Fuel 259 (2020) 116219

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Quaternary ammonium salts as thermodynamic hydrate inhibitors in the T

presence and absence of monoethylene glycol for methane hydrates
⁎
Ali Qasima,b, Muhammad Saad Khana,c, Bhajan Lala,b, , Mokhtar Che Ismaild, Khairul Rostanie
a
Chemical Engineering Department, Universiti Teknologi PETRONAS, Bandar Seri Iskandar 32610, Perak, Malaysia
b
CO2 Research Centre (CO2RES), Universiti Teknologi PETRONAS, Bandar Seri Iskandar 32610, Perak, Malaysia
c
Texas A&M University at Qatar, Doha, Qatar
d
Department of Mechanical Engineering, Centre for Corrosion Research, Universiti Teknologi PETRONAS, Bandar Seri Iskandar 32610, Perak, Malaysia
e
Gas Sustainability Technology R&D, PETRONAS Research Sdn Bhd, Malaysia

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO A B S T R A C T

Keywords: The thermodynamic hydrate inhibition (THI) effect induced by three different quaternary ammonium salts
Hydrate liquid vapor equilibrium (HLVE) (QASs) namely tetraethylammonium iodide (TEAI), tetramethylammonium bromide (TMAB) and tetra-
Methane hydrates ethylammoniun (TEAB) on the formation of methane gas hydrates has been investigated. The THI impact is also
Quaternary ammonium salts (QASs) determined in the presence of monethylene glycol (MEG) for TMAB and TEAB. The inhibition capacity is ex-
Thermodynamic hydrate inhibition (THI)
amined by finding the average suppression temperature. Experiments are performed in the pressure range of
Thermodynamic modeling
3.45–8.3 MPa at the concentration values of 1, 5 and 10 wt%. Hydrate liquid vapor equilibrium (HLVE) points
are calculated by using isochoric thermodynamic cycle (T-cycle) method. COSMO-RS study is also performed to
understand the THI behavior of the compounds by analyzing sigma profile. Along with it hydrate dissociation
enthalpies (ΔHdiss) are considered. The results revealed that inhibition effect increased with increase in con-
centration. TMAB showed better impact on hydrate suppression as compared to TEAB and TEAI in pure as well
as in mixture with MEG. Moreover, in thermodynamic modeling section, Dickens and Quinby-Hunt model is
employed to predict HLVE data. The equilibrium temperatures predicted by model are compared with the ex-
perimental results and found to be in good agreement. Hence these compounds can be employed in THI ap-
plications and flow assurance schemes.

⁎
Corresponding author at: Chemical Engineering Department, Universiti Teknologi PETRONAS, Bandar Seri Iskandar 32610, Perak, Malaysia.
E-mail address: bhajan.lal@utp.edu.my (B. Lal).

https://doi.org/10.1016/j.fuel.2019.116219
Received 6 July 2019; Received in revised form 20 August 2019; Accepted 16 September 2019
0016-2361/ © 2019 Elsevier Ltd. All rights reserved.
A. Qasim, et al. Fuel 259 (2020) 116219

1. Introduction coefficient models included NRTL, Pitzer-Mayorga and Pitzer-Mayorga-

Zavitsas-hydration models. The applied models were used to evaluate
Gas hydrates consist of host molecules of water and guest molecules water activity in ionic liquid aqueous solutions.
of gas which are trapped inside the host [1–4]. The gas hydrate struc- In preceding works for hydrate system, COSMO-RS software has
tures are formed as a result of hydrogen bonding of gas molecules and been employed in finding the intermolecular interactions of ionic li-
water [5–13]. In order for gas hydrates to form suitably, certain ne- quids and water [36,56]. It was established in previous studies that
cessary conditions of high pressure and lower temperatures are re- hydrogen bonding and sigma profile are two main factors which in-
quired [14–17]. The gas hydrate study involves the exploration and dicate the thermodynamic inhibition performance of ionic liquids and
search for the compounds which can be utilized in both the applications other compounds in COSMO-RS [57,58]. In case of imidazolium based
of mitigation and promotion of hydrate structures [18–20]. Some of the ionic liquids, Bavoh et al. [56] reported that sigma profile demonstrates
gases which commonly form hydrates are carbon dioxide, methane, the ability of a compound to be miscible and also hydrogen bonding
ethane, propane and hydrogen sulfide in pipeline conditions [21–25]. ability with water molecules. These are the most important properties
Gas hydrates form either structure I, structure II or structure H [26–30]. that define THI behavior of a compound.
Methane usually forms structure I crystal structures of hydrates Numerous ways have been employed in the literature to model the
[31,32]. systems which contain ionic liquids. Among these, one of the most
In oil and gas industry, hydrate formation poses a threat to normal commonly used approaches is activity coefficient method [30,59,60]. It
flow inside the flowline causing significant disruption [27,33–35]. It is stipulates that the thermodynamic hydrate inhibition is obtained due to
estimated that plugging of the pipeline by natural gas hydrate (NGH) the activity of the chemical in aqueous solution. Preceding modeling
costs the gas industry around $1 million per day estimated at the studies included pure methane systems along with electrolytes such as
shutting down of production [34,36]. Apart from the shutdown loss, it amino acids and in some cases ILs also [61,41]. The inhibition effect
causes safety concern to the personnel at the field and a possible risk to offered by amino acids and their thermodynamic modeling for methane
pipeline safety also [13,37,38]. Hydrate reserves present in deep oceans hydrate formation has previously been studied but in case of quaternary
and permafrost regions are a large natural source of methane [1,39]. It ammonium salts (QASs), modeling studies have not been performed
is assessed that hydrate deposits in permafrost regions are potential extensively. It will be useful to determine the practicability of the model
sources of fuel that can be utilized for power generation purposes for methane hydrates in the presence of quaternary ammonium salts.
[6,32]. Currently, oil and gas industry is focusing on methane hydrate For this reason it is necessary to examine the activity coefficient based
mitigation as methane is prime component of natural gas [12,40]. modelling approach which describes the hydrate formation mechanism
Methane gas hydrates are formed at relatively higher pressures as of quaternary ammonium compounds. For this purpose, Dickens and
compared to CO2 hydrates [41]. Quinby-Hunt predicted a thermodynamic model which is electrolyte
In oil and gas industry, different techniques are applied to avoid based [60]. It can be used to predict the hydrate liquid–vapor-equili-
hydrate formation inside the pipelines which may disrupt the normal brium (HLVE) conditions for methane hydrates with and without MEG
flow. At present, the techniques which are employed to stop forming for QAS solutions.
hydrates include thermal heating, dehydration, depressurization and In previous studies, imidazolium based ILs were focused mainly for
chemical method [11,42–44]. Among all the methods discussed above, thermodynamic gas hydrate mitigation but compounds with ammo-
chemical inhibition is considered as the most economical and cost-ef- nium base especially QASs are used scarcely in the presence of methane
fective method as it inhibits clathrate development without disturbing hydrates. However, the impact of MEG in the presence of QASs has
the normal pipeline flow [13,23,45]. never been reported for CH4 hydrates. MEG is considered in mixtures as
Chemical inhibitors are categorized into two types which are ther- it is a commercially used thermodynamic inhibitor. The addition of
modynamic hydrate inhibitors (THIs) and kinetic hydrate inhibitors MEG with QAS can provide useful information about their combined
(KHIs) [16]. In industrial applications, KHIs are used in low con- effect in hydrate inhibition process. Thus, in the current study, ther-
centrations varying from 1 to 2 wt% while for THI, concentration varies modynamic mitigation effect of three quaternary ammonium salts is
from 1 to as high as 50 wt% [4,21,46]. KHIs prevent formation of hy- probed for methane gas hydrates. The pressure range selected for this
drates by delaying the nucleation time thus avoiding the clathrate study is 3.4–8.30 MPa. This particular pressure range is selected as CH4
forming [47]. In case of THI, chemicals added for inhibition prevent hydrates are usually formed at high pressures. QASs are used in the
hydrate formation by shifting the equilibrium towards high pressure concentrations of 1, 5 and 10 wt%. Typically, thermodynamic hydrate
and lower temperatures thus enhancing the hydrate free region. For inhibitors are used in 5 and 10 wt% concentrations for commercial flow
thermodynamic inhibition, methanol and monoethylene glycol (MEG) assurance applications.
are used in commercial applications [48,49,1]. Various classes of
compounds have been engaged at laboratory scale for gas hydrate in- 2. Methodology
hibition in which ionic liquids find special attention due to their anionic
and cationic moieties [28,33,50–54]. 2.1. Chemicals required
The literature survey shows that ionic liquids and quaternary am-
monium salts (QAS) find application in hydrate mitigation practices. Sigma-Aldrich has provided all the QASs used for this study. Table 1
Xiao et al. reported that some of the imidazolium-based ionic liquids act represents the names and purities of all the chemicals. For experi-
as KHIs and THIs for methane hydrates [50]. Richard et al. [53] mea- mentation, methane gas was bought from Linde Malaysia Sdn Bhd.
sured hydrate dissociation of EMIM-Cl up to the solution concentration Deionized water was used to dilute the salt solutions in order to achieve
of 40% at pressure conditions of 10, 15 and 40 MPa. The high pressure different concentrations. It was already available in the laboratory.
was achieved using micro differential scanning calorimeter. They also
conducted experiments using mixed ionic liquids and other conven- 2.2. Experimental apparatus
tional inhibitors for studying dissociation conditions of methane hy-
drate. The conventional inhibitors included monoethylene glycol Fig. 1 demonstrates the schematics of apparatus employed to con-
(MEG) and sodium chloride (NaCl). The mixtures of two ionic liquids duct experiments. It comprises of six distinct units of identical rocking
namely EMIM-Cl and EMIM-Br were also employed in their study along cells which can bear high pressure and placed in a compartment where
with conventional inhibitors to probe the possible synergic effects [53]. the temperature can be controlled. The compartment or the tempera-
Avula et al. [55] performed a theoretical study using 21 different ionic ture controlled chamber contains a rocking mechanism in conjunction
liquids to predict dissociation of methane hydrates. The applied activity with the ball count method. The reactor has an internal volume of

