1 Introduction

Gas hydrates are ice-like crystalline compounds having different guest molecules surrounded by water cages formed at high pressure but low temperature conditions. These hydrates look like ice; however, they have different physical and chemical properties [1]. The guest molecules could be gases, such as small molecules (e.g., CH4 or CO2) or large molecules like propane. Guest molecules are entrapped and stabilized within cages due to intermolecular forces, and cages are formed by water molecules connected to each other via hydrogen bonding. Based on three-dimensional orientation, the hydrate structure can be divided into three types: s-I, s-II, and s-H. Gas hydrate formation is a crystallization process which occurs in different stages, mainly nucleation, growth and agglomeration [2]. Gas hydrates can store a large volume of gas such that 1 m3 of hydrate volume can store up to 163 m3 of gas. Gas hydrates also offer additional advantages over conventional gas storage technologies, such as being environmentally friendly, non-explosive, and low mantainence. Gas hydrates applications can be divided into four categories with respect to their origin and application, represented in Fig. 1.

Fig. 1
figure 1

Role of surfactant in different gas hydrate based applications

Full size image

Gas hydrates are found in nature within sediments and are considered a source of gas supply for future generation. These untouched gas hydrate reservoirs are in the continental shelf in the marine environment as well as cold regions on land, such as permafrost regions in Alaska, Canada, China, and Russia. It is estimated that these reservoirs contain 1.5 × 1016 m3 of gas [3, 4]. Hydrates are also believed to present on other planets, including Mars and Saturn [5]. Gas hydrates formation in oil and gas pipeline is considered an engineering challenge in the petroleum industry as hydrate formation could block and damage the pipelines [6, 7]. Studies in this context are focused on preventing hydrate formation by injecting chemicals known has hydrate inhibitors. Based on the mechanism, these inhibitors are categorized as high dosage thermodynamic hydrate inhibitors (THI, 20–40 wt%) and low dosage hydrate inhibitors (LDHI, 0.1–5 wt%). Due to environmental impact consideration, LDHI is used more frequently than THI [8]. The most common surfactants used under LDHI type are quaternary ammonium surfactants characterized having n-butyl or n-pentyl functional groups and long alkyl chains [9].

Opposite to hydrate inhibitors, a different class of chemicals, known as promoters, improve formation kinetics needed in different industrial applications [10], such as natural gas storage [11], hydrogen storage [12], hydrate-based pre- and post-combustion CO2 separation, capture, storage, transport [13,14,15], and hydrate-based desalination [16].

During laboratory-based hydrate studies, gas hydrate is formed at gas-water interface as gas solubility in water is slow and gas saturated water is crystallization process is very slow to happen. A thin hydrate layer forms first at gas-liquid interface which later grows towards the gas phase; however, growth is limited by mass transfer of gas molecule through the thin hydrate layer [17]. To move toward commercialization, the kinetics of gas hydrate formation require drastic improvement. This could be achieved through mechanical techniques as well as adding surface-active chemicals that could reduce the mass transfer barrier and enhance the gas-liquid contact area without changing the hydrate phase equilibrium [18]. These surface-active agents are used in small concentrations and known as hydrate promoters. Surfactants [19] and hydrophobic amino acids [20] are known to be hydrate promoters. Recently low dosage methanol is also considered as hydrate promoter due to its near similar behaviour as a surfactant at low concentration [21].

This chapter is focused on discussing the role of surfactants as surface-active agents during gas hydrate formation and dissociation. Formation kinetics depends on guest molecule, pressure, temperature, and reactor design. Surfactant performance is also system-dependent, including the difference in reactor design, pressure, and temperature conditions, the difference in hydrate forming gas mixture as well as the role of supporting material due to difference in thermal conductivity and surface-to-volume ratio [22,23,24]. In this chapter, we discuss the available mechanisms and current status of surfactant application in gas hydrate management briefly.

2 Role of Surfactant Molecular Structure on Hydrate Promotion

Surfactants are known to affect the kinetics of the hydrate formation. Kinetics of hydrate formation can be divided into different stages, starting from dissolution, nucleation, growth, and agglomeration [1]. Surfactants facilitate faster nucleation by reducing the surface free energy by absorbing into the aqueous-hydrocarbon (gas/liquid) interface [25]. Surfactants also enhance the mass transfer by improving the hydrocarbon solubility into the water. Surfactants play an essential role at the gas-liquid and liquid-hydrate interface. A surfactant could occupy the area at the interface and could also hinder hydrate formation. In this context, sodium dodecyl sulfate is considered to be the most effective surface-active agent to enhance nucleation and growth phases [26,27,28]. Surfactants are composed of hydrophobic “tails” and hydrophilic headgroups within a single molecule, which determine the surfactant properties. Surfactant molecules could diffuse from the bulk phase to the liquid-gas interface, such that the hydrophilic end stays in the liquid phase while the hydrophobic end stays in the gas phase. This would lead to a change in surface tension, modification in contact angle, and change in surface charge and surface viscosity [29]. At a given concentration, surfactant molecules bind together in different shapes and orientations, called micelles [30]. Above the critical micelle concentration (CMC), the hydrophilic part covers the hydrophobic group and supports the organic compound (methane or CO2) solubility [31]. Surfactants are well used chemical substances to enhance surface activity that control spreadability, wetting, foaming, etc. [19]; however, it is still unclear about the key mechanism responsible and the role of concentration during formation mechanism. Insights into the molecular structure of surfactant can enhance our understanding of its role in hydrate formation kinetics. The key component of the surfactant structure is given in Fig. 2.

