Doubly dual nature of ammonium-based ionic liquids for methane hydrates probed by rocking-rig assembly

Abstract

This work presents a systematic study of methane hydrate inhibition in the presence of five structurally variable ionic liquids (ILs) belonging to the ammonium family—viz., tetra-alkylammonium acetate (TMAA), choline butyrate (Ch-But), choline iso-butyrate (Ch-iB), choline hexanoate (Ch-Hex) and choline octanoate (Ch-Oct). For this purpose, hydrate equilibrium curves have been obtained for methane in pure water and in 1 and 5 wt% aqueous solutions of ILs using a rocking cell apparatus (RC-5). Thermodynamic hydrate inhibition and promotion efficiency in each IL has been thoroughly investigated by observing the shifts in the hydrate equilibrium curves and trends of the calculated hydrate suppression temperatures (ΔT). The results were also obtained for methanol to validate the apparatus and compare the inhibition efficiency of ILs. Molar hydrate dissociation enthalpies (ΔH_diss) were obtained for the studied systems using the Clausius–Clapeyron equation and showed that ILs do not participate in forming the hydrate cages. The induction time (t_ind) data reveal that some of the studied ILs delay the time of hydrate formation, thereby indicating a potential kinetic inhibitor effect. The tested ILs exhibited a variety of phenomena, such as hydrate inhibition, hydrate stabilization, kinetic effect and surfactant-like behavior, which emphasize the use of ILs as potential gas hydrate inhibitors.

---

Introduction

Gas hydrates, in the simplest sense, are crystalline solids composed of water and gas. The gas molecules are encased within cavities formed by hydrogen-bonded water molecules at low temperatures and high pressures.^1–3 Similar to ice hydrates, they form a crystalline structure; however, whereas ice forms as a pure component, gas hydrates will form only with the presence of correctly sized gas molecules. More than 130 compounds are known to form clathrates when coexisting with water; the majority of which, including those found within natural gas, form sI, sII and sH structures.⁴

Although chemically interesting, these materials are industrially problematic because they can have a significant impact on natural gas production and safety; conditions such as high pressure and low temperature found at the well-head are often favorable for their formation.^5,6 The most common molecules that form hydrates are methane, ethane, propane and carbon dioxide.⁷ Currently, the oil and gas sector is investing heavily in hydrate-related research for a number of reasons, including the growth of the natural gas sector, the discovery of new gas fields more likely to result in hydrate formation, the potential for hydrates to be utilized as a fuel source, and indeed their potential for the separation of gases.^8,9

Currently, the most common way to prevent hydrates is to add special chemicals into the pipelines (a.k.a. thermodynamic hydrate inhibitors (THIs)), which are generally polar molecules and salts. It was discovered that these materials could be added to the aqueous phase within a gas pipeline and would lower the three-phase temperature or increase the three-phase pressure of hydrate formation. The industry preference is for methanol and glycols, which may be injected into pipelines and processes without undesirable side reactions.^10–12 Methanol is widely used in the industry as THI injected at the wellhead to prevent hydrate formation; however, much of it is usually lost into vapor phase and thus causing their dosage in the pipelines being as high as 50 wt% based on free water. Glycols generally have much lower vapor pressure than methanol and a much higher molecular mass; thus, they can be separated from the gas stream and recycled more easily. However, their higher viscosity leads to additional pumping costs.

On the other hand, low dosage hydrate inhibitors (LDHIs) are also used as hydrate inhibitors. The first recorded usage of LDHIs was in 1970 and it was documented that hydrate formation was suppressed upon the addition of a surfactant in a small quantities. However, no mechanistic details were provided.¹³ Later, in the late 1980s, researchers found that certain substances acted as hydrate suppressors such as poly(N-vinylpyrrolidone), Gaffix VC-713 and poly-VCap (PVCap), which is considered as the benchmark nowadays for flow assurance against which the efficiency of other LDHIs are considered. The next breakthrough for LDHIs came when the company Shell tested the performance of a range of surfactants as LDHIs, including quaternary ammonium and phosphonium surfactants. These were known to be good dispersants but also acted as clathate hydrate inhibitors. KHIs based on PVP, PVCap and their analogues or anti-agglomerates based on surfactant molecules have been successfully applied in the field by various companies worldwide.^14,15 However, there are still challenges to be faced in the design of inhibitors, particularly relating to ultra-deepwater drilling and compliance with increasingly stringent environmental legislation.

The use of quaternary ammonium salts as hydrate inhibitors opens up the possibility of applying ionic liquids to this problem. Ionic liquids have been marketed as both designer solvents and green chemicals because their functionality and physical properties can be designed or tailored to a specific application, and they have negligible vapor pressure and thus produce no VOCs.^16,17 However, very little research has been carried out in this area as summarized in a recent review by Tariq et al.¹⁸ Selective cases reported in the literature^18–20 have featured ionic liquids designed or tailored for hydrate inhibition application; rather, samples have been chosen from a range of commonly available materials. With the wide variety of possible cation and anion combinations available, ionic liquids are ideally placed to benefit from modern chemical product design philosophies, in which the materials are deliberately designed and tailored to incorporate the required functionality at the outset.