2
A. Qasim, et al. Fuel 259 (2020) 116219

Table 1
List of chemicals.
Name of chemical Symbol Mol. Wt. (g/mol) Purity

Water H2O 18 Deionized

Methane CH4 16 99.95 mol%
Mono Ethylene Glycol (MEG) C2H6O2 62.07 Anhydrous, 99.8%
Tetraethylammonium iodide (TEAI) C8H20IN 257.16 98%
Tetraethylammonium bromide (TEAB) C8H20NBr 210.16 98%
Tetramthylammonium bromide (TMAB) C4H12BrN 154.05 98%

37.3 cm3, a height of 7.6 cm, wall thickness of 2 cm, and internal dia- quantity of CH4. The cells were purged three times. The cells were
meter of 2.5 cm. A built-in PLC regulator controls and adjusts all the vacuumed after purging so that any amount of air remaining inside
functions. It also records all the measurements performed during ex- them could be removed. Afterwards, the cells were filled with CH4 to
periment. The PLC controller is linked with the software for data pro- prescribed value. The pressure was selected in the range of
cessing. The data is recorded with an interval of 10 s. All the cells 3.4–8.30 MPa since these pressure conditions are frequently en-
contain glass balls within. It is useful to have glass balls inside the cells countered in the natural gas transmission line. Fast cooling technique
as these balls facilitate during shaking. The rotating angle of the cell is augments hydrate formation process as the temperature is reduced
30°. This angle (30°) is sufficient to move the glass ball from one end to swiftly. The temperature of system was maintained for 4 h after cooling
the other in order to enable proper mixing of gas and liquid phase. The in order to ensure the sufficient materialization of gas hydrates. The
diameter of each glass ball is about 1.8 cm. hydrate formation was indicated by a sudden pressure drop detected in
the recorded data. The system was heated stepwise after complete
2.3. Experimental procedure formation of gas hydrates as no more pressure drop occurred.
Fig. 2 shows the Pressure-Temperature diagram for hydrate forma-
The HLVE of CH4 hydrates is determined through an isochoric tion and dissociation which involves the cooling curve, pressure drop
Thermodynamic cycle or shortened as T-cycle method. This method due to hydrate formation, complete hydrate formation, fast heating and
involves the hydrate formation and dissociation process. In hydrate step heating processes for pure water at 4.55 MPa. The point where the
formation process, system temperature is decreased gradually allowing heating curve meets the cooling curve is hydrate liquid vapor equili-
the hydrates to form. Hydrate formation is marked by a sudden drop in brium (HLVE) point. The P-T figures for TMAB, TEAB, TEAI,
pressure. Hydrate dissociation is a slower gradual process. For hydrate TMAB + MEG, TEAB + MEG at 4.6 MPa for 10 wt% solution mixtures
dissociation step heating method is employed where the temperature is has been provided as Supplementary information.
increased stepwise with an increment of 0.5 K at each step with ade-
quate holding time. The holding is set to attain steady state equilibrium.
After experimentation a Pressure-Temperature phase diagram is ob- 2.4. Average suppression temperature (Ŧ) and hydrate dissociation enthalpy
tained. The point where heating and cooling curves meet is the equi- (ΔHdiss)
librium point. In presence of THIs, equilibrium point is shifted to the
left side of hydrate phase boundary which is a region of relatively In order to analyze the degree of thermodynamic inhibition ob-
higher pressure and lower temperature. Before starting the experi- tainable by using QASs, inhibition capacities of TEAI, TEAB and TMAB
ments, rocking cells were washed with distilled water and then dried. as pure compounds and as well as in mixture with MEG on CH4 hy-
All the cells were filled with 15 ml solutions of selected concentrations. drates are determined. The average suppression temperature (Ŧ) for
The solutions were prepared for pure compounds and in mixture with these chemicals is also examined at different concentrations of 1, 5 and
MEG also. After that, cooling of the system was performed to reach up 10 wt%. Eq. (1) is used to calculate the suppression temperature (ΔT).
to operational temperature. The cells were then filled with some

Fig. 1. Experimental apparatus.

3
A. Qasim, et al. Fuel 259 (2020) 116219

Fig. 2. Pressure-Temperature profile for water + CH4 mixture.

n
ΔT ∑ (T0, pi − T1, pi ) ∑i xip xi (σ )
T= = i=1 ps (σ ) =
n n (1) ∑i xi (3)