Fig. 2
figure 2

Role of molecular structure of surfactant optimal efficiency during gas hydrate formation

Full size image

2.1 Ionic Type

Surfactants tested for hydrate formation are from anionic, cationic, and nonionic natures. Studies confirmed that anionic surfactants produced better promotion compared to cationic and nonionic ones at lower concentration (100–500 ppm). At higher concentration (>800 ppm), the difference in promotion ability decreased [32, 33]. Among three homologuous anionic surfactants, Sodium dodecyl sulfate (SDS), Sodium tetradecyl sulfate (STS) and Sodium hexadecyl sulfate (SHS), it was found that SDS is most effective above 1000 ppm concentration for methane hydrate formation while STS has shown same promotion behaviour at 100 ppm [34]. SHS was not effective compared to SDS and SHS. In another study on the comparative performance of different surfactants (anionic/cationic/nonionic) during CO2 hydrate formation, anionic surfactant SDS was the most effective among all three. Nonionic surfactants are more effective compared to cationic surfactant [35]. In another study, when n-dodecyltrimethylammonium chloride (DTAC) and Tween 20 were used for the Tetrahydrofuran (THF/H2) system and THF/Methane system, both surfactants showed different behavior. This led to the conclusion that the role of surfactant during hydrate formation is dependent on guest molecules as well as the system (single guest molecule vs. mixed hydrate) [36]. Figure 3 provides the details of the key surfactant in each category. Apart from traditional classification, novel surfactants, including biosurfactants and Gemini surfactant, have also been used in hydrate-based studies [37,38,39]. Figure below describe some popular surfactants used frequently in gas hydrate studies.

Fig. 3
figure 3

Popular surfactans used in gas hydrate studies

Full size image

2.2 Properties of Hydrophobic Group

Size of the hydrophobic and hydrophilic groups controls the surfactant properties, such as interfacial tension. Large groups have lower interfacial tension than smaller groups [40, 41] Properties of the hydrophobic group affect the surfactant properties and are of greater research interest [42]. Kumar et al. [19] have provided a summary of the properties of the hydrophobic group and their effect on the surfactant properties. Three key elements include the change in the length of the hydrophobic group, branching, and unsaturation in the hydrophobic group and the presence of an aromatic nucleus in the hydrophobic group. The hydrophobic group-controlled solubility of surfactant in water and organic solvent, biodegradability and packing of surfactant at the interface. With an increase in the length of the hydrophobic group, the solubility of surfactant in water decreases, but in organic solvent it increases. Apart from that, biodegradability and surfactant absorption at the interface also increases as the length of the hydrophobic part increases [19]. Okutani et al. [34] studied the effect of alkyl chain length, using three surfactants (SDS, STS, and SHS) having the same headgroup (-OSO3-Na+) but different carbon chains (12, 14, or 16, respectively). They concluded that surfactant with larger carbon number could be useful even at lower concentrations. On the other hand, Dicharry et al. [43] have tested the effect of carbon chain for sulphate-based surfactants and found out that higher carbon chain-based surfactants could readily absorb on hydrate surfaces by forming hemimicelles, hence promoting hydrate formation. The difference in carbon chain length creates different solubility and packing area. By looking at the available research, it can be concluded that sulphate-/sulfonate-based surfactants have shown the best promotion capabilities with 12–14 carbon chain as an optimal solution. As per adsorption and mass transfer theory, an increase in chain length could decrease the surfactant hydrate promotion efficiencies.

3 General Theories Behind the Surfactant-Based Promotion

Kalogerakis et al. [18] were one of the first to study the role of surfactants during hydrate formation. During nucleation, hydrate film formed at the gas-liquid interface, which further isolates gas phase from the liquid phase and allows only gas molecule reaching to liquid phase through diffusion. The exact mechanism behind the role of surfactant during hydrate formation is not yet agreed upon. Many theories have been proposed. In the following section, we have discussed some well-known theories available in the literature and describe in the Fig. 4 [31].

Fig. 4
figure 4

Surfactant based key hydrate formation theories

Full size image

3.1 Micelles Formation Theory

In one notable research proposing micelles theory, critical micelle concentration (CMC) was reported as 242 ppm using SDS during natural gas hydrate formation [28]. Experiments also suggest that the CMC value of natural gas-water solution decreases as the pressure decreases [28]. Other studies suggested that when surfactant SDS concentration was above the CMC, a 700-fold increase in the rate of hydrate formation was observed due to enhanced guest molecule solubilities [44]. Change in gas solubility due to change in surfactant concentration was also measured for the ethylene and SDS system, both at ambient pressure and under hydrate formation conditions [45]. Solubility behavior for ethylene and methane was similar under hydrate formation conditions in the presence of SDS [46]. Presence of CMC for SDS during methane hydrate formation was also confirmed [47]. The CMC-based hydrate formation mechanism is explained in the figure below. However, some researchers have disputed this theory, and citing that a decrease in the rate of methane hydrate formation above CMC was also observed for cationic and anionic surfactants [48, 49]. Few studies also suggested that SDS at even very low concentration (10 ppm) could promote methane hydrate formation [17] or a single molecule itself can assist in hydrate formation [50]. In another study, it was observed that the CMC value of surfactant does not depend on the hydrate formation conditions and remains unchanged [51].

3.2 Capillary Driven Growth

Presence of capillary action during hydrate formation was demonstrated by Watanabe et al. using SDS and Difluoromethane (HFC-32) gas [52]. Visual observation suggested that crystals were initially formed both at the liquid-gas interface as well as reactor sidewall at different spots. Thereafter, when crystals grew in size, coalesence took place with each other, and crystals started to grow at the reactor sidewall. After that, hydrate grew in a downward direction in the solution phase to maintain contact and cause a decrease in the solution surface. Additionally, rippling motion of liquid at the reactor wall also confirmed the presence of capillary action during the formation. Other studies also confirmed moving of gas-liquid interface in upward direction along the reactor wall in the presence of SDS and did not change the hydrate thermodynamics [17, 53]. This behaviour was not observed in the absence of surfactant for the pure water-gas system.

The capillary mechanism is said to be caused by either change in hydrate morphology [54,55,56] or due to change in the wettability [57,58,59,60]. Wang et al. have used anionic surfactants SDS, Sodium dodecyl benzene sulfonate (SDBS) and Sodium dodecyl sulfonate (SDSN) having different wettability on the reactor side and found that SDBS performed poorly compared to other surfactants due to weaker wettability on the reactor wall [60]. Wang et al. [59] have also achieved directional hydrate growth by controlling and varying the wettability of the solid surface inside the glass tube. Wettability of surfactant solution is also controlled by surfactant concentration. SDS wettability also changed due to the difference in its concentration [57]. NMR and Raman studies also confirmed that SDS has two different growth mechanisms at 25 ppm and 500 ppm. At higher concentration, water converted into an intermediate solid-state and then combined with methane gas. This was not observed at the lower concentration [61].