The above facts motivated us to carry out a systematic study on the effect of structural variations within the ILs and its effect on the methane hydrate inhibition performance of ionic liquids. Thus, in this work, five ionic liquids (ILs) belonging to the biocompatible, biodegradable ammonium family²¹ but structurally and functionally different were tested for their methane hydrate inhibition ability using a mesoscale rocking-rig apparatus (Scheme 1). The first IL studied was tetra-methylammonium acetate (TMAA), which is very similar to a tetra-alkylammonium salt that is a well-known hydrate inhibitor. The other four ILs belong to choline, also known as the substituted alkyl-ammonium family, in which one of the alkyl substitutions of TMA is replaced by a hydroxy-ethyl functionality. The four choline ILs were attached to anions that are slightly different in nature—viz., butyrate and iso-butyrate are isomeric counterparts—whereas butanoate, hexanoate and octanoate have slight differences in the chain length. The hydrate liquid vapor equilibrium (HLVE) and induction time data have been collected to test the strength of thermodynamic and kinetic inhibition. Quantitative analysis has also been performed by calculating the hydrate suppression temperature and molar hydrate dissociation enthalpies from the Clausius–Clapeyron equation. All the above research reveals a unique, doubly dual nature of ILs toward methane hydrate systems.

Scheme 1 Structures of the ionic liquids (ILs) used in this work. (a) Choline cation = Ch = (2-hydroxyethyl) trimethylammonium; (b) butyrate anion = But; (c) hexanoate anion = Hex; (d) octanoate anion = Oct; (e) iso-butyrate anion = iB and (f) tetramethylammonium acetate IL = TMAA.

Experimental

Materials

All ionic liquids (ILs) used in this work—viz., choline butyrate, choline iso-butyrate, choline butanoate, choline hexanoate, choline octanoate, and tetramethylammonium acetate—were synthesized and purified at Queen's University Belfast, U.K. using standard procedure reported elsewhere.²² The purity of all synthesized ILs was further checked by NMR, and except for traces of water (all ILs are hygroscopic), no other impurities were found. The ILs were vacuum dried before use and contain water contents less than 300 ppm. The structural details of the ILs are furnished in Scheme 1. Methanol (AR grade) was procured from Fischer Scientific. Methane was purchased from Buzware Scientific and Technical Gases, Qatar, with a stated purity of 99.9%. Throughout the experiments, Millipore Quality water was used for making the solutions. All IL inhibitor solutions were prepared by weighing on an electronic balance with a precision of ±0.00001 g.

Methods

The rocking-rig assembly (PSL Systemtechnik, Germany), Fig. 1, used in this study contains five test cells that perform 5 runs simultaneously. Each cell is connected with a pressure and temperature sensor. Similar to pressure sensors, temperature sensors are placed in the close proximity of the tested sample such that the accurate values have been obtained from the experiments. Each cell has a screw lid that provides an access to the measurement chamber and sample is filled or removed accordingly. A stainless steel ball is inserted into each cell for the mixing and rolls back and forth to create the desired turbulence in the tested sample mixture. To start a fresh experiment, the test cell was opened with a jaw wrench. The ball casing of the test cell and the mixing ball were washed carefully several times with distilled water and ethanol and dried. Later, the cell was filled with the prepared test mixture (i.e., ultrapure water and/or inhibitors) according to the experimental requirement. The test cell was then sealed properly and installed in its corresponding place on the platform axis in RC-5 bath, and connected to the temperature sensor and pressure supply. The test procedure with this experimental setup was started by filling each cell (with a maximum volume of 40 cm³) with 15 ml of sample solution. Each cell was pressurized directly with methane to the target pressure. Cylinder pressure is used for initial pressurization and hand pump is used to reach the desired pressure at each reaction cell.

Fig. 1 Different parts of the rocking rig assembly. (a) A typical rocking cell; (b) various parts of the rocking cell including temperature sensor, place for putting pressure sensor, steel ball for mixing and lid to close the cell; (c) a glimpse of mixing mechanism inside a typical rocking cell unit; (d) all the five cells immersed in the water bath and mounted on top of the rocking axis.

Before charging the cells with methane, each test cell with its aqueous content was flushed with the same gas twice to remove the undesired atmospheric gases from the test chamber. After the desired temperature/pressure conditions within the each test cell were stabilized, RC-5 software developed by PSL Systemtechnik, Germany was used to design the whole experimental protocol. The script was edited for an isothermal pressure search method.²³ The liquid–vapor mixture within each cell was left for at least 1 h for equilibration at 20 °C. After achieving stability in pressure/temperature values, the mixture was cooled down to 2 °C at a rate of 1.8 °C h⁻¹. The mixture was left for 24 h at 2 °C, and then the heating ramp was started at 0.1 °C h⁻¹ to dissociate the formed hydrates and close the loop. The mixing module was programmed according to the following parameters: rocking rate: 10 rocks per min, rocking angle: 30°. While the script is running, pressure, temperature and time data can be monitored in the RC-5 main window. Further details on the rocking cell assembly can be found elsewhere.^24–26 A schematic diagram is presented in Fig. 2 for the whole rocking-rig assembly used in this work.

Fig. 2 Schematic diagram of the rocking cell (RC-5) assembly connected to the thermostat, gas bottle and data acquisition computer.