where T0, pi is the temperature at equilibrium conditions for gas in a The distribution of the division given on the sigma (σ) is called σ-
blank sample which does not contain any compound. The equilibrium profile (ps(σ)). The σ-profile of the solvent ps(σ) is defined as the mole
temperature of gas containing QASs is denoted as T1, pi. In both cases fraction (xi) weighted sum of the σ -profiles of its compounds xi, pxi
the dissociation temperatures must be determined at the same pressure respectively in Eq. (3).
values. Symbol ‘n’ represents the number of points for pressure con-
sidered in the experimentation. 2.6. Theoretical understanding of thermodynamic modeling
ΔHdiss signifies the gas hydrate dissociation enthalpy. Dissociation
enthalpies are figured out by employing Clausius-Clapeyron equation. In order to determine the influence of additional chemicals and salts
For this purpose, derivative of the experimental HLVE data is obtained. on water activity, different thermodynamic models are suggested and
Dissociation enthalpy is calculated by finding slope of hydrate liquid available. In this article, a thermodynamic model given by Dickens and
vapor equilibrium data using Clausius-Clapeyron equation. Preceding Quinby Hunt [60] is applied to determine the hydrate liquid vapor
researches have adopted the same approach and reported about it equilibrium for CH4 hydrates in case of pure compounds and QAS-MEG
[13,41,62]. solution mixtures. The aforementioned model is an extended version of
model used by Pieroen [65]. Partoon and others have used this model
d ln P ΔHdiss
1
= previously in case of amino acids and ionic liquid solutions [30,61]. In
dT zR (2) the presence of CO2, this particular model has accurately predicted the
boundaries for hydrate phases [66]. The model has also successfully
Wherein T is the equilibrium temperature and P is pressure at which been applied for mixed gas combination of CO2-CH4 as (70/30)% and
equilibrium temperature is inspected. Universal gas constant is pre- (30/70)% in case of ammonium based ionic liquids (AILs) [36]. It was
sented as R and z is the compressibility factor of the gas used. established that quaternary ammonium compounds taken into con-
sideration influenced the hydrate phase boundary by decreasing the
water activity (aw). It was also found that ΔHdiss almost remained
2.5. COSMO-RS analysis of QAS-water system
constant for shorter ranges of temperature. All particulars about the
model can be found in the works of Dickens et al. and Pieroen [60,65].
To understand the inhibition mechanism of THI, molecular inter-
The effect on hydrate dissociation temperature due to the presence
action is observed by obtaining sigma profile of QAS and water mole-
of inhibitors for methane hydrates can be represented according to the
cules. The quantum chemistry-based software namely Conductor-like
following relation:
Screening Model for Real Solvents (COSMO-RS) is used to determine
the sigma profile of the compounds taken into consideration. Sigma ΔHdiss
ln a w = [1/ Tw − 1/ Tinhibitors]
profile shows the charge distribution on the surface of QAS. This charge nR (4)
distribution is an indicator of the hydrogen bonding and hydrophilicity
ΔHFUS (i)
of the compounds. The COSMO-RS predictions were performed using ln a w = [1/ Tfi − 1/ Tf ]
R (5)
COSMOthermX, Version C3.0, release 17.01 software. In order to obtain
the sigma profiles, molecules of water and QAS were selected from the where aw shows the water activity, n is hydration number for CH4
list with the parameter file BP_TZVP_C21_0111.ctd (COSMOlogic GmbH hydrates. The value of n is 5.75 as reported in previous study [41].
& Co KG, Leverkusen, Germany) by the lowest energy conformer ΔHdiss for CH4 hydrates is found to be 57.44 kJ/mol. R is universal gas
[63,64]. constant and Tw represents the hydrate formation temperature. It is

4
A. Qasim, et al. Fuel 259 (2020) 116219

used for both types of solution mixtures i.e. in the absence and presence thermodynamic inhibition performance with an average suppression
of MEG. ΔHFUS(i) is the heat of ice fusion and its value is 6.008 KJ/mol. temperature of 1.5 K [57]. It was due to the presence of OH− anion
The freezing point temperatures of water (273.15 K) without the ad- which showed exceptional capability to form hydrogen bonding with
dition of inhibitors is written as Tf and with addition of inhibitors it is water. In this work, the suppression performance of TMAB is high and
represented as Tf(i). Dickens et al. [60] stated the value of cryoscopic comparable with TMAOH as it showed the average value of 1.34 K at
constant to be 1.853 K.kg/mol for the calculation of Tf. The in- 10% concentration.
corporation of Eqs. (3) and (4) expresses the temperature at constant The thermodynamic inhibition effect of QAS-MEG mixtures on hy-
pressure where the hydrate phase of CH4 is in equilibrium along with drate formation has also been calculated and reported in Fig. 5. The
the equilibrium conditions and ice-water phase temperature for QAS/ effect of TEAB and TMAB in combination with MEG has been taken into
QAS-MEG mixtures. Eq. (5) shows the calculation. account. The obtained results demonstrate that QAS-MEG mixtures
performed better than pure QACs. QAS-MEG mixtures shifted the
⎡1 − 1 ⎤ = nΔHFUS (i) [1/ Tfi − 1/ Tf ] equilibrium temperature more towards the left of the graph i.e., lower
⎢ T
⎣ w Tinhibitors ⎥
⎦ ΔHdiss (6) side at the same experimental pressure values used in the scheme of
In the presence of inhibitors, the temperature of hydrate dissocia- pure compounds. The inhibition performance of MEG and QAS is ob-
tion can be examined using Eq. (5). Average absolute error (AAE) is also served due to their ability to form hydrogen bond with water. For
taken into consideration as it is an indicator of the suitability of the TMAB-MEG and TEAB-MEG mixtures, the inhibition shown by them is
model. Eq. (6) shows the calculation for AAE. greater than pure quaternary ammonium compounds but still the effect
of inhibition observed is not synergic. Fig. 4 shows that the 10 wt%
n
1 TMAB-MEG mixture accomplished better inhibition effect as compared
AAE =
n
∑ [TExp − TCal ]i to the other compounds considered in this study. It is due to the reason
i=1 (7)
that TMAB and MEG themselves in their individual capacities act as
effective THIs. So in a situation where these compounds are added to-
3. Results and discussion gether, an effective inhibition effect is observed.

3.1. Measurement for thermodynamic hydrate inhibition 3.2. Average suppression temperature

THI measurement comprises the results regarding hydrate liquid Table 3 shows the quantified effect of QACs on thermodynamic
vapor equilibrium points. The significance of HLVE is due to simulta- inhibition of hydrate formation for methane hydrates. For this purpose,
neous coexistence of all phases. Table 2 shows all the points obtained at average suppression temperature (ΔT) is reported for 5 and 10 wt%
equilibrium for QASs mentioned in Table 1. The equilibrium points are QACs solutions. Among these compounds, The THI influence of TMAB
considered within the pressure range of 3.4–8.3 MPa. In this selected is found to be of highest order followed by TEAB. The influence of TEAI
range of pressure values, equilibrium is reached within the temperature as THI is less than TMAB and TEAB. Table 3 shows that TEAI offered the
range of 274–284.6 K. suppression up to an extent of 0.45 K at 5 wt% while 0.82 K at 10 wt%
The HLVE points of methane hydrates for pure QACs which are in comparison to that of water. TEAB suppressed the hydrate formation
mentioned in Table 2 are shown in Fig. 3. The HLVE points are de- to an extent of 0.55 K at 5 wt% and 1.07 K at 10 wt%. For TMAB, the
termined at 1, 5 and 10 wt% of solution concentrations. The results degree of suppression achieved at 5 wt% is 0.76 K and 1.34 K at 10 wt%.
from Fig. 3 proved that QASs have the ability to show thermodynamic Among these inhibitors, TMAB clearly showed the maximum amount of
influence by moving HLVE in lower temperature regions in comparison suppression at each concentration (At 5 wt% ΔT = 0.76 K; At 10 wt%,
to that of pure water HLVE curve. For CH4 hydrates, the THI effect of ΔT = 1.34 K) which is ascribed to the presence of shortest alkyl chain
QASs is concentration dependent. With the increase in concentration of among all QACs. Previous results show the same trend as obtained in
solution, activity of compounds in aqueous phase is also increased. The here. Results agree with the previous literature data available for ionic
increased activity is responsible for higher inhibition at 10 wt% con- liquids and QACs. Preceding works showed that the length of alkyl
centration in comparison to that of 5 wt%. In order to quantify THI and chain plays an important role and affects the phase stability of hydrate
examine the effect of pressure on phase behavior, Fig. 3 illustrates the formation [17]. Alkyl chain puts influence as larger chain carries hy-
HLVE data of various compositions of CH4-QAS-H2O at different ex- drophobicity and prohibits cation in QAS to have a bond with the hy-
perimental pressures. According to it, THI impact of CH4-QAS-H2O droxyl ion. Same trend was observed for ILs containing imidazolium in
hydrates enhanced with increased concentration as it is varied from 1 to preceding studies done by Yaqub et al. [16] and Xiao et al. [68]. It was
5 and 10 wt%, respectively. The ability to suppress hydrate formation found that thermodynamic inhibition capacity of IL with the same
emanates due to the formation of hydrogen bonds of cations and anions anion decreased as the alkyl chain length increased from ethyl to butyl
present in quaternary ammonium compounds with water. group. It reaffirmed the findings of the previous studies wherein re-
Moreover Fig. 4 compares the performance of studied QAS systems searchers found that THI capacity of ionic liquids with shorter alkyl
with earlier reported ILs and QAS system. As observed from the Fig. 4, chain showed better results in comparison to that of longer alkyl chain
the inhibition performance of QASs is compared and found to be in line keeping the anion constant. Khan et al. [7] observed the same effect as
with the systems reported in the literature [37,41,57,67,68,9,69]. The the ILs employed for study revealed the decrease in inhibition cap-
HLVE points obtained in this experimental work are compared with the ability with the increase in alkyl chain length. In case of methane hy-
results from some previously available literature on thermodynamic drates, Tariq et al. studied the effect of ammonium based ionic liquids
inhibition of pure water, some ILs and QASs. These include TMACl, on depression temperature and found that results were in compliance
TMAOH, EMMor-Br, EMMor-BF4, EMPip-Br, EMPip-BF4 and BMIM-BF4 with the identified behavior [70]. TEAB showed better suppression
in presence of CH4 as shown in Fig. 4. In order to compare the results, performance as compared to TEAI.
maximum concentration of 10 wt% is taken into consideration. Pre- Table 4 shows the average suppression temperature for TEAB-MEG
ceding literature involved piperidinium, morpholinium and imidazo- and TMAB-MEG solution mixtures. The collective performance of QAS-
lium based compounds while quaternary ammonium salts are ammo- MEG solutions is found to be better than pure QAS solution mixtures. It
nium based compounds. The comparison of ammonium based can also be seen that the inhibition performance increases with the
compounds with ILs used in preceding studies can provide a useful increase in concentration. The trend obtained for solution mixtures is
insight regarding their inhibition performance. It can be noticed from identical as observed in pure compounds. As expected, TMAB also ex-
the previous studies that TMAOH as a pure compound offered the best hibited better inhibition behavior in mixture with MEG as it showed in