The key limitation of this theory is that most of the observations regarding the capillary mechanism are for anionic surfactants, including SDS, STS, and SHS [34, 53] and for specific guest molecules, methane and ethane. In the presence of the CO2 molecule and an anionic surfactant such as SDS, no capillary mechanism was observed at high driving force [62]. Other studies involving various surfactants, such as lithium dodecyl sulfate (LDS), dodecyl benzene sulfonic acid (DBSA), sodim oleate (SO), dodecyl alcohol ethoxylates (AEO), cetyltrimethylammonium bromide (CTAB) from cationic and non-ionic categories, did not confirm the capillary mechanism responsible for hydrate growth [63,64,65]. Therefore, it can be concluded that capillary-based hydrate growth is dependent on surfactant type, surfactant concentration, and guest molecules. For example, Molokitina et al [62] performed microscopic investigation of the CO2 hydrate formation mechanism in bulk water phase in the presence of SDS under different mass transfer barrier and visualized the hydrate growth pattern as shown in Fig. 5. It can be seen that driving force changes the mechanism of gas hydrate formation at the gas/liquid interface, such that capillary-driven movement is observed at low driving force. Further research is required to understand the factors influencing the capillary mechanism and how this mechanism could be improved to achieve enhanced growth.

Fig. 5
figure 5

a CO2 hydrate film propagation along the gas-liquid interface (capillary-driven) in the presence of SDS (1000 ppm) and low mass transfer driving force. b CO2 hydrate film propagation along the gas-liquid interface on both gas and liquid sides under high mass transfer driving force (not capillary-driven) in the presence of SDS (1000 ppm) [62]

Full size image

3.3 Adsorption Theory

Theory of surfactant adsorption on the hydrate surface was first proposed based on zeta potential measurements [66, 67] and is proposed as a cause behind improved hydrate formation kinetics [26, 66, 68]. Different types of surfactants were absorbed through a different mechanism and due to difference in surfactant concentration. For example, an ionic surfactant is considered to absorb on the hydrate surface under the influence of electrostatic forces while a non-ionic surfactant adsorbs through hydrogen bonding [69]. At concentrations below the CMC, adsorption behavior follows Henry’s law, confirmed by Scamerhorn et al. [70]. Above the CMC, surfactants form hemimicelles which are an aggregate form of surfactant due to tail-tail interactions between surfactant molecules. Also above the CMC, adsorption is independent of the concentration [71].

3.4 Interfacial Tension and Adhesion Energy

This theory is based on the few studies that suggest that during the gas hydrate formation, the contact angle between the liquid-gas interface and solid-state change in the presence of the surfactant [53, 72]. Addition of surfactant decreases the surface tension of the aqueous phase, thus decreasing the contact angle. This causes creation of a film-like interface along the reactor wall and becomes a preferred location for hydrate nucleation and growth [53]. In the presence of surfactant, the solid surface becomes more water-wet due to the decrease in surface tension [72]. Song et al. suggested that the reduction in contact force and interfacial tension due to the presence of surfactant caused enhance hydrate growth [73].

3.5 Zeta Potential Measurement

Zeta potential measurement was used to explain the synergy between THF and SDS as reported by Torre et al. [74] for gas hydrate based CO2 capture studies. In another study, Torre er al. [75] has suggested no mass transfer barrier was observed during the gas to liquid mass transfer in the presence of both SDS and SDS+THF. Zeta potential measurement has suggested that in the mixture of SDS and THF, THF hydrate stays dispersed due to electrostatic repulsion caused by adsorption of DS- anions on the hydrate surface hence porous texture allow CO2 diffusion into the liquid phase [67].

During our recent study focused on kinetics of methane hydrate formation in the presence of SDS and effect of SDS concentration on the formation kinetics. We observed change in formation kinetic behavior around 2000–3000 ppm concentration. To suggested that behavior can be explained in terms of dual effect of absorption and surface tension. Trend in key kinetic properties as a function of SDS concentration is illustrated in Figs. 6 and 7 [47].

Fig. 6
figure 6

a Key kinetic properties trend during methane hydrate formation as a function of SDS concentration (500-3000 ppm). Experimental observation confirm the presence of CMC between 2000–3000 ppm [47]. b Effect of change in SDS concentration on the absorption and surface tension (500–3000 ppm) [47]

Full size image
Fig. 7
figure 7

Different association of surfactants as copromoter

Full size image

4 Role of Surfactant with Co-promoting Gas Hydrate Formation

Surfactants have also played a key role as co promoter along with different agents and medium to enhance and stablise hydrate formation. Figure below describe their key associations.

4.1 Microemulsions

Surfactants also serve as supports for other systems that could achieve rapid hydrate formation, including microemulsions [76, 77], dry water, and nanofluids. In water-in-oil emulsions, methane molecules disperse into the oil phase and later diffuse and reach the water droplet surface to react. Each water droplet serves as an isolated nucleation front, thus able to create highly efficient water-to-hydrate conversion. Many factors, such as pressure, temperature, stirring, and initial water volume, control the stability and droplet size of the emulsion. Water cut controls not only the gas-liquid contact interface but also controls the water droplet size [78,79,80,81]. There are some challenges, such as hydrate film formation at the water droplet surface could retard gas transfer into water droplet, as well as formation and dissociation cycles could potentially destabilize the hydrate [82, 83].

4.2 Nano-Fluids

Nano-fluids are seen as the potential alternative to accelerate gas hydrate formation due to their superior thermal conductivity that addresses the exothermic and thermal inhibition during formation [84, 85]. Among different nanofluids, metal nanofluids such as silver nanoparticles have been tested repeatedly. Silver nanoparticles have been used to study methane and ethane hydrate nucleation and improvement in gas uptake, and the induction time was recorded. It is suggested that nanoparticles help during formation by providing nucleation sites, reducing effective interfacial tension and wetting angle between hydrate and particle [86,87,88]. A key challenge during the application of nanoparticles for gas hydrate formation and dissociation is their stability during cycling. This could destroy nanoparticle stability and, thus, reusability. To stabilized nanoparticles, surfactants such as SDS are also used. SDS in low concentration has been used along with Al2O3-, ZnO-, CuO-based nanoparticles [89,90,91].