The equilibrium points for each sample run were calculated from the specific pressure–temperature loops obtained after the completion of the experiments (Fig. 3(a)). This is the point at which a plateau in the P–T values was reached to close the loop. The two segments were fitted to linear equations and then solved to obtain the equilibrium (P,T) points. The method has been described in detail elsewhere.²³ The uncertainty in the reported equilibrium points is ±0.75%. Similarly, the induction time (or hydrate formation point) in the presence of inhibitors was also calculated from the pressure–time curves obtained for a specific sample solution as indicated in Fig. 3(b). The induction time is the time that indicates the first sign of hydrate crystal formation through an abrupt decrease in pressure from the initial values. The uncertainty in the reported induction times is ±0.5 h.

Fig. 3 Typical loops obtained from rocking-rig experiments to calculate (a) the hydrate liquid equilibrium data for methane in presence of 5 wt% tetra-methylammonium acetate (TMAA) and (b) to calculate the hydrate formation time for methane in presence of 1 wt% choline octanoate (Ch-Oct). Orange dots in (a) and (b) are calculated equilibrium point and induction times, respectively. Solid lines are just guide to eye.

Results and discussion

Apparatus validation

To test the suitability of the selected ILs for gas hydrate inhibition, in this study, we have first calibrated and validated the rocking-cell assembly (RC-5). For this purpose, pure methane hydrate equilibrium data points in deionized water were obtained, and the same experiment was repeated twice to establish the reproducibility of the results. The obtained results were then compared with the values reported in the literature.^27,28 One can see from Fig. 4(a) that there is excellent agreement in the values obtained using RC-5 and those reported in the literature. The two sets of results are also in good agreement and indicate high reproducibility of the apparatus. This also shows no significant difference in using either average temperature or individual cell temperature because all cells are placed in the same thermostatic bath. This fact has also been supported by Lone et al.²⁹ in a recent work. To check the accuracy in reproducing the HLVE data in the presence of hydrate inhibitor, similar runs using methane in the presence of 10 wt% methanol were carried out. The results are presented in Fig. 4(b) along with those reported in the literature.^30–32 It is clear from Fig. 4 that the results obtained using RC-5 are similar to that reported in the literature. This exercise established that the RC-5 assembly is suitable for such inhibitor efficiency tests.

Fig. 4 Hydrate equilibrium curves obtained from rocking-rig assembly used in this work for pure methane (a) and for methane in presence of 10 wt% methanol (b) along with the data collected from literature for validation purposes.

Hydrate equilibrium in the presence of inhibitors

The methane hydrate equilibrium curves were obtained in the absence and presence of different concentrations of ionic liquids (ILs) to look for the shifts in the equilibrium curves towards higher pressures and lower temperatures (typical action of a thermodynamic inhibitor). In Fig. 5, the results are presented for both tested concentrations—viz., 1 and 5 wt%. The first plot shows the effect of 1 wt% ILs on the hydrate equilibrium. It is clear from Fig. 5(a) that at lower pressures (P < 80 bar), there is a negligible effect of IL presence on the methane hydrate equilibrium curve (represented by the solid line). However, at higher pressures (P > 80 bar), a completely unexpected situation occurred. The equilibrium shifted towards the inverse side—i.e., lower pressures and higher temperatures. This indicates that instead of inhibiting the formation of hydrates, the ILs at this concentration (1 wt%) seem to act as hydrate promoters. Such behavior at lower concentrations has also been reported for other compounds.^33,34 and recently for a long-chain ionic liquid, 1-decyl-3-methylimidazolium chloride,²⁰ which is known to form micelles in aqueous solution.³⁵

Fig. 5 Hydrate vapour–liquid equilibrium curves obtained from rocking cell for methane in presence of 1 wt% ionic liquids (a) and 5 wt% ionic liquids along with same concentration of methanol (b). Solid lines are the methane P–T data fitted to second order polynomial (P = a + b × T + c × T²).

From Fig. 5(a), it is clear that at a pressure of ∼120 bar, choline octanoate (Ch-Oct) shows the highest hydrate promoting effect followed by choline butanoate (Ch-But), tetra-methyl ammonium acetate (TMAA), choline iso-butyrate (Ch-iB) and choline hexanoate (Ch-Hex). Ch-Oct is an amphiphilic IL and has more than six carbon atoms in the alkyl chain; thus, it is most likely to form micelles/aggregates in aqueous solutions at a concentration of 1 wt%.³⁶ Gayet et al.³⁷ also showed that a common surfactant prevents hydrate particles from agglomerating and forming a rigid hydrate film at the liquid–gas interface that hinder further hydrate formation. Therefore, Ch-Oct acts as a hydrate promoter rather than inhibitor at this concentration. Ch-Hex is also a potential aggregate former in water; however, owing to the shorter chain length compared with Ch-Oct, it will not form micelles at this concentration and will act as a simple electrolyte. Ch-Hex also shows the sign of hydrate promotion, but its hydrate promotion performance is poor compared to its octanoate counterpart (Ch-Oct). For other ILs showing signs of hydrate promotion, it can be assigned to their typical structure, which is very close to that of tetra-alkylammonium halide salts (TAAH), which are known to stabilize the hydrate cages.^38,39

Fig. 5(b) presents the methane hydrate equilibrium data obtained in the presence of 5 wt% ILs. The situation here is in complete contrast to what was observed for 1 wt% ILs as discussed previously. The equilibrium is shifted towards lower temperatures and higher pressures in the presence of all ILs—a typical thermodynamic inhibitor action. The ILs form hydrogen bonding with the ‘free’ water molecules available in the process and, because of the competition, do not allow the hydrate cages to stabilize at the given conditions, which results in a shift of the hydrate equilibrium. Thus, much lower temperatures and higher pressures are required to stabilize the hydrates as indicated in the plots. The order of inhibition strength for the ILs used in this work is as follows: TMAA > Ch-iB > Ch-Hex > Ch-But > Ch-Oct. These trends seems quite obvious because Ch-Oct is most likely to form aggregates in aqueous solution at the given concentration (5 wt%). Thus, very few ions will be available to integrate in the hydrogen-bonded network of water, which results in a negligible shift in the hydrate equilibrium at lower pressures (P > 80 bar), whereas small shifts can be noticed at higher pressures (P < 90 bar). Similar trends were observed for Ch-Hex because it is highly possible that at this concentration it also forms micelles owing to the surfactant-like character. Ch-But and Ch-iB show similar results in the whole experimental pressure range. However, the most striking results were observed for TMAA, which belongs to the family of tetra-alkylammonium salts, known as typical thermodynamic inhibitors.^38,39 At lower pressures, the results for TMAA are similar to other short-chained (>C6) ILs used in this work, but at higher pressures (P < 90 bar), there is a clear-cut boost in its performance. For comparison, hydrate equilibrium in the presence of 5 wt% methanol has also been studied and plotted along with the IL results. It can be envisaged here that at a pressure of ∼120 bar, the performance of 5 wt% Ch-Oct is similar to that of 5 wt% methanol, which is an encouraging result in itself. How does the structure of the studied ILs exhibit a dual functionality and play an instrumental role in suppressing and promoting the hydrate equilibrium? – will be discussed in the next sections.

Hydrate suppression temperatures

The IL inhibition efficiency was qualitatively discussed in the previous section. In this section, the calculated trends of ΔT will allow us to probe correct quantification of the efficiency of each studied IL inhibitor in the working experimental conditions.

The hydrate equilibrium data obtained in the absence and presence of different concentrations of the ionic liquids studied in this work are reported in Table 1. The data have been used to calculate the hydrate suppression temperature⁴⁰ by fitting the equilibrium pressure against temperature in a simple logarithmic expression T = a × ln(P) + b; where a and b are fitting coefficients. The difference between the values of methane hydrate dissociation temperature in deionized water from the hydrate dissociation temperature in a given concentration of IL will give us the hydrate suppression temperature (ΔT). Fig. 6 depicts the values of methane hydrate suppression temperatures in the presence of 1 and 5 wt% ILs in the studied pressure range. In Fig. 6(a), the trends of ΔT in the presence of 1 wt% ILs are presented. It can be seen from the plots that the studied ILs exhibit distinct trends. The inhibition performance of some of them is highly pressure dependent, such as that of TMAA, Ch-But and Ch-Oct, whereas some of them do not show any pressure dependence, such as Ch-iB and Ch-Hex. It can be inferred that at lower pressures (P < 55 bar), three of the ILs act as hydrate inhibitors; however, the magnitude of shift (ΔT) is very small or negligible, falling in the range of 0.5 to 0.1 °C in the pressure range of 40 to 55 bar. At the studied high pressure range, 55 to 120 bars, there is a role reversal in the behavior of these ILs. The magnitude of ΔT begins to decrease in the negative zone (55 bar onwards), thus indicating a hydrate-promoting effect that further increases with increasing pressure. The best inhibitor in the lower pressure range becomes the best promoter in the high pressure range as indicated in the plots. The promoting strength of ILs follow the series Ch-Oct > Ch-But > TMAA, which is simply reversed at lower pressures. However, it is clear from the plots that the performance of Ch-iB and Ch-Hex does not depend on the working pressure. Both of these ILs show similar hydrate suppression temperatures falling in the range of −0.1 to −0.3 °C; which in fact indicates that both of these ILs work as methane hydrate promoters at this concentration (1 wt%) and in the studied pressure range (40–120 bar).

Table 1 Hydrate dissociation data for methane in presence of various concentrations of ionic liquids obtained through rocking-rig assembly

TMAA		Ch-But		Ch-iB		Ch-Hex		Ch-Oct
T/°C	P/bar	T/°C	P/bar	T/°C	P/bar	T/°C	P/bar	T/°C	P/bar
1 wt%
4.12	38.47	8.13	57.43	14.50	117.74	9.18	63.47	4.01	38.07
9.56	65.87	11.13	78.03	10.73	75.36	11.05	76.87	8.35	58.83
11.17	79.21	12.99	94.77	7.48	53.18	13.44	96.38	11.06	76.62
13.29	97.56	14.99	114.09			14.18	114.39	13.46	96.39
14.60	116.17							15.64	115.93
5 wt%
4.09	41.13	7.99	59.59	13.55	113.09	3.95	39.02	10.04	69.96
7.72	59.71	10.51	76.92	11.74	93.62	8.50	60.03	11.06	78.25
10.44	79.98	11.98	95.12	10.22	78.64	10.67	78.54	12.44	97.56
11.84	97.24	13.49	114.09	7.53	58.29	13.13	97.71	14.10	122.26
12.19	117.59					13.28	115.40

---

Fig. 6 Calculated methane hydrate suppression temperature after fitting the curves against pressure presented in Fig. 5 using T = a − b × ln P; in presence of 1 wt% ionic liquids (a) and 5 wt% ionic liquids and same concentration of methanol (b) and its variation against pressure.

The picture for 5 wt% ILs in Fig. 6(b) is in complete contrast to what we witnessed earlier for 1 wt% ILs in Fig. 6(a). There is a clear indication that the magnitude of ΔT is always positive in the studied pressure range (40–120 bar), indicating their hydrate-inhibiting behavior. However, the distinct behavior of each IL has been observed along the studied pressure range. TMAA and Ch-But show enhanced hydrate inhibition efficiency with increasing pressure, with magnitudes of 0.7–1.4 °C and 0.2–0.6 °C, respectively, in the pressure range of 40–120 bar. In contrast, Ch-iB, Ch-Hex and Ch-Oct show decreased efficiency with magnitudes of 0.8–0.5 °C, 0.4–0.2 °C and 0.5–0.1 °C, respectively, in the similar pressure range described previously. However, the behavior of methanol is the same throughout the studied pressure range, showing a suppression of ∼2 °C in hydrate dissociation temperatures. Many theoretical studies^41–45 are available in the literature where the inhibition mechanisms of methane hydrates in presence of classical inhibitors are reported. It has been proposed by Anderson et al.⁴⁵ after studying the local structure of liquid water in presence of an inhibitor molecule under hydrate forming conditions that there are two characteristics that leads to strong inhibition effect: (i) charge distribution on the edge of the inhibitor and (ii) appropriate size of the inhibitor with respect to the available space at the hydrate-surface binding site. The ILs can be easily designed containing both these characteristics and thus have the potential to be potential alternative inhibitors.

Dissociation enthalpy of methane hydrates

The dissociation enthalpies, ΔH_diss, of gas hydrates are generally experimentally determined using calorimetric measurements and are necessary to predict the heat requirement for hydrate dissociation. An alternative way to obtain the values of ΔH_diss is through the Clausius–Clapeyron equation by differentiation of the phase equilibrium data. The methane hydrate formation/dissociation equilibrium can be represented by the equation:

CH₄ + nH₂O ↔ CH₄·nH₂O (1)

where n is the mole number of water in crystalline hydrate. As proposed by various groups,^46–48 ΔH_diss values above 0 °C can be determined from the univariate slope of the phase equilibrium line obtained through the Clausius–Clapeyron equation:where P is the pressure, T is the temperature, z is the compressibility factor of methane for average gas temperature and pressure, R is the universal gas constant, and ΔH_diss is the molar dissociation enthalpy of methane hydrates. The (P,T) data presented in Table 1 were plotted as ln P versus 1000/T, which exhibits linear trends showing satisfactory correlation coefficients (R²) for all concentrations of ILs, methanol and water. The ΔH_diss values were calculated from the obtained slopes of eqn (2) and are presented in Table 2. It is clear from Table 2 that the value of ΔH_diss for methane hydrates is in excellent agreement with that reported in the literature⁴⁴ obtained in the same pressure and temperature range. The data for the ILs can be divided into two groups for different concentrations. At 1 wt%, Ch-iB and Ch-Hex show very similar values of ΔH_diss compared with that obtained for neat methane hydrates, whereas TMAA, Ch-But and Ch-Oct show values that are less than the ΔH_diss of neat methane hydrates. Similar values of ΔH_diss indicate that those ILs do not participate in the hydrate cage formation, and only the type sI methane hydrates are formed in the system. It can also be envisaged that most of the energy is consumed to break down the hydrogen-bonded network of water present in hydrate cages.^19,49 The fact that the other three ILs at this concentration exhibit slightly lower values of ΔH_diss means that these ILs do participate and affect the hydrate cages, and the trends are very similar to the ΔT trends shown in Fig. 6(a). Tetra-alkylammonium salts are known to participate in the hydrate cages;^38,39 however, the thermodynamic stability of the cages is questionable (see previous section).

Table 2 Calculated molar hydrate dissociation enthalpies ΔH_diss (kJ mol⁻¹) of methane in pure water and in presence of various concentrations of ionic liquids aqueous solutions obtained using Clausius–Clapeyron equation in the studied temperature–pressure range

Components	ΔHdiss/kJ mol^−1
Methane + water	57.09	57.09 (ref. 43)
Methane + water + methanol (5 wt%)	—	57.20

---

(Methane + water) + IL	1 wt%	5 wt%
TMAA	51.84	59.50
Ch-But	50.78	59.75
Ch-iB	57.05	55.46
Ch-Hex	56.45	54.20
Ch-Oct	48.02	49.88

---

At 5 wt%, the ΔH_diss values for most of the ILs including methanol at a similar concentration are similar to the neat methane hydrates. However, the slight difference in the ΔH_diss values is consistent with the trends of ΔT presented in the last section. The lowest value of ΔH_diss has been shown by Ch-Oct, which is known to be a surfactant and seems to act as a kinetic inhibitor (see next section). Therefore, the trends of ΔH_diss support our previously proposed hypothesis. It must be stressed that the slight difference in the ΔH_diss values should be considered with due care. These small differences (a maximum of 9 kJ mol⁻¹ for Ch-Oct) could be due to the uncertainty in measured equilibrium (P,T) values resulting in large errors in calculation of z and ΔH_diss.^19,50,51

Hydrate induction time

In this section, the hydrate formation time in the absence of any inhibitor and in the presence of different concentrations of IL inhibitors were obtained through the pressure–time loops as presented in Fig. 3(b). The induction time is the time when the onset of hydrate crystal formation occurs.^52,53 The data obtained from the loops are presented in Fig. 7, which gives us some information about the kinetic inhibiting behavior of the studied ILs when present in the system at concentrations of 1 and 5 wt%. It can be inferred from Fig. 7(a) that there is very little or almost negligible delay in the induction time for 1 wt% ILs at lower pressures. At high pressures of 100 and 120 bars, Ch-Oct and Ch-But show little kinetic inhibition. However, at a concentration of 5 wt%, Fig. 7(b), the trends are better in resolution to extract information about ILs kinetic inhibition ability. Ch-Oct is clearly performing as the best kinetic inhibitor among the other IL samples tested in this work by delaying the hydrate formation time by 1.3 h at higher pressure (120 bar) and 0.8 h at lower pressure (40 bar). Ch-Oct kinetic inhibition performance is followed by the TMAA, Ch-Hex, Ch-But and Ch-iB, respectively. The trends can be explained on the basis of proposed mechanism of hydrate inhibition using polymers and surfactants.^54,55 Ch-Oct and Ch-Hex has long alkyl chain and form micelles. Surface active molecules are known to adsorb strongly on the surface of a propagating hydrate crystal or pre-nuclear hydrate like clusters. In this process, they change the energy of the surface of crystal or cluster and thereby change its growth.⁵⁴ Thus, the analysis shows that the studied ILs have the ability to perform as mild kinetic inhibitors too.

Fig. 7 Induction time for methane hydrate formation in presence of 1 wt% (a) and 5 wt% ionic liquids (b) at different pressures. The solid lines are just a guide to eye.

Conclusions

In the present work, five ionic liquids (ILs) that belong to the same ammonium family but are structurally and functionally different were tested with regard to their methane hydrate inhibition ability using a mesoscale rocking-rig apparatus. The first IL studied was tetra-methylammonium acetate (TMAA), which is very similar to a tetra-alkylammonium salt that is a well-known hydrate inhibitor/promoter. The other four ILs belong to choline, also known as substituted alkyl-ammonium family, in which one of the alkyl substitutions of TMA is replaced by a hydroxy-ethyl functionality. The other four choline ILs were attached to anions that are different in nature—viz., butyrate and iso-butyrate are isomeric counterparts—whereas hexanoate and octanoate have a difference in the chain length.

It has been found that TMAA is the best thermodynamic inhibitor among the five tested ILs because of its hydrogen bonding capability with water. Ch-Oct owing to its micelles forming ability, has given the best performance as a kinetic inhibitor among the studied ILs in delaying the hydrate formation time by a span of more than 1 h. The working concentration and pressure range are also important factors: at elevated pressures and low concentration (1 wt%); Ch-Oct, Ch-But and TMAA acts as hydrate promoters. Calculated molar hydrate dissociation enthalpies indicate that the ILs do not participate in the hydrate cages. Thus, it must be emphasized that slight structural variations in the structure of ILs reveal their doubly dual nature for methane hydrate systems—viz., thermodynamic inhibition, hydrate promotion, kinetic inhibition and surfactant/anti-agglomerate character. Thereby this study shows the potential of ILs to be used for flow assurance strategies.

Acknowledgements

This work was made possible by NPRP grant #5-590-2-238 and #6-330-2-140 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors.

References

1. E. D. Sloan Jr., Fundamental principles and applications of natural gas hydrates, Nature, 2003, 426, 353–359.
2. W. Kondo, K. Ohtsuka, R. Ohmura, S. Takeya and Y. H. Mori, Clathate-hydrate formation from a hydrocarbon gas mixture: compositional evolution of formed hydrate during an isobaric semi-batch hydrate-froming operations, Appl. Energy, 2014, 113, 864–871.
3. R. Ludwig, Water: from clusters to the bulk, Angew. Chem., Int. Ed., 2001, 40, 1808–1828.
4. E. D. Sloan, Clathate Hydrates of Natural Gases, Marcel Dekker Inc., New York, 1998.
5. C. Moon, P. C. Taylor and P. M. Rodger, Molecular dynamics study of gas hydrate formation, J. Am. Chem. Soc., 2003, 125, 4706–4707.
6. R. Anderson, A. Chapoy and B. Tohidi, Phase relations and binary clathrate hydrate formation in the system H₂-THF-H₂O, Langmuir, 2007, 23, 3440–3444.
7. S. Mokhatab, W. A. Poe and J. Y. Mak, Handbook of Natural Gas Transmission and Processing: Principles and Practices, 3rd edn, 2015, ch. 2.
8. A. Demirbas, Methane Gas Hydrates, Springer, 2010.
9. C. Giavarini and K. Hester, Hydrates seen as a problem for oil and gas industry, in Gas Hydrates: Immense energy potential and environmental challenges, 2011, pp. 97–116.
10. S. Mokhatab, W. A. Poe and J. G. Speight, Handbook of natural gas transmission and processing, Elsevier Inc., 2006.
11. F. E. Anderson and J. M. Prausnitz, Inhibition of gas hydrates by methanol, AIChE J., 1986, 32, 1321–1333.
12. P. G. Lafond, K. A. Olcott, E. D. Sloan, C. A. Koh and A. K. Sum, Measurements of methane hydrate equilibrium in systems inhibited with NaCl and methanol, J. Chem. Thermodyn., 2012, 48, 1–6.
13. A. Perrin, O. M. Musab and J. W. Steed, The chemistry of low dosage clathate hydrate inhibitors, Chem. Soc. Rev., 2013, 42, 1996–2015.
14. M. A. Kelland, History of the development of low dosage hydrate inhibitors, Energy Fuels, 2006, 20, 825–847.
15. C. M. Perfeldt, P. C. Chua, N. Daraboina, D. Friis, E. Kristiansen, H. Ramlov, J. M. Woodley, M. A. Kelland and N. V. Solms, Inhibition of gas hydrate nucleation and growth: efficacy of an antifreeze protein from the longhorn beetle Rhagiummordax, Energy Fuels, 2014, 28, 3666–3672.
16. N. V. Plechkova and K. R. Seddon, Applications of ionic liquids in the chemical industry, Chem. Soc. Rev., 2008, 37, 123–150.
17. J. M. S. S. Esperança, J. N. C. Lopes, M. Tariq, L. M. N. B. F. Santos, J. W. Magee and L. P. N. Rebelo, Volatility of aprotic ionic liquids – A Review, J. Chem. Eng. Data, 2009, 55, 3–12.
18. M. Tariq, D. Rooney, E. Othman, S. Aparicio, M. Atilhan and M. Khraisheh, Gas hydrate inhibition: a review of the role of ionic liquids, Ind. Eng. Chem. Res., 2014, 53, 17855–17868.
19. Z. Long, X. Zhou, X. Shen, D.-L. Li and D.-Q. Liang, Phase Equilibria and Dissociation Enthalpies of Methane Hydrate in Imidazolium Ionic Liquid Aqueous Solutions, Ind. Eng. Chem. Res., 2015, 54, 11701–11708.
20. C.-K. Chu, S.-T. Lin, Y.-P. Chen, P.-C. Chen and L.-J. Chen, Chain length effect of ionic liquid 1–alkyl–3–methylimidazolium chloride on the phase equilibrium of methane hydrate, Fluid Phase Equilib., 2015, 413, 57–64.
21. M. Petkovic, K. R. Seddon, L. P. N. Rebelo and C. S. Pereira, Ionic liquids: a pathway to environmental acceptability, Chem. Soc. Rev., 2011, 40, 1383–1403.
22. M. Petkovic, J. L. Ferguson, H. Q. N. Gunaratne, R. Ferreira, M. C. Leitao, K. R. Seddon, L. P. N. Rebelo and C. S. Pereira, Novel biocompatible cholinium-based ionic liquids—toxicity and biodegradability, Green Chem., 2010, 12, 643–649.
23. B. Tohidi, R. W. Burgass, A. Danesh, K. K. Østergaard and A. C. Todd, Improving the Accuracy of Gas Hydrate Dissociation Point Measurements, Ann. N. Y. Acad. Sci., 2000, 912, 924–931.
24. N. Daraboina, C. Malmos and N. V. Solms, Synergistic kinetic inhibition of natural gas hydrate formation, Fuel, 2013, 108, 749–757.
25. A. Lone and M. A. Kelland, Exploring kinetic hydrate inhibitor test methods and conditions using a multicell stell rocker rig, Energy Fuels, 2013, 27, 2536–2547.
26. M. F. Mady, J. M. Bak, H. Lee and M. A. Kelland, The first kinetic hydrate inhibition investigation on fluorinated polymers: poly(fluoroalkylacrylamide)s, Chem. Eng. Sci., 2014, 119, 230–235.
27. L. Keshavarz, J. Javanmardi, A. Eslamimanesh and A. H. Mohammadi, Fluid Phase Equilib., 2013, 354, 312–318.
28. K. Tumba, P. Reddy, P. Naidoo and D. Ramjugernath, Phase Equilibria of Methane and Carbon Dioxide Clathate Hydrates in the Presence of Aqueous Solutions of Tributylmethylphosphonium Methylsulfate Ionic Liquid, J. Chem. Eng. Data, 2011, 56, 3620–3629.
29. A. Lone, Establishing a new high pressure steel multi-cell rocker rig for kinetic hydrate inhibitor testing, M. S. thesis, University of Stavanger, Norway, 2011.
30. H. Haghighi, A. Chapoy, R. Burgess, S. Mazloum and B. Tohidi, Fluid Phase Equilib., 2009, 278, 109–116.
31. H.-J. Ng and D. B. Robinson, Fluid Phase Equilib., 1985, 21, 145–155.
32. A. H. Mohammadi and D. Richon, Phase Equilibria of Methane Hydrates in the Presence of Methanol and/or Ethylene Glycol Aqueous Solutions, Ind. Eng. Chem. Res., 2010, 49, 925–928.
33. M. Aliabadi, A. Rasoolzadeh, F. Esmaeilzadeh and A. Alamdari, Experimental Study of Using CuO Nanoparticles as a Methane Hydrate Promoter, J. Nat. Gas Sci. Eng., 2015, 27, 1518–1522.
34. B. Partoon and J. Javanmardi, Effect of Mixed Thermodynamic and Kinetic Hydrate Promoters on Methane Hydrate Phase Boundary and Formation Kinetics, J. Chem. Eng. Data, 2013, 58, 501–509.
35. M. Tariq, F. Moscoso, F. J. Deive, A. Rodriguez, M. A. Sanromán, J. M. S. S. Esperança, J. N. C. Lopes and L. P. N. Rebelo, J. Chem. Thermodyn., 2013, 59, 43–48.
36. M. Blesic, M. H. Marques, N. V. Plechkova, K. R. Seddon, L. P. N. Rebelo and A. Lopes, Self-aggregation of ionic liquids: micelle formation in aqueous solution, Green Chem., 2007, 9, 481–490.
37. P. Gayet, C. Dicharrya, G. Mariona, A. Graciaa, J. Lachaisea and A. Nesterovb, Experimental determination of methane hydrate dissociation curve up to 55Mpa by using a small amount of surfactant as hydrate promoter, Chem. Eng. Sci., 2005, 60, 5751–5758.
38. M. Arjmandi, A. Chapoy and B. Tohidi, Equilibrium data of hydrogen, methane, nitrogen, carbon dioxide, and natural gas in semiclathate hydrates of tetran-butyl ammonium bromide, J. Chem. Eng. Data, 2007, 52, 2153–2158.
39. A. Joshi, J. S. Sangwai, K. Das and N. A. Sami, Experimental investigations on the phase equilibrium of semiclathate hydrates of carbon dioxide in TBAB with small amount of surfactant, Int. J. Energy Environ. Eng., 2013, 4, 11.
40. K. K. Østergaard, R. Masoudi, B. Tohidi, A. Danesh and A. C. Todd, A general correlation for predicting the suppression of hydrate dissociation temperature in the presence of thermodynamic inhibitors, J. Pet. Sci. Eng., 2005, 48, 70–80.
41. X. Cao, Y. Huang, W. Li, Z. Zheng, X. Jiang, Y. Su, J. Zhao and C. Liu, Phase diagrams for clathrate hydrates of methane, ethane, and propane from first-principles thermodynamics, Phys. Chem. Chem. Phys., 2016, 18, 3272–3279.
42. M. Atilhan, N. Pala and S. Aparicio, J. Mol. Model., 2014, 20, 2182.
43. S. Pal and T. K. Kundu, DFT-based inhibitor and promoter selection criteria for pentagonal dodecahedron methane hydrate cage, J. Chem. Sci., 2013, 125, 1259–1266.
44. T. Yagasaki, M. Matsumoto and H. Tanaka, Adsorption Mechanism of Inhibitor and Guest Molecules on the Surface of Gas Hydrates, J. Am. Chem. Soc., 2015, 137, 12079–12085.
45. B. J. Anderson, J. W. Tester, G. P. Borghi and B. L. Trout, Properties of Inhibitors of Methane Hydrate Formation via Molecular Dynamics Simulations, J. Am. Chem. Soc., 2005, 127, 17852–17862.
46. M. B. Rydzy, J. M. Schicks, R. Naumann and J. Erzinger, J. Phys. Chem. B, 2007, 111, 9539–9545.
47. K. M. Sabil, G. J. Witkamp and C. J. Peters, Fluid Phase Equilib., 2010, 290, 109–114.
48. E. D. Sloan and F. Fleyfel, Fluid Phase Equilib., 1992, 76, 123–140.
49. K. M. Sabil, O. Nashed, B. Lal, L. Ismail and A. Japper-Jaafar, Experimental investigation on the dissociation conditions of methane hydrate in the presence of imidazolium-based ionic liquids, J. Chem. Thermodyn., 2015, 84, 7–13.
50. Q. Nasir, K. K. Lau and B. Lal, Enthalpies of Dissociation of Pure Methane andCarbon Dioxide Gas Hydrate, Int. J. Chem. Mol. Nucl. Mat. Metal. Eng., 2014, 8, 838–841.
51. P. Skovborg and P. Rasmussen, Comments on: hydrate dissociation enthalpy and guest size, Fluid Phase Equilib., 1994, 96, 223–231.
52. A. Vysniauskas and P. R. Bishnoi, A kinetic study of methane hydrate formation, Chem. Eng. Sci., 1983, 38, 1061–1072.
53. P. Skovborg, H. Ng, P. Rasmussen and U. Mohn, Measurement of induction times for the formation of methane and ethane gas hydrates, Chem. Eng. Sci., 1993, 48, 445–453.
54. N. J. Phillips and M. A. Kelland, Industrial Applications of Surfactants IV, ed. D. R. Karsa, Royal Society of Chemistry, 1999, pp. 244–259.
55. B. Kvamme and R. Asnes, Mechanism for kinetic hydrate inhibitors, Proc. 4^th WSEAS Conf. Heat Mass Transf., Gold Coast, Queensland, Australia, 2007.

---