kozielski K. A. (2019). Kinetics of sII and mixed sI/sII, gas hydrate growth for a methane/propane mixture using neutron diffraction. J. Phys. Chem. C123, 2703–2715. doi: 10.1021/acs.jpcc.8b06693

  • CrossRef
  • Google Scholar

• 11

  Konno Y. Uchiumi T. Oyama H. Jin Y. Nagao J. Masuda Y. et al . (2012). Dissociation behavior of methane hydrate in sandy porous media below the quadruple point. Energy Fuels26, 4310–4320. doi: 10.1021/ef300628c

  • CrossRef
  • Google Scholar

• 12

  Li C. Hu G. Zhang W. Ye Y. Liu C. Li Q. et al . (2016b). Influence of foraminifera on formation and occurrence characteristics of natural gas hydrates in fine-grained sediments from Shenhu area, South China Sea. Sci. China Earth Sci.59, 2223–2230. doi: 10.1007/s11430-016-5005-3

  • CrossRef
  • Google Scholar

• 13

  Li X. Xu C. Zhang Y. Ruan X. Li G. Wang Y. (2016a). Investigation into gas production from natural gas hydrate: A review. Appl. Energy172, 286–322. doi: 10.1016/j.apenergy.2016.03.101

  • CrossRef
  • Google Scholar

• 14

  Liu C. Lu H. Ye Y. Ripmeester J. A. Zhang X. (2008). Raman spectroscopic observations on the structural characteristics and dissociation behavior of methane hydrate synthesized in silica sands with various sizes. Energy Fuels22, 3986–3988. doi: 10.1021/ef800440s

  • CrossRef
  • Google Scholar

• 15

  Liu C. Meng Q. He X. Li C. Ye Y. Lu Z. et al . (2015). Comparison of the characteristics for natural gas hydrate recovered from marine and terrestrial areas in China. J. Geochem. Explor.152, 67–74. doi: 10.1016/j.gexplo.2015.02.002

  • CrossRef
  • Google Scholar

• 16

  Lu H. Seo Y. Lee J. Moudrakovski I. Ripmeester J. A. Chapman N. R. et al . (2007). Complex gas hydrate from the Cascadia margin. Nature445, 303–306. doi: 10.1038/nature05463

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  Mao P. Schicks J. M. Pan M. Wu N. (2025). CH₄-C₃H₈ mixed gas hydrates formation in marine mud and foraminifera-rich sand from the South China Sea: an experimental approach. Front. Mar. Sci.12. doi: 10.3389/fmars.2025.1510050

  • CrossRef
  • Google Scholar

• 18

  Medvedev V. I. Gushchin P. A. Yakushev V. S. Semenov A. P. (2015). Study of the effect of the degree of overcooling during the formation of hydrates of a methanepropane gas mixture on the equilibrium conditions of their decomposition. Chem. Technol. Fuels Oils51, 470–479. doi: 10.1007/s10553-015-0627-4

  • CrossRef
  • Google Scholar

• 19

  Naeiji P. Luzi-Helbing M. Schicks J. M. Pan M. (2024). Dissociation behavior of sI and sII gas hydrates in response to environmental changes-Investigations on the self-preservation effect. Appl. Energy374, 124042. doi: 10.1016/j.apenergy.2024.124042

  • CrossRef
  • Google Scholar

• 20

  Naeiji P. Pan M. Luzi-Helbing M. Alavi S. Schicks J. M. (2023). Experimental and simulation study for the dissociation behavior of gas hydrates-Part I: CH₄ hydrates. Energy Fuels.37, 4484–4496. doi: 10.1021/acs.energyfuels.2c03984

  • CrossRef
  • Google Scholar

• 21

  Pan M. Naeiji P. Schicks J. M. (2023). Experimental and simulation study for the dissociation behavior of gas hydrates─Part II: sII mixed gas hydrates. Energy Fuels37 (6), 4497–4514. doi: 10.1021/acs.energyfuels.2c03985

  • CrossRef
  • Google Scholar

• 22

  Placzek G. (1934). Handbuch der Radiologie Vol. 6. Ed. MarxE. (Leipzig: Akad. Verlagsgesellschaft).

  • Google Scholar

• 23

  Prasad P. S. R. Chari V. D. Sharma D.V.S.G.K. Murthy S. R. (2012). Effect of silica particles on the stability of methane hydrates. Fluid Phase Equilibria318, 110–114. doi: 10.1016/j.fluid.2012.01.012

  • CrossRef
  • Google Scholar

• 24

  Riestenberg D. West O. Lee S. McCallum S. Phelps T. J. (2003). Sediment surface effects on methane hydrate formation and dissociation. Mar. Geol.198, 181–190. doi: 10.1016/S0025-3227(03)00100-2

  • CrossRef
  • Google Scholar

• 25

  Schicks J. M. Luzi-Helbing M. (2013). Cage occupancy and structural changes during hydrate formation from initial stages to resulting hydrate phase. Spectrochim. Acta A. Mol. Biomol. Spectrosc.115, 528–536. doi: 10.1016/j.saa.2013.06.065

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26

  Schicks J. M. Pan M. Giese R. Poser M. Ismail N. A. Luzi-Helbing M. et al . (2020). A new high-pressure cell for systematic in situ investigations of micro-scale processes in gas hydrates using confocal micro-Raman spectroscopy. Rev. Sci. Instrum.91, 115103. doi: 10.1063/5.0013138

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  Schrader B. (1995). Infrared and Raman spectroscopy: Methods and applications (Weinheim, Germany: VCH Verlagsgesellschaft mbH). doi: 10.1002/9783527615438

  • CrossRef
  • Google Scholar

• 28

  Schrötter H. W. Klöckner H. W. (1979). “ Raman scattering cross sections in gases and liquids,” in Raman Spectroscopy of Gases and Liquids, Topics in Current Physics. Ed. WeberA. ( Springer Berlin Heidelberg, Berlin, Heidelberg), 123–166. doi: 10.1007/978-3-642-81279-8_4

  • CrossRef
  • Google Scholar

• 29

  Sloan E. D. Koh C. A. (2008). Clathrate Hydrates of Natural Gases. 3rd Edition (Boca Raton, FL: CRC Press-Taylor and Francis Group LLC).