4.3 Thermodynamic Promoters

During the last decade, much attention has been placed on non-mechanical techniques to improve gas uptake and reduce nucleation time and the stochastic nature of hydrate nucleation and formation. These non-mechanical techniques consist of using chemicals categorized into kinetic and thermodynamic promoters. Thermodynamic promoters allow hydrate formation at moderate pressure and temperature condition by reducing the required formation pressure and increasing the temperature, making them sought for novel hydrate-based industrial applications. Frequently used thermodynamic promoters include tetrahydrofuran (THF), Tetra n-butyl ammonium bromide (TBAB), and cyclopentane (CP). The key disadvantage of using thermodynamic promoters includes loss of gas uptake due to the occupancy of the cages by the promoter molecule and slower formation kinetics. To overcome these challenges, surfactants such as SDS have been used along with thermodynamic promoters, including THF and CP, and have received greater attention due to more efficient performance compared to thermodynamic promoters.

Kumar et al. [92] have studied the role of SDS and THF on formation and dissociation kinetics of the methane hydrate in an unstirred reactor configuration. Their results concluded that SDS and THF could provide optimal configuration for methane hydrate storage and transportation at atmospheric pressure. Veluswamy et al. [93] used SDS (100 ppm) with THF (5.6 mol%) to achieve methane hydrate formation within 1 h at ambient temperature 293.2 K and 7.2 MPa. They also highlighted the synergetic effect between SDS and THF was visible only at ambient temperature. When the temperature was reduced to 283.2 K, the addition of SDS decreased the gas uptake by 20% and 60% at 72 bar and 30 bar, respectively. Mech et al. [94] also studied SDS along with THF and TBAB at 276.15 K and different pressures from 30 bar to 75 bar in a stirred tank reactor. They concluded that SDS at 600 ppm concentration, THF plus SDS had higher uptake at 75 bar while TBAB plus SDS had higher gas uptake at 30 bar. Kakati et al. [95] had tested the mixture of SDS and THF on the mixture of methane, ethane and propane and found that a THF and SDS combined system can be used to improve formation and thermodynamics of the natural gas storage in the form of the hydrates.

Additionally, SDS is used with another thermodynamic promoter in applications such as gas separation, CO2 capture, desalination, and hydrogen energy storage, and it is further discussed in §5.

4.4 Porous Medium

Pan et al. [96] have discussed the presence of a porous medium and SDS on formation kinetics. Presence of porous medium improve the heat transfer as well as greater higher surface area leads to quicker 2D nucleation. It is usually challenging to form methane hydrate in the porous medium as a porous medium act as a thermodynamic inhibitor and controlled by the pore size and salinity [97]. It is general practice to add surfactant to enhance the kinetics of hydrate formation [17] however, the surfactant may influence wettability as well as can create an air bubble during gas injection; therefore, the surfactants have been ignored traditionally during the modelling of the hydrate formation process. Presence of surfactant leads to enhanced formation rate and much higher gas uptake. Results suggest that both particle size and water saturation play an important role in the hydrate formation kinetics. Particle size control interface-specific area, pore volume and pore size distribution whereas water saturation control water migration as well as hydrate distribution which intern controls formation and dissociation kinetics [96]. Addition of the surfactant in the liquid phase lowers the mass transfer resistance at the gas-liquid interface and reduce the surface tension. SDS, in particular, found effective because of hydrophobic active groups in the ionized SDS which would accumulate around the surface of the silica sand to avoid any contact with water in the solution. Many surfactant micelles are formed that further help methane dissolves more due to micelle solubilization [98]. In another study for SDS and water system, the presence of SDS increase the ethane hydrate dissociation rate and weaken the self-preservation tendency of the gas hydrate [99]. Some of the key research results are summarized in Table 1.

Table 1 Key experimental work focused on the kinetics of hydrate formation in the presence of porous media and surfactant
Full size table

Amino acids are seen as eco friendly replacement of surfactant for gas hydrate based research and hydrophobic amino acids are seen as good alternative of SDS. Figures 8 and 9 compare the kinetics of methane hydrate formation between SDS and Amino acids at same concentration (3000 ppm) in different sands with four different sand particle sizes.

Fig. 8
figure 8

Pressure variation during isothermal experiments at 100 bar and 1℃ starting condition comparing the methane hydrate formation kinetics for SDS and four selected amino acids (L-valine, L-methinonine, L- histidine, L-arginine). Results suggest that SDS and L-methionine hydrate promotion capabilities are near similar in porous medium with different physical properties [104]

Full size image
Fig. 9
figure 9

Pressure variation during the isothermal experiments at 100 bar and 1℃ comparing the methane hydrate formation kinetics for SDS and four selected amino acids (L-valine, L-methinonine, L- histidine, L-arginine) at given type of sand. Results indicate the hydrate formation rate increase as sand particle size increases in the presence of SDS and hydrophobic amino acids [104]

Full size image

5 Application of Surfactant

Surfactants have been used as both as inhibitors as well as promoter in different industries and application and their usage is summarize in Fig. 10.

Fig. 10
figure 10

Usage of Surfactant in different industries and applications

Full size image

5.1 Surfactant-Based Hydrate Inhibition

Quaternary ammonium surfactants in low dosage amount are used as anti-agglomeration (AA) compounds for hydrate inhibition [9] to avoid hydrate plug formation in oil and gas pipelines. AA compounds form a well-structured thin layer that further slows down hydrate aggregation and stops the crystal growth process due to mass transfer barrier [105, 106]. The thin layer could be formed either between water and oil or oil and hydrates and containssurfactants and alkanes from the oil phase [107,108,109,110]. Apart from ionic surfactants, Sorbitan type Span-20 to Span-80 are also used as AA compounds [111]. When used along with thermodynamic inhibitor (MeOH) or salts in the aqueous phase, they improve the inhibition efficiency [112, 113]. Increase in salinity increases ionic AA inhibition efficiency without disturbing emulsion stability [114]

Molecular simulation shows that AA could also promote hydrate growth [115] which could be used in hydrate based novel application such as natural gas storage and transportation, desalination, and other emerging applications.

5.2 Natural Gas Storage and Transportation

Most crucial industrial development is in the field of natural gas (NG) storage and transportation, as gas hydrates offer additional benefits compared to traditional technologies like liquidified natural gas (LNG) and compressed natural gas (CNG). Methane gas hydrates offer high gas storage capacity in a solid-state, such that 170× methane gas per volume of hydrates (V/V) can be stored at moderate low-temperature (260–270 K) at atmospheric pressure due to self-preservation tendency shown by pure methane hydrates [116]. The storage and transport of NG in solid form could be alternative to CNG [11]. In this regard, hydrate pelletization technology has been evaluated for the storage and transportation of NG [117, 118]. Stability was further improved in the presence of SDS [119] and maintained for 256 h with decomposition measured to be only 0.04% at 1 atm and 268.2 K. This ultrastability was caused due to different hydrate morphology in the presence of surfactant. Therefore, gas hydrates at subzero temperature show reduced gas leakage and offer added advantages compared to CNG and LNG transportation.

5.3 Hydrate Based Desalination and Produced Water Treatment

The feasibility of produced water and seawater treatment via gas hydrate formation was first demonstrated during the 1960 s [120]. This technique is based on the fundamental understanding of gas hydrates, that the chemical structure of gas hydrates includes only water and host molecules and excludes all salts and other impurities in unreacted water [121, 122]. Hydrate-based desalination has shown better efficiency compared to traditional desalination techniques, such as reverse osmosis and multistage distillation, at higher salinity levels up to 25%. Such high salinity in produced water has been reported in shale gas projects and CO2 injection-based projects. Due to several reasons, including energy consumption, technology immaturity, low efficiency, hydrate-based technology was not used in desalination plants [123]. Recent studies using cyclopentane (CP) have shown promising results to be considered for desalination studies as CP forms hydrates with pure water under atmospheric pressure at 7 °C and is immiscible with pure water; therefore, it can be recycled after hydrate dissociation [124,125,126]. When water and CP come together, an emulsion is formed, and the use of CP is advantageous as it can be recovered at the end of dissociation. In the presence of promoter the hydrate formation rate, gas uptake, water recovery, and salt rejection improve. To make desalination more attractive, it is also suggested to combine cyclopentane with another guest molecule-based hydrate application, such as gas separation or gas capture to optimize energy consumption and improve the salt removal efficiency [127, 128]. Using surfactant with CP can bring more drawbacks than an advantages because its presence makes the hydrate former difficult to separate after dissociation [129]. Erfani et al. [130] studied the effect of 14 nonionic surfactants on the formation kinetics of CP hydrate and found that presence of surfactant decreased induction time and enhanced the hydrate formation rate. The surfactant, which generates an oil-in-water emulsion, performed better than water-in-oil emulsion. Lim et al. [24] found that SDS changes the CP hydrate morphology, which includes rectangular tree-like or fiber-like crystals, and no change in CP hydrate shell thickness was observed in the presence of surfactant [131]. To summarise based on results, key surfactants tested along with CP hydrate for desalination include LAE8EO, TritonX-100, NPE6EO, SDS, Dodecyl-trimethylammonium bromide (DTB), Span-20, DDBSA (Dodecyl Benzene Sulfonic Acid) and Tween 20 when used at a suitable concentration. Many surfactants, when added, modify CP hydrate morphology and physical properties. In the presence of surfactant CP solubility also enhances in water, and, hence, the removal of surfactant from the water at the end of dissociation is required as surfactants traditionally are toxic and not environmentally friendly. Use of bio-surfactants could be recommended in this application as they are biodegradable [37, 38, 132]

5.4 Hydrate Based CO2 Separation, Capture and Storage

CO2 separation, capture, and storage are important technology considerations to reduce greenhouse gas emission from industrial plants, including chemical, power, cement, etc. This technology includes pre and post-combustion CO2 capture. A pre-combustion gas mixture contains a CO2/H2 gas mixture, also known as fuel gas [14], while the post-combustion gas mixture includes CO2/N2 mixture known as flue gas. Hydrate-based CO2 capture is proposed as a novel technique for CO2 separation from fuel and flue gas mixtures [133,134,135,136]. The difference in CO2 concentration in hydrate and in vapor phase acts as the main driver to separate CO2 from the gas mixture during hydrate-based separation [15]. The key thermodynamic promoter used to achieve moderate operating conditions includes tetrahydrofuran (THF), tetrabutylammonium bromide (TBAB), tetra-n-butylammonium nitrate (TBANO3), tetrabutylammonium fluoride (TBAF), and dodecyl trimethyl ammonium chloride (DTAC). Among these, THF is the most extensively used thermodynamic promoter for CO2 separation and capture from the gas mixture.

Thermodynamic promoters are able to lower operational pressure or increase the temperature but do not have the influence of kinetics of the hydrate formation, which is essential for commercialization. Therefore, kinetic promoters including surfactants like SDS and SDBS (sodium dodecylbenzene sulfonate) have been used extensively along with thermodynamic promoters [137]. Among all tested surfactants, SDS was the most efficient. The CO2–water system in the presence of SDS has shown lower induction time for CO2 hydrate formation. CO2 solubility increases in the presence of SDS due to a decrease in surface tension at the liquid-gas interface. Higher solubility causes faster nucleation and a further decrease in induction time [19, 62, 138] The growth rate is highest at 500–1000 ppm concentration, and higher concentration does not improve the growth rate and gas uptake [20]. In another study, it is suggested that SDS concentration has no effect on gas separation efficiency and only affects the rate of hydrate formation [139, 140]. When SDS is used with cyclopentane, no improvement in the kinetics of CO2 hydrate formation are observed [136]. Presence of SDS during CH4-CO2 hydrate swapping can enhance CO2 storage into methane hydrate reservoirs without disturbing geological formation [141] (Figure 11).

Fig. 11
figure 11

Change in hydrate morphology in the presence of SDS 500 ppm in bulk water. Change in morphology is recorded before and after CO2 injection into methane hydrate. Pictures also show the porous methane hydrate morphology. Methane hydate formed are porous in nature (Fig. 1). Figures 2–5 shows the morphgology change after CO2 injection and 72 h after CO2 injection [141]

Full size image

5.5 Hydrate Based Hydrogen Storage

Application of hydrogen hydrates for stationary hydrogen storage has not taken off as hydrogen hydrates are formed at very high pressure (at the scale of GPa) at given ambient temperature. Research is focused on the use of kinetic and thermodynamic promoters and co-guest molecules to achieve moderate operating condition as well as faster formation kinetics [12, 142,143,144]. Some of the thermodynamic chemicals being tested repeatedly include TBAB, TBANO3, THF, and CP [145,146,147,148]. Thermodynamic promoters occupy cages and reduces the hydrogen storage volume within hydrate. Apart from low hydrogen storage, low formation kinetics and risk of hydrogen diffusion through cages hinder adopting hydrate-based hydrogen storage methods at commercial scale [149,150,151]. Some researchers have tried to improve the hydrogen storage efficiency through different techniques [152,153,154]; however, there less attention is given to kinetics improvement, and very few studies have discussed the role of surfactant during hydrogen hydrate formation. SDS was found to be effective at the small concentration (5–500 ppm) during mixed hydrogen/propane hydrate formation studies, and two-stage hydrate growth was observed [155], a significant finding showing that micelles are unnecessary to impact hydrate formation (Fig. 12). Profile et al. [156] have invented a new technology with the help of aerosol OT surfactant(AOT), THF, and water and with the use of nanotechnology and isooctane.

Fig. 12
figure 12

Evidence of hydrate nucleation and growth at low concentration of SDS (25 ppm) within a mixed hydrogen−propane hydrate system performed at 274.2 K and 8.5 MPa in a stirred tank reactor. Reprinted from [156]

Full size image

5.6 Drawbacks of Surfactants

Surfactants have been studied extensively as kinetic promoters for hydrate formation; however, few studies have discussed the disadvantages of surfactants. The key disadvantage is that surfactants create foam even at low concentration (100–1000 ppm) during the degassing operation [157,158,159]. Due to foam formation, gas production rates can be very slow,which could be undesirable for industrial-scale applications. There are also concerns about surfactant biodegradability and their effect on environment [160]; therefore, current research is focused on environmentally friendly substitutes, such as amino acids [20] or new classes of biosurfactants [38] having similar kinetic promotion capabilities as anionic surfactants.

6 Computational Studies on the Role of Surfactant During Hydrate Formation

The first attempt to understand the blockage of pipelines due to gas hydrates has been by Hammerschmidt [6] in 1934. Among the first additives used to prevent the formation of gas hydrate included anti-freezing agents, such as methanol or ethylene glycol (EG). [161] Their effect in the solution resulted in a shift toward lower temperatures and higher pressures, due to a leftward change in the equilibrium phase boundary conditions. This came about because the hydrogen bonding between water and additive molecules affected the activity of water and the propensity to form hydrate cages [162], giving rise to the class of chemicals that are known as thermodynamic inhibitors. Such inhibitors have become commonplace in the oil and gas industry as preventative measures for gas hydrate formation and consequent pipeline blockage [163]. Conversely, there are chemicals that instead improve the formation behavior of hydrate by causing a shift of the phase boundary to the right, and they are thus known as thermodynamic promoters. These promoters are typically trapped in the hydrate cage along with the gas molecules, aiding in the stabilization of the hydrate crystalline structure at higher temperatures and/or lower pressures.

Thus, it is very valuable for oil and gas industries to be able to predict which additives and the correct amount that would be necessary to remove the risk of pipeline blockage resulting from gas hydrates. Hammerschmidt developed an early formula, considering the temperature that would suppress hydrate formation because of the inclusion of inhibiting chemicals, \( \Delta T_{H} \), as described in the work [6].

$$ \Delta T_{H} = \frac{{k_{H} w_{\text{add}} }}{{M_{\text{add}} \left( {1 - w_{\text{add}} } \right)}} $$

Here, \( k_{H} \) is a dimensionless constant that depends on the type of inhibitor, \( w_{\text{add}} \) is the mass fraction additive in aqueous solution, and \( M_{\text{add}} \) is its molecular weight. The equation is fairly simple and has relatively good accuracy, which makes it popular even to this day, despite more advanced models that have been developed since the 1950 s. However, because the accuracy of suppression temperature depends on estimation of the hydrate equilibrium temperature considering purity of samples, there is a higher chance of errors in calculation. In comparison, thermodynamic models primarily depend on the chemical potentials of each chemical in every phase being equal. Thus, thermodynamic models make it possible to include additives into calculations, insofar as parameters are provided for predicting their chemical potentials.

In the last decades, there has been a larger prevalence of first principles-based modeling since computational power has greatly increased over that time. This has made it possible to predict the properties of long chain molecules, typical of surfactants, ionic liquids, or amino acids, through methods like density functional theory (DFT). DFT typically has \( N_{e}^{2 - 3} \) scaling, where Ne is the number of electrons, which is why it has been a more recent endeavor for such molecules. More specifically, much research has been performed to understand the mechanisms of thermodynamic and kinetic hydrate inhibition (THI and KHI, respectively) arising from gas hydrate additives, both from experimental and computational aspects [164,165,166,167]. Very recently, Lee et al. were able to find synergistic effects when more than one inhibitor is utilized through a combination of both methods [168]. Statistical thermodynamics has gained much traction in the past years, since it can incorporate DFT-calculated properties into its calculation routine (e.g. the conductor-like screening model with real solvation, COSMO-RS) [169] to give information like reaction constant, activities, and Henry constants, to name a few [170,171,172].

Further implementation of first principles methods has been seen in the realm of artificial intelligence, as machine learning methods have been able to make predictions about properties using training sets with both experimental and computational data. The advantage of machine learning is that it can provide savings in computational time, given a large enough training set to give accurate results. Going into the future, as large scale operations will begin to implement gas hydrate production for various applications, computational fluid dynamics (CFD) and plant-scale simulations will benefit from incorporating these into a multi-scale approach, since each of these techniques span different time- and length-scales (see Fig. 13). Already, there has been some CFD simulations studying hydrate formation, and inclusion of additives will become increasingly prevalent.

Fig. 13
figure 13

How multi-scale modeling could link methods of different length- and time-scales

Full size image

Some key features of each of the methods above can be summarized in Table 2. In the following sections, the atomistic to micro-scale methods described above are presented with examples.

Table 2 Key usages and applications of the different computational methods
Full size table

6.1 Empirical Modelling—The Electrolyte Model

One particularly important application of models and property prediction, prior to the use of computationally-assisted models, pertains to equilibrium phase diagrams for gas hydrates and their additives, and the most common method of doing so—with good accuracy—has included estimated values for the activity of water as a function of additive. Dickens and Quinby-Hunt developed an electrolyte-based model [173] that could generate the equilibrium points for dissolved salts as additives, although it could be extended to ionic liquids and surfactants that can be viewed as separated charges [174], as has been proposed by Nashed et al. [175]. regarding ionic liquids as additives for methane hydrates. The electrolyte-based model is an adaptation of Pieroen’s model [176], and has been utilized by many other works [177,178,179,180]. These models are based on classical thermodynamics, so assumptions, like negligible amount of gas in the hydrate and hydrate dissociation enthalpy \( (\Delta H_{\text{dis}} ) \) is constant over a small temperature range, are made, showing that additives decrease water activity \( (a_{w} ) \). Thus, Nashed et al. made the following relationship between hydrate formation temperatures of pure water and the additive, \( T_{w} \) and \( T_{\text{add}} \), respectively [175]:

$$ \ln a_{w} = \frac{{\Delta H_{\text{dis}} }}{nR}\left( {\frac{1}{{T_{w} }} - \frac{1}{{T_{\text{add}} }}} \right) $$

where n is methane hydrate hydration number [181], and R is the universal gas constant. CSMGem software was used to calculate the water- methane hydrate dissociation temperature. The activity is also related to the change in freezing point due to additive, \( T_{f} \) and \( T_{{f,{\text{ice}}}} \), for with and without additive, respectively, and the heat of fusion of ice:

$$ \ln a_{w} = \frac{{\Delta H_{{{\text{fus}},{\text{ice}}}} }}{R}\left( {\frac{1}{{T_{{f,{\text{ice}}}} }} - \frac{1}{{T_{f} }}} \right) $$

Combining the two equations above, one would be able to determine the effect of additives on the hydrate formation temperature.

In the work by Nashed et al. they were able to determine the phase boundaries of methane hydrates in the presence of 10 wt% ionic liquid solution experimentally, and found good agreement with the models above [175]. It was found that their impact caused a shift toward lower temperature and higher pressure, and a mean temperature reduction of 0.37–1.52 K was observed within their pressure range (5.1–11.1 MPa), depending on the type of ionic liquid used. More importantly, they drew relationships between the inhibition effect with chemical structure (e.g. cation/anion combinations and alkyl chain length), leading them to conclude that thermodynamic inhibitors are not involved in the formation of hydrate cages.

6.2 Quantum Mechanics

The presently most popular implementation of computational quantum mechanics is density functional theory (DFT), although others are used to lesser degrees, like configuration interaction, coupled cluster, and increasingly Møller-Plesset perturbation theory. One of the basic types of calculation that can be performed with DFT is the determination of the relaxed structure of a molecule and its corresponding energy. Using this information, one could, for example, use DFT (and, similarly, moleculdar dynamics) to determine the interaction energy, \( E_{\text{int}} \), of a gas hydrate molecule with an inhibitor, given the gas hydrate energy, \( E_{\text{GH}} \), the additive energy, \( E_{\text{add}} \), and the energy of the interacting species, via:

$$ E_{\text{int}} = E_{{{\text{GH}} + {\text{add}}}} - \left( {E_{\text{GH}} + E_{\text{add}} } \right) $$

As an example, Lee et al. calculated the interaction energy between a cage and inhibitor molecule (amino acids and ionic liquids), and they found that the ionic liquid 1-butyl,3-methylimidazolium tertrafluoroborate had a greater probability of hydrate inhibition due to a stronger interaction energy compared to amino acid glycine [168]. This and many other binding energies can be compared in Table 3. Such relationships could be used in the reserve, to also determine improved hydrate formers, such as some surfactants (e.g. sodium dodecyl sulfate, SDS). Furthermore, these DFT calculations are used as a database to calculate larger systems, such as molecular dynamics simulations or statistical thermodynamics. Transition state theory is a method to determine the free energy barrier of reactions, of which the formation of gas hydrates could be applied, and DFT is commonly applied to it. However, molecular dynamics can be utilized the same way, as has been performed by Sicard et al. to understand how anti-agglomerants control methane transport with hydrates [106].

Table 3 Binding free energies calculated in literature from DFT or molecular dynamics simulations
Full size table

6.3 Molecular Dynamics

Molecular dynamics simulation is a very diverse and expanding field that can be utilized to calculate properties like surface tension, surfactant (reverse-) micellization in water/oil systems, bilayer and thin film formation, and looking just for such properties of SDS, there are already several sources [189,190,191,192,193,194,195,196,197,198]. Other studies of surfactant molecules have found a difference in the self-assembled structures due to changing conditions [199, 200].

More recently, a study by Choudhary et al. looked into the role of SDS (1 wt%) on methane hydrate formation in comparison with pure water [201], and it is one example of how simulations can be used to predict the influence of surfactants on gas hydrate growth behavior (Fig. 14). It was found that SDS tended to adsorb onto the hydrate surface because of its hydrophobic tail binding to openings in the cages of growing hydrates, which they report caused it to stabilize the nascent nuclei. This was expected to reduce the surface energy and thus also the nucleation barrier and induction time. They believe that these adsorbed surfactant molecules might change the morphology of the hydrate as it grows of larger length and time scales. The porosity could increase the mass transfer of guest molecule, leading to improved growth kinetics of the gas hydrate.

Fig. 14
figure 14

Molecular dynamics simulation of a a mixture of methane, water, and 1% SDS and b a snapshot after 1 μs at 270 K and 100 bar, showing significant 512 (black cage) and 51262 (red cage) hydrate formation (water molecules hidden in b). Key for dots: liquid water (light red), seed water (dark red), methane (blue), methane of seed (magenta), carbon of hydrophobic SDS tail (green), sulfur (yellow), hydrogen (white), and oxygen of SDS (red). Reprinted from [201]

Full size image

Another molecular dynamics study of SDS with methane hydrates found that micellization of the molecule was not required during the formation of gas hydrates [50].

Great interest has been invested in understanding formation kinetics using molecular simulations [202]. One such study by Walsh et al. [203] found that during the nucleation process the interaction of guest molecules with the faces and surfaces of partial hydrate cages led to the formation of the full gas hydrate. This and the many other studies have been able to improve the understanding of how additives may affect hydrate formation, such the simulations of CO2 and CH4 gas hydrates [204] leading to the further study on the impact of tetrahydrofuran on their formation [205].

In another important study, Carver et al. [206], using Monte Carlo simulations, found that the effect of PVP as an inhibitor depended on pendant hydrogens on the hydrate surface being available as adsorption sites, since PVP would lie along the surface and block these sites. Furthermore, Bui et al. studied how anti-agglomerants could either enhance or impair hydrate formation [107]. Similarly, other groups studied how sodium chloride might influence the adsorption behavior of anti-agglomerants [207, 208], surfactants like SDS, and hydrocarbons [209]. In another study by Bui et al. [105], they were able to reproduce micromechanical force experiments using equilibrium molecular dynamics.

Large systems have also been utilized with stochastic models like Kinetic Monte Carlo simulations. Understandings of agglomeration [210] and surface diffusion [108] behaviors of hydrates due to surfactacts have been particularly impacted by such Big Picture approaches.

6.4 Continuum Solvation Model

As mentioned above, conductor-like screening model with real solvation (COSMO-RS) can be employed to calculate many thermodynamic properties, in an approach different from molecular dynamics [169]. Taking advantage of the large database and promising nature (low vapor pressure, tunability, and bifunctionality) of ionic liquids, Bavoh et al. screened potential candidates for gas hydrate inhibition, rather than relying on trial and error, which are time intensive [211].

Athough group contribution (GC) methods, such as UNIQUAC Functional-group Activity Coefficients (UNIFAC), are commonplace and reliable for thermodynamic property prediction, they are lacking in accuracy when the groups are less studied experimentally or have long chains, such as in the case of ionic liquids. Thus, using a tool which is based on first principles becomes more attractive, considering its high accuracy [212]. It is already commonly used in pharmaceutical research and chemical engineering [213,214,215].

Bavoh et al. presented COSMO-RS as a novel prescreening tool for ionic liquid-based hydrate mitigation by correlating their calculated hydrogen bonding energies \( (E_{\text{HB}} ) \) with suppression temperature, \( \Delta T_{H} \), in comparison with induction time studies. Their work was able to describe the factors that impact the \( E_{\text{HB}} \) of ionic liquids in terms of hydrogen bond donors and acceptors, through the use of sigma profiles and potentials that are generated with COSMO-RS.

Similar studies could be performed with a focus, for example, on screening of surfactant molecules that can aid in gas hydrate formation, since the work by Bavoh et al. is one of the first to apply COSMO-RS calculations for gas hydrate applications, and it was limited to ionic liquids. There are still many ionic liquids that are not included in the database of commercially available software, which would also expand the scope of ionic liquids screening.

6.5 Machine Learning

An example of the application of machine learning for studying gas hydrates can be found in the work done by Xia et al. [216]. They incorporated a fusion modeling method that could be used to predict CO2 solubility in hydrates as related to nine ionic liquids. Using data collected from literature, they divided them into three sets, some for training, some for validation, and the rest for testing. With the training set, they were able to incorporate the back propagation neural network, support vector machine, and extreme learning machine. From these, three sub-models with the best performance were chosen according to the validation set. Afterwards, linear fusion models were included via the minimum square error and information entropy methods. Lastly, the prediction performance of these last sets of models was evaluated with the test set, which found that their linear fusion model was the best performing, with the information entropy method having better predictions. However, they do clarify that although their predictions worked well, they were not guaranteed to work on ionic liquids outside of the nine they tested, but it is something that is amenable to future expansion and has tremendous time savings compared to the alternatives, and it could rival COSMO-RS in the prediction of thermodynamic properties.

7 Closing Remarks and Future Prospects

In this chapter, an introduction into surfactant-based hydrate promotion studies was provided. Key areas that were touched include a summary of the theories proposed and how they have been implemented computationally, as well as discussions on surfactant-supported promoters. Based on the discussion, the following remarks can be added as the conclusion and future prospects

  • Many theories are suggested to explain the role of surfactants during hydrate promotion. Micellar and capillary-based theories are the two most prominent ones. The capillary effect in particular plays a critical role.

  • The molecular structure of a surfactant controls the hydrate formation efficiency, and surfactants with optimal structures would outperform the rest of the surfactants. Key factors that play a part in enhancing the efficiency include the ionic strength and chain length.

  • Surfactants also play a key role along with other co-promoters since surfactants affect formation in different ways, including stabilization, kinetic improvements, etc., at a suitable concentration.

  • The success of hydrate-based industrial applications depends on optimal use of surfactants along with other promoters and mechanical techniques.

  • Use of biosurfactants is gaining attention due to their biodegradability. More research is required regarding the use of biosurfactant in desalination.

  • Use of surfactants to improve hydrate formation kinetics is not well discussed in emerging technologies, including hydrogen storage, cold storage, etc.

  • Computational modeling has given a large number of insights into the understanding experimental observations, and they have assisted in smart design selection of surfactants and other additives.

  • Multi-scale modeling from quantum-scale to plant-scale will prove to be a major breakthrough in hydrate formation and storage, since it would be able to predict the appropriate surfactants and physical conditions that make the applications feasible and efficient.