5
A. Qasim, et al. Fuel 259 (2020) 116219

Table 2
HLVE data at 1, 5 and 10 wt% inhibitors for CH4 gas with and without MEG.
Water

T (K) P (MPa)

276.2 3.45
278.95 4.55
282.55 6.5
283.58 7.35
284.6 8.21

10 wt% TEAI 5 wt% TEAI 1 wt% TEAI

T (K) P (MPa) T (K) P (MPa) T (K) P (MPa)
275.5 3.5 275.8 3.5 276 3.5
278.1 4.6 278.5 4.6 278.7 4.6
281.7 6.5 282.1 6.5 282.3 6.5
282.7 7.4 283.12 7.4 283.25 7.4
283.8 8.2 284.1 8.2 284.3 8.2
10 wt% TMAB 5 wt% TMAB 1 wt% TMAB
T (K) P (MPa) T (K) P (MPa) T (K) P (MPa)
274.9 3.45 275.5 3.45 275.8 3.5
277.7 4.6 278.2 4.6 278.55 4.6
281.1 6.5 281.6 6.5 282.05 6.5
282.3 7.4 282.9 7.4 283.1 7.4
283.2 8.25 283.9 8.25 284.15 8.25
10 wt% TEAB 5 wt% TEAB 1 wt% TEAB
T (K) P (MPa) T (K) P (MPa) T (K) P (MPa)
275.05 3.5 275.65 3.5 275.9 3.5
278 4.6 278.4 4.6 278.6 4.6
281.5 6.5 282 6.5 282.2 6.5
282.5 7.4 283 7.4 283.2 7.4
283.5 8.2 284.1 8.2 284.25 8.2
10 wt% TMAB-MEG 5 wt% TMAB-MEG 1 wt% TMAB-MEG
T (K) P (MPa) T (K) P (MPa) T (K) P (MPa)
274 3.45 274.6 3.45 275.2 3.5
276.5 4.6 277.3 4.6 277.9 4.6
280.2 6.5 280.9 6.5 281.5 6.5
281.2 7.4 281.9 7.4 282.4 7.5
282.2 8.2 282.9 8.2 283.6 8.3
10 wt% TEAB-MEG 5 wt% TEAB-MEG 1 wt% TEAB-MEG
T (K) P (MPa) T (K) P (MPa) T (K) P (MPa)
274.35 3.5 275 3.45 275.4 3.45
277 4.55 277.7 4.6 278.1 4.6
280.5 6.5 281.3 6.5 281.7 6.5
281.5 7.4 282.3 7.4 282.6 7.4
282.6 8.2 283.3 8.2 283.8 8.2

Expanded uncertainties U(T) = ± 0.1 K, U(P) = ± 0.01 MPa, U(mass fraction) = ± 0.0001 g (0.95 level of confidence).

Fig. 3. HLVE data for TEAI, TMAB and TEAB at studied concentrations (1, 5 wt% and 10 wt%).

6
A. Qasim, et al. Fuel 259 (2020) 116219

Fig. 4. Comparison of studied QASs with literature reported ILs and QAS systems for 10 wt% concentration.

Fig. 5. HLVE data of TMAB-MEG and TEAB-MEG systems at 1, 5 wt% and 10 wt%.

pure form. with the suppression value of 4.21 K. In litereature, better performance
For comparitive analysis, Fig. 6 represents the average suppression of methanol and MEG in terms of THI is attributed to the attraction
(Ŧ) temperaure at 10 wt% of the studied sytems with with the com- between inhibitor oxygen atoms and surrounding water molecules [69].
mercial thermodynamic inhibitors (methanol and MEG) obtained from The presence of OH− anion attached with a short alkyl chain of methyl
commercial prediction software which is CSMGem. According to results in methanol structure accounts for its better results in THI applications.
shown in Fig. 6, the inhibition performance of methanol is unsurpassed The inhibition performance shown by TMAB-MEG mixture is closer to

Table 3
Average suppression temperatures (Ŧ) of TMAB, TEAB and TEAI at different experimental pressure conditions.
ΔT (K) TMAB ΔT (K) TEAB ΔT (K) TEAI

P (MPa) 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt%

3.45 0.40 0.70 1.30 0.30 0.55 1.15 0.20 0.40 0.70
4.55 0.40 0.75 1.25 0.35 0.55 0.95 0.25 0.45 0.85
6.5 0.50 0.95 1.45 0.35 0.55 1.05 0.25 0.45 0.85
7.35 0.48 0.68 1.28 0.38 0.58 1.08 0.33 0.46 0.86
8.21 0.45 0.70 1.40 0.35 0.50 1.10 0.30 0.50 0.85
Ŧ (K) 0.45 0.76 1.34 0.35 0.55 1.07 0.27 0.45 0.82

7
A. Qasim, et al. Fuel 259 (2020) 116219

Table 4 profile generated of ions and MEG is shown in Fig. 7.

Average suppression temperatures (ΔT) of QAS-MEG systems at various pres- Sigma profile displays important information on the hydrophilicity
sures. and the hydrogen bond acceptor or donor tendency of an ionic com-
ΔT (K) (TMAB-MEG) ΔT (K) (TEAB-MEG) pound [56,71]. According to Klamt [64], three distinct and interactive
regions can be identified in sigma profile. The region mentioned on the
P (MPa) 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt% left side of the profile is electropositive. The electropositive region is
hydrogen bond donor region. The nonpolar area lies intermediately
3.45 1.00 1.60 2.20 0.80 1.20 1.85
4.55 1.05 1.65 2.45 0.85 1.25 1.95 which is between −0.01 e/Å2 and 0.011 e/ Å2. The right hand side of
6.5 1.05 1.65 2.35 0.85 1.25 2.05 Fig. 7 represents highly electronegative region. It is hydrogen bond
7.35 1.18 1.68 2.38 0.98 1.28 2.08 acceptor region. COSMO-RS results showed that QASs with similar
8.21 1.00 1.70 2.40 0.80 1.30 2.00 hydrogen bond donor and acceptor affinities can interact efficiently
Ŧ (K) 1.06 1.66 2.36 0.86 1.26 1.99
with water molecules [72,73]. Similarly, for stronger intermolecular
interactions, ionic compounds should hold lower peaks in the nonpolar
pure MEG with the value of 2.36 K but the effect is not greater than area and retain broader peaks at the hydrogen bond acceptor and donor
MEG itself which shows the suppression of 2.65 K. areas [64].
Table 5 demonstrates the enthalpy of hydrate formation for all the According to the Sigma profile obtained in Fig. 7 it shows that QAS
systems taken into consideration in this article. Preceding works re- with shortest alkyl chain which is TMA+ exhibits the highest hydrogen
ported that ΔHdiss depends upon two main factors which include cage bond donor ability. It is due to the reason that it lies in the hydrogen
occupancy of gas molecules and the ability of clathrate structure to bond donor region with least peak difference (21.82 e/Å2) among water
form hydrogen bond [4]. As results show that dissociation enthalpy in nonpolar regions. The peak difference among QAS and water extends
values do not change considerably so it can be argued that the com- further with the increase in alkyl chain length. The hydrogen bond
pounds used for inhibition did not contribute significantly in the pro- affinity of cations is found to be in the order of TMA+ > TEA+. It
cess of hydrate crystallization. The results revealed for QAS and QAS- shows that tetramethylammonium cation is more suitable in thermo-
MEG solution mixtures in case of CH4 hydrates provide an indication of dynamic hydrate inhibition applications as it shows higher hydrogen
structure I hydrates. The thermodynamic inhibition influence of these bonding affinity.
compounds in pure as well as in mixture is due to their hydrogen Similarly Fig. 7 also shows that all the considered anions show clear
bonding capability. peaks in hydrogen bond acceptor regions. The bromide and iodide
anions hold strong peaks in the polar hydrogen bond acceptor regions.
For anions the small difference in peaks with higher hydrogen bond
3.3. COSMO-RS analysis of QAS-water system acceptor affinity marks a better performance in terms of THI applica-
tions. The hydrogen bond donor affinity of bromide anion is relatively
COSMO-RS analysis is performed to observe the hydrogen bonding greater than iodide. The bromide and iodide anions are used for ther-
affinity of QAS and water molecules. For cations, only shorter alkyl modynamic hydrate inhibition in this work as they maintain strong
chain cations are considered. The cations include tetra- electrostatic charges and have the ability to form hydrogen bond with
methylammonium (TMA+), tetraethylammonium (TEA+). These ca- water.
tions are taken into consideration because shorter cations are suitable
in gas hydrate mitigation especially in THI applications [8,50]. Ad-
ditionally, in gas hydrate mitigation applications, anions are required to 3.4. Thermodynamic modeling for pure QAS and QAS-MEG solution
form sufficient hydrogen bonding with water molecules [13,55,57]. mixtures
The anions prescreened in this work are mainly composed of the halide
group which included bromide [Br]−, and iodide [I] −. The sigma A comparison of hydrate liquid vapor equilibrium points for CH4 gas

Fig. 6. Comparison of the studied QAS and QAS-MEG systems on methane hydrate dissociation temperature with MEG and methanol (predicted by CSMGem).

8
A. Qasim, et al. Fuel 259 (2020) 116219

Table 5
ΔHdiss data for all the studied systems (QAS-CH4 and QAS-MEG-CH4).
TMAB TEAB TEA1

Pressure (MPa) CH4 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt%

3.45 60.69 60.22 60.52 60.72 60.68 60.55 60.76 61.16 60.75 60.97
4.55 59.13 58.68 58.97 59.14 59.10 58.97 59.18 59.58 59.18 56.47
6.5 56.67 56.26 56.51 56.65 56.71 56.54 56.70 56.17 56.76 56.91
7.35 55.76 55.28 55.53 55.67 55.72 55.57 55.73 55.33 55.77 55.92
8.21 54.95 54.53 54.73 54.85 54.96 54.79 54.92 55.34 54.99 55.13
ΔHdiss (kJ/mol) 57.44 56.99 57.25 57.41 57.43 57.28 57.46 57.52 57.49 57.08

TMAB-MEG TEAB-MEG

Pressure (MPa) CH4 1 wt% 5 wt% 10 wt% 1 wt% 5 wt% 10 wt%

3.45 60.69 60.39 60.17 60.82 60.57 60.41 60.34

4.55 59.13 58.79 58.61 59.21 58.88 58.80 58.72
6.5 56.67 56.37 56.15 56.70 56.48 56.35 56.26
7.35 55.76 55.39 55.15 55.70 55.52 55.37 55.26
8.21 54.95 54.63 54.36 54.89 54.74 54.67 54.53
ΔHdiss (kJ/mol) 57.44 57.12 56.88 57.46 57.24 57.12 57.02

Expanded uncertainties U(T) = ± 0.1 K, U(P) = ± 0.01 MPa, U(mass fraction) = ± 0.0001 g (0.95 level of confidence).

Fig. 7. Sigma profile generated by COSMO-RS software for anions and cations.

hydrates present in solution mixtures of pure QACs are shown in Fig. 8. The AAE values for TMAB-MEG and TEAB-MEG mixtures are also
The equilibrium point values include both experimental and predicted calculated and shown in Table 7. In case of TMAB-MEG mixture, at 1, 5,
by the model in between the pressure range of 3.4–8.3 MPa. The results 10 wt%, AAE values are obtained to be 0.8 K, 0.35 K and 0.25 K re-
disclose that experimental and model predicted values show a good spectively as depicted in Table 7. For TEAB-MEG mixture, AAE values
agreement with each other. are found to be 0.62 K, 0.06 K and 0.41 K at 1, 5 and 10 wt% respec-
Average Absolute Error (AAE) values for pure QACs at all con- tively. Table 7 shows that all the R-squared values obtained for QAS-
centrations are shown in Table 6. For TEAI, AAE values at 1, 5 and MEG mixtures are greater than 0.998. The high values of R2 which are
10 wt% are 0.20 K, 0.14 K and 0.2 K respectively. The AAE values cal- closer to 1 show a very good agreement between experimental and
culated in case of TMAB at 1, 5 and 10 wt% are obtained to be 0.24 K, modeling data.
0.24 K and 0.30 K respectively. For TEAB, the values of AAE are at-
tained as 0.27 K, 0.17 K and 0.31 K at 1, 5 and 10 wt% respectively.
From Table 6, it can be noticed that R-squared values are higher than 4. Conclusion
0.998 in all cases which shows a good fit and modeling results comply
with the experimental results. The THI performance of three quaternary ammonium salts which
A comparison between experimental and predicted data for TMAB- include TEAI, TMAB and TEAB is evaluated for CH4 hydrates at high
MEG and TEAB-MEG solution mixtures is presented in Fig. 9 at 1, 5 and pressure range of 3.4–8.3 MPa. Thermodynamic hydrate inhibition is
10 wt% concentrations. Fig. 9 shows that the model accurately pre- determined for pure compounds and along with MEG in case of TMAB
dicted the equilibrium temperature as both the values are close to each and TEAB. The inhibition effect shown by TMAB is found to be un-
other at the selected pressure range. Average Absolute Error (AAE) and surpassed both in pure and in mixture with MEG. The average sup-
R-squared values are also calculated and reported in Table 7. pression temperature for TMAB as a pure compound is achieved to be
1.34 K at the maximum concentration of 10 wt%. In mixture with MEG,

9
A. Qasim, et al. Fuel 259 (2020) 116219

9.0
Water
1 % TEAI
1 % TEAI Predicted
8.0 5 % TEAI
5 % TEAI Predited
10 % TEAI
10 % TEAI Predicted
1 % TMAB

Pressure (MPa)
7.0
1 % TMAB Predicted
5 % TMAB
5 % TMAB Predicted
6.0 10 % TMAB
10 % TMAB Predicted
1 % TEAB
1 % TEAB Predicted
5 % TEAB
5.0 5 % TEAB Predicted
10 % TEAB
10 % TEAB Predicted
4.0

3.0
274 276 278 280 282 284 286
Temperature (K)
Fig. 8. CH4 hydrates model predicted and experimental HLVE data at 1, 5 and 10 wt%.

Table 6 Table 7
Model predicted equilibrium data (parameters) for QAS systems in case of CH4 The Model predicted HLVE data (parameters) for QAS-MEG systems in the
hydrates. presence of CH4 hydrates.
1 wt% 1 wt%
TEAI TMAB TEAB TMAB-MEG TEAB-MEG
Temperature Range (K) 276.14–284.54 275.8–284.15 276.13–284.52 Temperature Range (K) 275.95–284.33 275.97–284.36
AAE (K) 0.2 0.24 0.27 AAE (K) 0.80 0.62
No. of data points 5 5 5 No. of data points 5 5
R2 Value 0.9987 0.9987 0.9987 R2 Value 0.9989 0.9987
5 wt% 5 wt%
TEAI TMAB TEAB TMAB-MEG TEAB-MEG
Temperature Range (K) 275.90–284.28 275.5–283.9 275.83–284.21 Temperature Range (K) 274.94–283.27 275.05–283.38
AAE (K) 0.14 0.24 0.17 AAE (K) 0.35 0.06
No. of data points 5 5 5 No. of data points 5 5
R2 Value 0.9986 0.9987 0.9987 R2 Value 0.9991 0.9991
10 wt% 10 wt%
TEAI TMAB TEAB TMAB-MEG TEAB-MEG
Temperature Range (K) 275.60–283.96 275.2–283.54 275.47–283.82 Temperature Range (K) 273.36–281.93 273.89–282.15
AAE (K) 0.2 0.3 0.31 AAE (K) 0.25 0.41
No. of data points 5 5 5 No. of data points 5 5
R2 Value 0.9986 0.9986 0.9987 R2 Value 0.9987 0.9987

Expanded uncertainties U(T) = ± 0.1 K, U(P) = ± 0.01 MPa, U(mass fraction) Expanded uncertainties U(T) = ± 0.1 K, U(P) = ± 0.01 MPa, U(mass fraction)
= ± 0.0001 g (0.95 level of confidence). = ± 0.0001 g (0.95 level of confidence).

Fig. 9. Model predicted and experimental HLVE data for CH4 hydrates in presence of different concentrations of aqueous QAS-MEG solutions (1, 5 and 10 wt%).

10
A. Qasim, et al. Fuel 259 (2020) 116219

the average suppression temperature of TMAB is 2.36 K at 10 wt%. The [15] Akhfash M, Arjmandi M, Aman ZM, Boxall JA, May EF. Gas hydrate thermodynamic
better inhibition effect is observed due to the presence of shorter alkyl inhibition with MDEA for reduced MEG circulation. J Chem Eng Data
2017;62:2578–83. https://doi.org/10.1021/acs.jced.7b00072.
chain in TMAB. After TMAB, TEAB showed better inhibition in com- [16] Yaqub S, Lal B, Partoon B, Mellon NB. Investigation of the task oriented dual
parison to that of TEAI at respective concentrations. COSMO-RS study function inhibitors in gas hydrate inhibition: a review. Fluid Phase Equilib
showed the suitability of TMA+, TEA+ cations and Br− and I− anions 2018;477:40–57. https://doi.org/10.1016/j.fluid.2018.08.015.
[17] Gupta P, Chandrasekharan Nair V, Sangwai JS. Phase equilibrium of methane hy-
in thermodynamic hydrate inhibition applications. Besides that, hy- drate in the presence of aqueous solutions of quaternary ammonium salts. J Chem
drate dissociation enthalpy (ΔHdiss) values are also calculated for all the Eng Data 2018;63:2410–9. https://doi.org/10.1021/acs.jced.7b00976.
compounds at 1, 5 and 10 wt%. Results revealed that the change in [18] Cha M, Shin K, Kim J, Chang D, Seo Y, Lee H, et al. Thermodynamic and kinetic
hydrate inhibition performance of aqueous ethylene glycol solutions for natural gas.
dissociation enthalpy is up to a very minor extent. It does not vary Chem Eng Sci 2013;99:184–90. https://doi.org/10.1016/j.ces.2013.05.060.
significantly which indicates non-participation of QASs in forming hy- [19] Nashed O, Partoon B, Lal B, Sabil KM, Mohd A. Review the impact of nanoparticles
drate crystal structure. Dickens and Quinby-Hunt model is used to on the thermodynamics and kinetics of gas hydrate formation. J Nat Gas Sci Eng
2018;55:452–65. https://doi.org/10.1016/j.jngse.2018.05.022.
predict HLVE and compared with the experimental results. Higher va-
[20] Nashed O, Koh JCH, Lal B. Physical-chemical Properties of Aqueous TBAOH
lues of AAE show a good fit between experimental and modeling data. Solution for Gas Hydrates Promotion. Procedia Eng 2016;148:1351–6. https://doi.
Therefore, it can be asserted that QACs can be employed in flow as- org/10.1016/j.proeng.2016.06.586.
surance applications for the purpose of thermodynamic hydrate in- [21] Sloan ED, Koh CA, E. Dendy Sloan, Koh CA, Sloan ED, Koh CA, et al. Clathrate
Hydrates of Natural Gases. vol. 87. third. Boca Raton; London; New York: CRC Press
hibition. Taylor & Francis; 2008.
[22] Ward ZT, Deering CE, Marriott RA, Sum AK, Sloan ED, Koh CA. Phase equilibrium
Acknowledgement data and model comparisons for H 2 S hydrates. J Chem Eng Data 2015;60. https://
doi.org/10.1021/je500657f.
[23] Abolala M, Varaminian F. Thermodynamic model for predicting equilibrium con-
All the authors acknowledge Universiti Teknologi PETRONAS, ditions of clathrate hydrates of noble gases + light hydrocarbons: Combination of
Malaysia for providing necessary experimental facilities required for Van der Waals-Platteeuw model and SPC-SAFT EoS. J Chem Thermodyn
2015;81:89–94. https://doi.org/10.1016/j.jct.2014.09.013.
this work. We are specially obliged to the chemical engineering de- [24] Avula VR, Gardas RL, Sangwai JS. An efficient model for the prediction of CO2
partment of Universiti Teknologi PETRONAS (YUTP Grant no. 015LCO- hydrate phase stability conditions in the presence of inhibitors and their mixtures. J
154) to cater for the financial support throughout the study. Chem Thermodyn 2015;85:163–70. https://doi.org/10.1016/j.jct.2015.01.009.
[25] Sa JH, Lee BR, Park DH, Han K, Chun HD, Lee KH. Amino acids as natural inhibitors
for hydrate formation in CO2 sequestration. Environ Sci Technol 2011;45:5885–91.
Appendix A. Supplementary data https://doi.org/10.1021/es200552c.
[26] Corak D, Barth T, Høiland S, Skodvin T, Larsen R, Skjetne T. Effect of subcooling
and amount of hydrate former on formation of cyclopentane hydrates in brine.
Supplementary data to this article can be found online at https://
Desalination 2011;278:268–74. https://doi.org/10.1016/j.desal.2011.05.035.
doi.org/10.1016/j.fuel.2019.116219. [27] Khan MN, Warrier P, Peters CJ, Koh CA. Advancements in hydrate phase equilibria
and modeling of gas hydrates systems. Fluid Phase Equilib 2018;463:48–61.
References https://doi.org/10.1016/j.fluid.2018.01.014.
[28] Khan MS, Lal B, Partoon B, Keong LK, Bustam AB, Mellon NB. Experimental
Evaluation of a Novel Thermodynamic Inhibitor for CH4 and CO2 Hydrates.
[1] Sloan D, Koh C, Sum A. Natural gas hydrates in flow assurance. Gulf Professional Procedia Eng 2016;148:932–40. https://doi.org/10.1016/j.proeng.2016.06.433.
Publishing 2010. [29] Khan MS, Lal B, Bavoh CB, Keong LK, Bustam A, Mellon NB. Influence of ammo-
[2] E. Dendy Sloan, Koh CA. Gas Hydrates of Natural Gases. third. London; New York: nium based compounds for gas hydrate mitigation: a short review. Indian J Sci
CRC Press LLC, 2000 Corporate Blvd., N.W., Boca Raton, FL 33431, USA Orders Technol 2017;10:1–6. https://doi.org/10.17485/ijst/2017/v10i5/99734.
from the USA and Canada (only) to CRC Press LLC; 2008. [30] Partoon B, Wong NMS, Sabil KM, Nasrifar K, Ahmad MR. A study on thermo-
[3] Zerpa LE, Salager J-L, Koh CA, Sloan ED, Sum AK. Surface chemistry and gas hy- dynamics effect of [EMIM]-Cl and [OH-C2MIM]-Cl on methane hydrate equilibrium
drates in flow assurance. Ind Eng Chem Res 2011;50:188–97. https://doi.org/10. line. Fluid Phase Equilib 2013;337:26–31. https://doi.org/10.1016/j.fluid.2012.09.
1021/ie100873k. 025.
[4] Koh CA, Sloan ED, Sum AK, Wu DT. Fundamentals and applications of gas hydrates. [31] Circone S, Stern LA, Kirby SH, Durham WB, Chakoumakos BC, Rawn CJ, et al. CO2
Ann Rev Chem Biomolec Eng 2011;2:237–57. https://doi.org/10.1146/annurev- Hydrate: synthesis, composition, structure, dissociation behavior, and a comparison
chembioeng-061010-114152. to structure I CH4 hydrate. J Phys Chem 2003:5529–39. https://doi.org/10.1021/
[5] Zanota ML, Dicharry C, Graciaa A. Hydrate plug prevention by quaternary ammo- jp027391j.
nium salts. Energy Fuels 2005;19:584–90. https://doi.org/10.1021/ef040064l. [32] Sa J-H, Kwak G-H, Han K, Ahn D, Cho SJ, Lee JD, et al. Inhibition of methane and
[6] Park J, Lee H, Seo Y, Tian W, Wood CD. Performance of polymer hydrogels in- natural gas hydrate formation by altering the structure of water with amino acids.
corporating thermodynamic and kinetic hydrate inhibitors. Energy Fuels Sci Rep 2016;6:1–9. https://doi.org/10.1038/srep31582.
2016;30:2741–50. https://doi.org/10.1021/acs.energyfuels.5b02978. [33] Qureshi MF, Atilhan M, Altamash T, Tariq M, Khraisheh M, Aparicio S, et al. Gas
[7] Khan MS, Lal B, Shariff AM, Mukhtar H. Ammonium hydroxide ILs as dual-func- hydrate prevention and flow assurance by using mixtures of ionic liquids and sy-
tional gas hydrate inhibitors for binary mixed gas (carbon dioxide and methane) nergent compounds: combined kinetics and thermodynamic approach. Energy Fuels
hydrates. J Mol Liq 2018;274:33–44. https://doi.org/10.1016/j.molliq.2018.10. 2016;30:3541–8. https://doi.org/10.1021/acs.energyfuels.5b03001.
076. [34] Sloan ED. A changing hydrate paradigm - From apprehension to avoidance to risk
[8] Khan MS, Bavoh CB, Partoon B, Nashed O, Lal B, Mellon NB. Impacts of ammonium management. Fluid Phase Equilib 2005;228–229:67–74. https://doi.org/10.1016/j.
based ionic liquids alkyl chain on thermodynamic hydrate inhibition for carbon fluid.2004.08.009.
dioxide rich binary gas. J Mol Liq 2018;261:283–90. https://doi.org/10.1016/j. [35] Lim D, Park S, Ro H, Hyery JL, Minchul K, Huen K. Thermodynamic and Kinetic
molliq.2018.04.015. Effect of a Dual-Function Inhibitor on Gas Hydrate Formation. Int Offshore Polar
[9] Nashed O, Dadebayev D, Khan MS, Bavoh CB, Lal B, Shariff AM. Experimental and Eng 2014;3:23–8.
modelling studies on thermodynamic methane hydrate inhibition in the presence of [36] Khan MS, Lal B, Keong LK, Sabil KM. Experimental evaluation and thermodynamic
ionic liquids. J Mol Liq 2018;249:886–91. https://doi.org/10.1016/j.molliq.2017. modelling of AILs alkyl chain elongation on methane riched gas hydrate system.
11.115. Fluid Phase Equilib 2018;473:300–9. https://doi.org/10.1016/j.fluid.2018.07.003.
[10] Bavoh CB, Khan MS, Ting VJ. The effect of acidic gases and thermodynamic in- [37] Nasir Q, Lau KK, Lal B, Sabil KM. Hydrate dissociation condition measurement of
hibitors on the hydrates phase boundary of synthetic malaysia natural gas IOP. CO2-rich mixed gas in the presence of methanol/ethylene glycol and mixed me-
Conf. Ser. Mater. Sci. Eng. Pap. 2018:1–10. https://doi.org/10.1088/1757-899X/ thanol/ethylene glycol + electrolyte aqueous solution. J Chem Eng Data
458/1/012016. 2014;59:3920–6. https://doi.org/10.1021/je5008313.
[11] Tariq M, Rooney D, Othman E, Aparicio S, Atilhan M, Khraisheh M. Gas hydrate [38] Broni-bediako E, Amorin R, Bavoh CB. Gas Hydrate Formation Phase Boundary
inhibition: a review of the role of ionic liquids. Ind Eng Chem Res Behaviour of Synthetic Natural Gas System of the Keta Basin of Ghana Abstract :
2014;53:17855–68. https://doi.org/10.1021/ie503559k. 2017;10:64–72. doi:10.2174/1874834101701010064.
[12] Atilhan M, Aparicio S, Benyahia F, Deniz E. Natural gas hydrates- Advances in [39] Sloan ED, Koh CA, Sum AK. Gas hydrate stability and sampling: the future as related
natural gas technology. Ind Eng 2012;542. https://doi.org/10.5772/67458. to the phase diagram. Energies 2010;3:1991–2000. https://doi.org/10.3390/
[13] Qasim A, Khan MS, Lal B, Shariff AM. Phase equilibrium measurement and mod- en3121991.
eling approach to quaternary ammonium salts with and without monoethylene [40] Jokandan EF, Naeiji P, Varaminian F. The synergism of the binary and ternary
glycol for carbon dioxide hydrates. J Mol Liq 2019;282:106–14. https://doi.org/10. solutions of polyethylene glycol, polyacrylamide and Hydroxyethyl cellulose to
1016/j.molliq.2019.02.115. methane hydrate kinetic inhibitor. J Nat Gas Sci Eng 2016;29:15–20. https://doi.
[14] Makogon YF, Holditch SA, Makogon TY. Natural gas-hydrates - A potential energy org/10.1016/j.jngse.2015.12.016.
source for the 21st Century. J Pet Sci Eng 2007;56:14–31. https://doi.org/10.1016/ [41] Khan MS, Lal B, Keong LK, Ahmed I. Tetramethyl ammonium chloride as dual
j.petrol.2005.10.009. functional inhibitor for methane and carbon dioxide hydrates. Fuel

11
A. Qasim, et al. Fuel 259 (2020) 116219

2019;236:251–63. https://doi.org/10.1016/j.fuel.2018.09.001. [58] Khan MS, Liew CS, Kurnia KA, Cornelius B, Lal B. Application of COSMO-RS in
[42] Mokhatab S, Wilkens RJ, Leontaritis KJ. A review of strategies for solving gas-hy- investigating ionic liquid as thermodynamic hydrate inhibitor for methane hydrate.
drate problems in subsea pipelines. Energy Sources, Part A Recover Util Environ Eff Procedia Eng 2016;148:862–9. https://doi.org/10.1016/j.proeng.2016.06.452.
2007;29:39–45. https://doi.org/10.1080/009083190933988. [59] Srinivasan S, Kane RD. Critical issues in the application and evaluation of a cor-
[43] Vedachalam N, Ramesh S, Srinivasalu S, Rajendran G, Ramadass GA, Atmanand rosion prediction model for oil and gas systems. NACE - Int Corros Conf Ser
MA. Assessment of methane gas production from Indian gas hydrate petroleum 2003:1–18.
systems. Appl Energy 2016;168:649–60. https://doi.org/10.1016/j.apenergy.2016. [60] Dickens GR, Quinby-Hunt MS. Methane hydrate stability in pore water: a simple
01.117. theoretical approach for geophysical applications. J Geophys Res Solid Earth
[44] Chen W. Status and challenges of Chinese deepwater oil and gas development. Pet 1997;102:773–83. https://doi.org/10.1029/96jb02941.
Sci 2011;8:477–84. https://doi.org/10.1007/s12182-011-0171-8. [61] Bavoh CB, Nashed O, Khan MS, Partoon B, Lal B, Sharif AM. The impact of amino
[45] Bavoh CB, Khan MS, Lal B, Bt Abdul Ghaniri NI, Sabil KM. New methane hydrate acids on methane hydrate phase boundary and formation kinetics. J Chem
phase boundary data in the presence of aqueous amino acids. Fluid Phase Equilib Thermodyn 2018;117:48–53. https://doi.org/10.1016/j.jct.2017.09.001.
2018;478:129–33. https://doi.org/10.1016/j.fluid.2018.09.011. [62] Khan MS, Cornelius BB, Lal B, Bustam MA. Kinetic assessment of tetramethyl am-
[46] Yarveicy H, Kariman A, Ghiasi MM. Practical use of statistical learning theory for monium hydroxide (ionic liquid) for carbon carbon dioxide, methane and binary
modeling freezing point depression of electrolyte solutions : LSSVM model. J Nat mix gas hydrates. Recent Adv. Ion. Liq. 2018:159–79. https://doi.org/10.5772/
Gas Sci Eng 2014;20:414–21. https://doi.org/10.1016/j.jngse.2014.06.020. intechopen.77262.
[47] Ke W, Kelland MA. Kinetic hydrate inhibitor studies for gas hydrate systems: a [63] Diedenhofen M, Klamt A. COSMO-RS as a tool for property prediction of IL mix-
review of experimental equipment and test methods. Energy Fuels tures—a review. Fluid Phase Equilib 2010;294:31–8. https://doi.org/10.1016/j.
2016;30:10015–28. https://doi.org/10.1021/acs.energyfuels.6b02739. fluid.2010.02.002.
[48] Brustad S, Løken KP, Waalmann JG. Hydrate Prevention using MEG instead of [64] Klamt A. COSMO-RS for aqueous solvation and interfaces. Fluid Phase Equilib
MeOH: Impact of experience from major Norwegian developments on technology 2016;407:152–8. https://doi.org/10.1016/j.fluid.2015.05.027.
selection for injection and recovery of MEG. Offshore Technol Conf 2005:10. [65] Pieroen A. Gas hydrates - approximate relations betvveen heat. Recl Des Trav Chim
[49] Hemmingsen PV, Burgass R, Pedersen KS, Kinnari K, Sørensen H. Hydrate tem- Des Pays-Bas 1955;74:995–1002.
perature depression of MEG solutions at concentrations up to 60wt%. Experimental [66] Bavoh CB, Partoon B, Lal B, Gonfa G, Foo Khor S, Sharif AM. Inhibition effect of
data and simulation results. Fluid Phase Equilib 2011;307:175–9. https://doi.org/ amino acids on carbon dioxide hydrate. Chem Eng Sci 2017;171:331–9. https://doi.
10.1016/j.fluid.2011.05.010. org/10.1016/j.ces.2017.05.046.
[50] Xiao C, Wibisono N, Adidharma H. Dialkylimidazolium halide ionic liquids as dual [67] Keshavarz L, Javanmardi J, Eslamimanesh A, Mohammadi AH. Experimental
function inhibitors for methane hydrate. Chem Eng Sci 2010;65:3080–7. https:// measurement and thermodynamic modeling of methane hydrate dissociation con-
doi.org/10.1016/j.ces.2010.01.033. ditions in the presence of aqueous solution of ionic liquid. Fluid Phase Equilib
[51] Sen Li X, Liu YJ, Zeng ZY, Chen ZY, Li G, Wu HJ. Equilibrium hydrate formation 2013;354:312–8. https://doi.org/10.1016/j.fluid.2013.05.007.
conditions for the mixtures of methane + ionic liquids + water. J Chem Eng Data [68] Xiao C, Adidharma H. Dual function inhibitors for methane hydrate. Chem Eng Sci
2011;56:119–23. https://doi.org/10.1021/je100987q. 2009;64:1522–7. https://doi.org/10.1016/j.ces.2008.12.031.
[52] Gupta P, Sakthivel S, Sangwai JS. Effect of aromatic/aliphatic based ionic liquids on [69] Sabil KM, Nashed O, Lal B, Ismail L, Japper-Jaafar A, Japper- A, et al. Experimental
the phase behavior of methane hydrates: Experiments and modeling. J Chem investigation on the dissociation conditions of methane hydrate in the presence of
Thermodyn 2018;117:9–20. https://doi.org/10.1016/j.jct.2017.08.037. imidazolium-based ionic liquids. J Chem Thermodyn 2015;84:7–13. https://doi.
[53] Richard AR, Adidharma H. The performance of ionic liquids and their mixtures in org/10.1016/j.jct.2014.12.017.
inhibiting methane hydrate formation. Chem Eng Sci 2013;87:270–6. [70] Tariq M, Connor E, Thompson J, Khraisheh M, Atilhan M, Rooney D. Doubly dual
[54] Del Villano L, Kelland MA. An investigation into the kinetic hydrate inhibitor nature of ammonium-based ionic liquids for methane hydrates probed by rocking-
properties of two imidazolium-based ionic liquids on Structure II gas hydrate. Chem rig assembly. RSC Adv 2016;6:23827–36. https://doi.org/10.1039/C6RA00170J.
Eng Sci 2010;65:5366–72. https://doi.org/10.1016/j.ces.2010.06.033. [71] Gonfa G, Bustam MA, Sharif AM, Mohamad N, Ullah S. Tuning ionic liquids for
[55] Avula VR, Gardas RL, Sangwai JS. Modeling of methane hydrate inhibition in the natural gas dehydration using COSMO-RS methodology. J Nat Gas Sci Eng
presence of green solvent for offshore oil and gas pipeline. Proc Int Offshore Polar 2015;27:1141–8. https://doi.org/10.1016/j.jngse.2015.09.062.
Eng Conf 2014;3:49–54. https://doi.org/10.1016/j.chembiol.2007.07.009. [72] Kurnia KA, Sintra TE, Neves CMSS, Shimizu K, Canongia Lopes JN, Gonçalves F,
[56] Bavoh CB, Lal B, Nashed O, Khan MS, Keong LK, Bustam MA, et al. COSMO-RS: an et al. The effect of the cation alkyl chain branching on mutual solubilities with
ionic liquid prescreening tool for gas hydrate mitigation. Chinese J Chem Eng water and toxicities. Phys Chem Chem Phys 2014;16:19952–63. https://doi.org/10.
2016;24:1619–24. https://doi.org/10.1016/j.cjche.2016.07.014. 1039/c4cp02309a.
[57] Khan MS, Partoon B, Bavoh CB, Lal B, Mellon NB. Influence of tetra- [73] Kurnia KA, Lima F, Cláudio AFM, Coutinho JAP, Freire MG. Hydrogen-bond acidity
methylammonium hydroxide on methane and carbon dioxide gas hydrate phase of ionic liquids: an extended scale. Phys Chem Chem Phys 2015;17:18980–90.
equilibrium conditions. Fluid Phase Equilib 2017;440:1–8. https://doi.org/10. https://doi.org/10.1039/c5cp03094c.
1016/j.fluid.2017.02.011.

12