  • Google Scholar

• 30

  Song Y. Cheng C. Zhao J. Zhu Z. Liu W. Yang M. et al . (2015). Evaluation of gas production from methane hydrates using depressurization, thermal stimulation and combined methods. Appl. Energy145, 265–277. doi: 10.1016/j.apenergy.2015.02.040

  • CrossRef
  • Google Scholar

• 31

  Sun Y.-H. Su K. Li S.-L. Carroll J. J. Zhu Y.-H. (2018). Experimental investigation into the dissociation behavior of CH₄-C₂H₆-C₃H₈ hydrates in sandy sediments by depressurization. Energy Fuels32, 204–213. doi: 10.1021/acs.energyfuels.7b02949

  • CrossRef
  • Google Scholar

• 32

  Takeya S. Fujihisa H. Gotoh Y. Istomin V. Chuvilin E. Sakagami H. et al . (2013). Methane clathrate hydrates formed within hydrophilic and hydrophobic media: Kinetics of dissociation and distortion of host structure. J. Phys. Chem. C117, 7081–7085. doi: 10.1021/jp312297h

  • CrossRef
  • Google Scholar

• 33

  Tang C. Zhou X. Li D. Zhao X. Liang D. (2018). In situ Raman investigation on mixed hydrate dissociation in the presence of polyvinylpyrrolidone. Fuel214, 505–511. doi: 10.1016/j.fuel.2017.11.063

  • CrossRef
  • Google Scholar

• 34

  Truong-Lam H. S. Seo S. Kim S. Seo Y. Lee J. D. (2020). In situ Raman study of the formation and dissociation kinetics of methane and methane/propane hydrates. Energy Fuels34, 6288–6297. doi: 10.1021/acs.energyfuels.0c00813

  • CrossRef
  • Google Scholar

• 35

  Uchida T. Ebinuma T. Takeya S. Nagao J. Narita H. (2002). Effects of pore sizes on dissociation temperatures and pressures of methane, carbon dioxide, and propane hydrates in porous media. J. Phys. Chem. B106, 820–826. doi: 10.1021/jp012823w

  • CrossRef
  • Google Scholar

• 36

  Uchida T. Moriwaki M. Takeya S. Ikeda I. Y. Ohmura R. Nagao J. et al . (2004a). Two-step formation of methane-propane mixed gas hydrates in a batch-type reactor. AIChE J.50, 518523. doi: 10.1002/aic.10045

  • CrossRef
  • Google Scholar

• 37

  Uchida T. Takeya S. Chuvilin E. M. Ohmura R. Nagao J. Yakushev V. S. et al . (2004b). Dissociation of methane hydrates in sand, sandstone, clays, and glass beads: CH₄ hydrate dissociation in sediments. J. Geophys. Res. Solid Earth109:B05206. doi: 10.1029/2003JB002771

  • CrossRef
  • Google Scholar

• 38

  Wei J. Fang Y. Lu H. Lu H. Lu J. Liang J. et al . (2018). Distribution and characteristics of natural gas hydrates in the Shenhu Sea Area, South China Sea. Mar. Pet. Geol.98, 622–628. doi: 10.1016/j.marpetgeo.2018.07.028

  • CrossRef
  • Google Scholar

• 39

  Wei J. Wu T. Zhu L. Fang Y. Liang J. Lu H. et al . (2021). Mixed gas sources induced co-existence of sI and sII gas hydrates in the Qiongdongnan Basin, South China Sea. Mar. Pet. Geol.128, 105024. doi: 10.1016/j.marpetgeo.2021.105024

  • CrossRef
  • Google Scholar

• 40

  Wu P. Dong Z. Huang Z. Zhi Y. Li Y. Song Y. (2025). A novel high-pressure and low-temperature ring shear apparatus for large-scale deformation of hydrate-bearing sediment. Energy Fuels acs.energyfuels39 (39), 18870–18879. doi: 10.1021/acs.energyfuels.5c03776

  • CrossRef
  • Google Scholar

• 41

  Yang S. Lei Y. Liang J. Holland M. Schultheiss P. Lu J. et al . (2017). “ Concentrated gas hydrate in the Shenhu Area, South China Sea: results from drilling expeditions GMGS3 & GMGS4,” in Proceedings of 9th International Conference on Gas Hydrates. 25–30. Denver, CO, USA: International Conference on Gas Hydrates (ICGH).

  • Google Scholar

• 42

  Yu C. Sun B. Hasan M. Ma W. Chen L. (2024). Experimental investigation and modeling on the dissociation kinetics of methane hydrate in clayey silt cores in depressurization. Chem. Eng. J.486, 150325. doi: 10.1016/j.cej.2024.150325

  • CrossRef
  • Google Scholar

• 43

  Zhao J. Lv Q. Li Y. Yang M. Liu W. Yao L. et al . (2015). In-situ visual observation for the formation and dissociation of methane hydrates in porous media by magnetic resonance imaging. Magn. Reson. Imaging33, 485–490. doi: 10.1016/j.mri.2014.12.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar