J. Chem.

Thermodynamics 49 (2012) 7–13

Contents lists available at SciVerse ScienceDirect

J. Chem. Thermodynamics
journal homepage: www.elsevier.com/locate/jct

Experimental study and thermodynamic modelling of methane clathrate

hydrate dissociation conditions in silica gel porous media in the presence
of methanol aqueous solution q
Hamed Hashemi a, Jafar Javanmardi a,⇑, Mojdeh Zarifi a, Ali Eslamimanesh b, Amir H. Mohammadi b,c
a
Department of Chemical Engineering, Shiraz University of Technology, Shiraz 71555-313, Iran
b
MINES ParisTech, CEP/TEP – Centre Énergétique et Procédés, 35 Rue Saint Honoré, 77305 Fontainebleau, France
c
Thermodynamics Research Unit, School of Chemical Engineering, University of KwaZulu-Natal, Howard College Campus, King George V Avenue, Durban 4041, South Africa

a r t i c l e i n f o a b s t r a c t

Article history: In this work, the phase equilibria of clathrate hydrates of methane in the presence of pure water and
Available online 13 February 2012 0.035 mass fraction of methanol aqueous solution in confined silica gel pores with (10 and 15) nm mean
diameters are measured and reported. A thermodynamic model is also developed for prediction of the
Keywords: obtained experimental hydrate dissociation data. The Valderrama–Patel–Teja (VPT-EoS) equation of state
Gas hydrate (EoS) accompanied with the non-density dependent (NDD) mixing rules coupled with a previously devel-
Porous media oped activity model are applied to evaluate the fugacity of the species present and the activity coefficient
Experimental data
of water in methanol aqueous solution. Acceptable agreement between the reported data and the pre-
Thermodynamic model
Methanol
dicted results using the proposed model and an existing method reported in the literature demonstrates
Phase equilibria the reliability of the presented model.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction cesses, which contributes to ozone depletion. As a consequence

of these concepts, carbon dioxide capture and sequestration have
Clathrate hydrates (gas hydrates) are ice-like crystalline struc- been of great interest during the past decade [4–8]. Methane hy-
ture consisting of a framework of water molecules bounded to each drates can be replaced by those of carbon dioxide in deep ocean
other by hydrogen bonds that can trap molecules of appropriate sediments. This technique results in two advantages: (1) reducing
sizes like methane, some volatile liquids, etc. Gas hydrates gener- the carbon dioxide emissions; (2) assisting natural gas recovery
ally form at low temperatures and high pressures. These structures [1,2].
occur abundantly in nature in two geologic conditions within per- It should be noted that methane is a more effective green-house
mafrost and arctic regions as well as in deep sea sediments at mar- gas than carbon dioxide [4–8]. In the case of drastic global warm-
gins of every continent [1,2]. ing, the temperature rises and the earth may face with the decom-
Methane is a main constituent of natural gas and normally sta- position of methane hydrate reserves. On the other hand, methane
ble in sea floor sediments. It can also create a gas trap beneath the hydrate deposition may occur in the sediments. Furthermore, there
sea floor. Methane gas in the sediments may continually migrates is evidence of the positive effects of special kinds of porous media
to upper levels to fill the pores of the rocks. This process leads to for enhancement of gas hydrate formation rates and their subse-
production of highly concentrated methane hydrate reservoir [3]. quent effects on gas separation processes through gas hydrate
Since there are enormous quantities of gas hydrate resources be- crystallization [4].
neath the ocean sediments, within permafrost, and arctic regions, Therefore, a number of studies have focused on the investiga-
they are being looked upon as a future source of energy. tion of phase equilibria of methane hydrate formation in various
Methane is environmentally a clean fuel compared to oil and it kinds of porous media similar to the sea sediments. Perhaps, the
produces less carbon dioxide per unit of energy. Therefore, it is first experimental measurement on this subject was performed
generally suggested as a better future fuel supply. On the other by Handa and Stupin [9]. Other researchers including Uchida
hand, carbon dioxide is a major component in the gas streams et al. [10,11], Wilder et al. [12,13], Seshadri et al. [14], Smith et al.
emitted from the thermal plants and fossil fuel combustion pro- [15], Zhang et al. [16] and Kang et al. [17] continued the corre-
sponding efforts experimentally.
q Clennell et al. [18] and Henry and co-workers [19] proposed a
Paper for gas hydrate initiative.
⇑ Corresponding author. Tel.: +98 711 7279416; fax: +98 711 7354520. new interpretation of experimental data based on considering
E-mail address: [email protected] (J. Javanmardi). the capillary pressure effects on hydrate formation conditions.

0021-9614/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jct.2012.02.003
8 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13

Clarke et al. [20] showed that the effect of geometrical constraints 2.3. Procedure
is to lower the water activity so that the phase equilibrium bound-
ary shifts to the lower temperature and higher pressure conditions. In the beginning of each experimental run, silica gel was dried
Klauda and Sandler [21] proposed a model considering the pore at T = 423 K for 2 days before water sorption. Later, the silica gel
size distribution on the conditions of the gas hydrate formation sample was loaded with water as follows: Having measured the
in specific porous media. The Gibson–Thomson expression was dry weight of the silica gel, it was placed under vacuum in a desic-
successfully applied by Lameudo et al. [22] to develop a thermody- cator for two days. Definite amount of liquid water was added to
namic model for prediction of the dissociation conditions of clath- the evacuated desiccator, with indirect contact with silica gel. After
rate hydrates of methane/carbon dioxide/their relevant mixtures in (5 to 6) days, all pores became saturated (vapour + solid equilibria).
the presence/absence of methanol (MeOH) aqueous solution in Knowing the specific pore volume of the silica gel, the pore satura-
porous media. Li et al. [23], Aladko et al. [24] and Sun and Duan tion was estimated by weighing the sample. Pursuing this proce-
[25] are among the others, who theoretically investigated the gas dure, we can reach up to 99% saturation (volume basis) of silica
hydrate formation in different porous media. pores. Later, a weighted amount of wetted sample equivalent to
It should be noted that accurate knowledge of the simultaneous 30 cm3 was fed to the equilibrium cell. In order to saturate the sil-
effects of porous media and alcohol/electrolyte provides key infor- ica gel sample with methanol solution, direct contact of silica gel
mation for exploiting natural gas hydrate reserves from marine and methanol solution was used because in indirect contact proce-
sediments. Regarding the significance of the corresponding exper- dure using a desiccator, the amount of absorbed methanol to the
imental measurements and development of more accurate ther- silica gel from vapour phase may be unknown. Definite amount
modynamic models for prediction of the phase equilibria of of methanol solution equals to the pore volume of silica gel was
clathrate hydrates in porous media, the dissociation conditions of added to the silica gel sample, placed in a balloon. The balloon
methane hydrate in mesoporous silica gel containing methanol was sealed off to prevent methanol evaporation and left for (5 to
are measured and reported in this communication. Moreover, the 6) days to ensure the complete saturation of silica gel.
capability of a thermodynamic model for prediction of the mea- The hydrate dissociation conditions of the investigated systems
sured data and some selected ones reported in the literature is were measured employing the isochoric pressure search method
examined. [26–30]. After the cell was pressurized to the desired pressure with
methane as hydrate former, it was cooled to T = 263 K at a cooling
rate of 1.5 K  h1. The hydrate formation was confirmed through a
2. Experimental sudden pressure drop, resulted from gas trapping in the water
cages (hydrate cavities). When this pressure depression reached
2.1. Materials a steady-state condition and the system conditions became stabi-
lized, the cell temperature was stepwise increased by 0.3 K and
Methane with minimum purity of 0.9995 mole fraction was held for (3 to 8) h or sufficient time in each step to allow the sys-
supplied by Air Product. Double distilled water and methanol with tem to reach equilibrium conditions and stable pressure at each
purity of 0.999 mole fraction was used in all the experiments. The step. The temperature of the cell was raised until all of the meth-
spherical silica gels with nominal pore diameter of (10 and 15) nm ane hydrates crystals were decomposed. The details of experimen-
were purchased from Merck and Aldrich. The detailed properties of tal procedure of the isochoric pressure search method can be found
silica gels can be found in table 1. elsewhere [26–30].

3. Thermodynamic model
2.2. Apparatus
The previously developed model of Javanmardi et al. [31] has
The main part of the experimental setup, as sketched in figure 1, been applied for prediction of methane hydrate phase equilibria
is a high pressure equilibrium cell (sample container). This reactor in the presence of methanol aqueous solution. The model is based
is made of 316 stainless steel equipped with double side glasses. on the equality of the chemical potential of water in hydrate phase
The cell has the internal volume of 75 cm3. The reactor tempera- with that in water-rich liquid phase or ice, assuming negligible
ture is controlled using a thermostatted bath, Julabo FP 50, with amounts of water in gas phase [32]:
accuracy of ±0.1 K. A Pt100 platinum temperature probe inserted
in the cell interior is used to measure the temperature of the sys- lLw ðT; PÞ ¼ lHw ðT; PÞ; ð1Þ
tem. The uncertainty of the temperature measurements are
±0.1 K after calibration against a 25 X reference platinum resis- where l is the chemical potential, T is temperature, P is pressure,
tance thermometer. The pressure of the cell is measured by a pres- subscript w denotes the water component, and superscripts L and
sure transducer within a precision of ±0.25% of full scale, which has H stand for the liquid and hydrate phases, respectively. If the chem-
been calibrated by the dead weight method. The data acquisition ical potential of an empty hydrate lattice is taken as reference,
system is used for recording the system pressure and temperature equation (1) becomes [1,31,32]:
throughout the experiments.
DlLw ¼ DlHw ; ð2Þ

TABLE 1 where,
The properties of the silica gels investigated.

Sample 10 nma 15 nmb DlLw ¼ lMT  lLw ; ð3Þ

Mean pore diameter/nm 10.5 14.6
Max pore diameter/nm 17.2 22.6 DlHw ¼ lMT  lHw ; ð4Þ
Specific pore volume/cm3  g1 0.87 1.13
Specific surface area/m2  g1 350.9 308
where superscript MT refers to the empty hydrate lattice. The chem-
a
Mean and max pore diameters have been calculated using the proposed model. ical potential of water in hydrate phase can be evaluated using the
b
Values are taken from [42]. van der Waals–Platteeuw [33] equation:
 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13 9

FIGURE 1. A schematic diagram of the experimental apparatus. P-i: pressure transmitter-i, T-i: temperature transmitter-i, V-i: valve-i, R-i: pressure regulator-i, TCS:
Temperature Control System.

X X
lHw ¼ lMT
w  RT mm lnð1 þ C mj fj Þ; ð5Þ equals to 4, 5, 10 or 11. Also, r⁄ = r  2a, where r is the collision
m j diameter [34,36]. The Kihara potential parameters are given in
table 2.
where R is the universal gas constant, v m stands for the number of
Holder et al. [32] developed the below equation for determining
cavities of type m per water molecule in the unit hydrate cell, fj de-
the chemical potential difference of water in empty hydrate lattice
notes the fugacity of the gas component j, and Cmj is the Langmuir
and liquid water or ice:
constant. Numerical values for the Langmuir constant can be calcu-
lated by choosing a model for the guest–host interaction [1,32,34]: Z MTI=Lw
Z   DlMTI=L
w
w
lMT lI=L
w
w
Dl0w T
Dh w
4p 1
wðrÞ 2 ¼ w  ¼  dT
C jm ðTÞ ¼ exp  r dr; ð6Þ RT RT 0 RT 0 RT 0 T0 RT 2
kT 0 kT
Z
where k is the Boltzmann’s constant. The function w(r) is the spher-
P
DmMTI=L
w
w

þ dP  ln aw ; ð9Þ
ically symmetric cell potential in the cavity, with r measured from 0 RT
centre, and depends on the intermolecular potential function cho-
sen for describing the encaged gas–water interaction. In this work, where superscript 0 refers to reference property and superscript
the Kihara [35] potential function is applied to evaluate the Lang- MT  I/Lw stands for the difference property between empty hydrate
muir constant as follows [1,34,36]: lattice and water in liquid state (Lw) or in the ice (I). The Dl0w is the
"   # reference chemical potential difference between water in empty hy-
ðr Þ12 10 a 11 ðr Þ6  4 a 5  drate lattice and pure water in the ice phase at standard condition
wðrÞ ¼ 2ze d þ d  d þ d ; ð7Þ
R11 r R R5 r R (here it is T = 273.15 K) and aw stands for the activity of water. The
Dm is the molar volume difference, and Dh stands for the enthalpy
where, difference, which is calculated by the following expression
" N  N # [1,32,34]:
1 r a r a
dN ¼ 1   1þ  : ð8Þ Z
N R R R R MTI=L 0
T
Dhw ¼ Dh w þ DC Pw dT; ð10Þ
In the two preceding equations, z is the coordination number of the T0

cavity (the number of oxygen molecules at the periphery of each

cavity), e would be characteristic energy, a is the radius of spherical where Cp stands for the molar heat capacity, and Dh0 is the enthalpy
molecular core, R stands for the cavity radius, and N is an integer difference between the empty hydrate lattice and ice, at the ice
point and zero pressure. Additionally, the difference between the
heat capacity of the empty hydrate lattice and the pure liquid water
TABLE 2 is evaluated by the following equation [1,32,34]:
The Kihara potential parameters [1].

Component aa /nm r ⁄ b/nm (e/k)c/K DC Pw ¼ 37:32 þ 0:179ðT  T 0 Þ T > T 0 : ð11Þ

Methane 0.03834 0.314393 155.593
a
The radius of spherical molecular core.
The heat capacity difference is assumed to be zero when T 6 T0. The
b
r⁄ = r  2a, where r is the collision diameter. values of the reference properties are listed in table 3. Conse-
c
e is the characteristic energy and k is the Boltzmann’s constant. quently, the following summarized equation can be written [32]:
10 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13

TABLE 3 4. Results and discussion

Thermodynamic reference properties for hydrate
structure-I in the liquid water region [1].
4.1. Calculation of pore size distribution
Structure I

Dl0w a/(J  mol1) 1263 As indicated by Wilder et al. [12] and Smith and coworkers
Dh0w b/(J  mol1) 4858 [15,41], the heating curve data can be used for calculation of pore
Dm0w c/(cm3  mol1) 4.6 size distribution (PSD) as well as nitrogen desorption isotherms. In
a this work, we have pursued the same procedure to calculate the
The reference chemical potential difference
between water in empty hydrate lattice and pure PSD of the silica gel investigated [12,15,41]. The PSD curve for silica
water in the ice phase at standard condition (here it gel with nominal pore diameter of 10 nm have been measured
is T = 273.15 K). from dissociation data of methane hydrate in the presence of pure
b
Enthalpy difference. water, which is shown in figure 2. Figure 3 indicates that the mean
c
Molar volume difference.
pore diameter of the studied silica gel is about 10.5 nm, which is in
acceptable agreement with the reported nominal pore diameter by
Z MTI=Lw Z P
Dl0w T
Dh w DmMTI=L
w
w
Merck. Max pore diameter has been selected as the 99.9% of max-
 dT þ dP  ln aw
RT 0 T0 RT 2
0 RT imum cumulative pore volume. The properties of silica gel with
X X nominal pore diameter of 15 nm have been taken from the work
 mm lnð1 þ C mj fj Þ ¼ 0: ð12Þ
m
of Lee and Seo [42].
j
Gas hydrate dissociation conditions have been determined from
Javanmardi et al. [31] showed that the following expression can be the heating curve data points [3,9–15]. The heating curve in porous
applied to calculate the activity of water in liquid phase: media has normally the ‘‘S’’ shape trend, which is in agreement
with previous conclusions reported in the literature
xw uelal
w
aw ¼ 0
: ð13Þ [10,11,42,43]. As interpreted by Anderson et al. [43], the hydrate
uEoS
w
0
In this equation, uEoS
w is the fugacity coefficient of pure water at ref- 5.7
erence pressure and temperature and xw is the mole fraction of the
5.6
water. The fugacity coefficient of water in the presence of alcohol
aqueous solution and electrolytes, uelal
w can be written as follows 5.5
[31]:
P/Mpa

5.4
ln uelal
w ¼ ln uEoS el al
w þ ln cw þ ln cw ; ð14Þ
5.3
where uEoS al
w is the fugacity coefficient of water and cw is the activity
coefficient of water in methanol aqueous solution. In this work, uEoS 5.2
w
has been evaluated applying the Valderrama–Patel–Teja equation of 5.1
state (VPT-EoS) [37] and non-density dependent (NDD) mixing rule
[38], and cal
w has been calculated using two-suffix Margules activity
5
model [39]. Constants of the Margules activity model are repre-
4.9
sented in table 4. The detailed description of the applied EoS is gi- 274 275 276 277 278 279 280
ven elsewhere [34,37,38,40]. Binary interaction parameters of the
T/K
EoS are given in table 5.
The decrease in water activity in silica gel pores with pore ra- FIGURE 2. Methane hydrate dissociation data in 10 nm silica gel in the presence of
dius r due to the capillary effect can be written as: pure water obtained in this work for calculation of the PSD.

mL F rHW coshHW
lnðaw Þpore ¼ lnðaw Þ  ; ð15Þ
rTR
0.5 4
where vL is the molar volume of pure water, F denotes the shape
Cumulative pore volume/cm3 .g-1 nm-1

factor, hHW stands for the wetting angle, and rHW is the interfacial 3.5
energy between the hydrate and liquid water phases. Equations 0.4
(12) and (15) are solved simultaneously to predict the investigated 3
Pore volume/cm3

gas hydrates dissociation conditions.

0.3 2.5

TABLE 4 2
Two-suffix margules equation parameters for water–methanol system.
0.2 1.5
a1 a2/R1 b1 b2/R1
5.84663 15.04912 2.82261 10.91332
1
0.1
0.5
TABLE 5
Binary interaction parameters between polar–nonpolar and polar–polara components 0 0
using in the VPT-EoS [37] with NDD mixing rule [38]. 0 5 10 15 20 25
Component H2O (i) MeOH (i) Pore diameter/nm
kij = kji 0 1 kij = kji 0 1
lij lij lij lij
FIGURE 3. Pore volume and cumulative pore volume distributions reconstructed
CH4(j) 0.5028 1.8180 0.0049 0.2538 0.7319 0.000688 from experimental data in figure 2. Dashed line, cumulative pore volume; solid line,
pore size distribution; d, calculated pore volume; double arrow, dmean = 10.5 nm;
a
For (H2O + MeOH) system, kij = kji = 0.1 + 0.000185  {(T/K)273}. double dashed arrow, dmax = 17.2 nm.
 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13 11

6 4.5
14
3 3.5
5.5 4

2.5 10

P/MPa
dP/dT
P/MPa

5
1.5
2
6
4.5
0.5
1
2
4 -0.5
270 274 278 282 286 290
270 271 272 273 274 275 276 277 278
T/K
T/K
FIGURE 6. Methane hydrate dissociation conditions in 15 nm silica gel in the
FIGURE 4. Methane hydrate heating curve in 15 nm silica gel in the presence of
presence of 0.035 mass fraction of methanol aqueous solution. Experimental data:
0.035 mass fraction of methanol aqueous solution. (1) Mean pore equilibrium point;
This work d, 15 (22.6) nm 0.035 mass fraction MeOH; s, bulk 0.035 mass fraction
(2) max pore equilibrium point; (3) mean pore; (4) max pore. Solid line with d, dP/
MeOH; literature: N, 15 (21.5) nm 0.035 mass fraction MeOH [44]; h, bulk 0.035
dT; solid line with s, heating curve; solid lines, model predictions.
mass fraction MeOH [44]; solid line, model prediction; dashed line, predictions of
the correlation proposed in [45].

TABLE 6
The measured dissociation data for the clathrate hydrates of methane in (10, 15) nm
silica gel and bulk phase in the presence of 0.035 mass fraction of methanol aqueous
solution. 14

15 nm 10 nm Bulk phase
T/K P/MPa T/K P/MPa T/K P/MPa
10
277.5 5.91 278.1 6.47 275.0 3.37
P/MPa
276.3 5.25 275.6 5.12 277.1 4.10
274.1 4.13 272.9 3.98 277.6 4.33
272.2 3.14 280.3 5.25 6
281.1 6.12
281.6 6.53

2
270 275 280 285 290
T/K

14 FIGURE 7. Prediction of methane hydrate dissociation conditions in (9.2 and

30) nm silica gel in the presence of 0.035 mass fraction of methanol aqueous
solution. Experimental data: j, 30 (52.5) nm, 0.035 mass fraction MeOH [44]; N, 9.2
(13.5) nm, 0.035 mass fraction MeOH [44]; solid line, model predictions; dashed
line, predictions of the correlation proposed in [45].
10
P/ MPa

6 14

2 10
270 275 280 285 290
P /MPa

T/K

FIGURE 5. Methane hydrate dissociation conditions in 10 nm silica gel in the 6

presence of 0.035 mass fraction of methanol aqueous solution. Experimental data:
This work j, 10 (17.2) nm 0.035 mass fraction MeOH; s, bulk 0.035 mass fraction
MeOH; literature: N, 9.2 (13.5) nm 0.035 mass fraction MeOH [44]; h, bulk 0.035
mass fraction MeOH [44]; solid line, model prediction; dashed line, predictions of 2
the correlation proposed in [45]. 270 275 280 285 290
T /K
dissociation point in the presence of pure water was chosen as FIGURE 8. Prediction of methane hydrate dissociation conditions in (15 and 30) nm
equilibrium point that corresponds to the inflection point of the silica gel in the presence of methanol aqueous solution. Experimental data: j, 15
heating curve. Regarding the hydrates formed in the presence of (21.5) nm, 0.036 mass fraction MeOH [44]; N, 30 (52.5) nm, 0.039 mass fraction
alcohol aqueous solution, Østergaard et al. [44] concluded that MeOH [44]; solid line, model predictions; dashed line, predictions of the correlation
proposed in [45].
the better agreement with experiment can be corresponded to
the maximum dissociation points of the heating curve because
during the hydrate growth, methanol concentration is increased as shown in figure 4 and table 6. Note that the obtained hydrate
in the aqueous solution. In this study, maximum hydrate dissocia- dissociation points through this method represent the conditions
tion points from heating curves have been chosen for this purpose of maximum pore size of porous media, dmax.
12 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13

TABLE 7 0
The absolute average relative deviations of the
predicted results from the obtained hydrate dissoci-
ation data for methane + 0.035 mass fraction of -0.002
methanol aqueous solution.

Pore diameter/nm AARD%a -0.004

ΔTp /Tb
10 (17.2) 0.1
15 (22.6) 0.05
-0.006
a PN jT Pred:ðiÞ T Exp:ðiÞ j
AARD% ¼ 100
N i ; where superscripts
T Exp:ðiÞ
Pred. and Exp. denote the predicted and experi-
mental values, respectively, and N is the number of
-0.008
experimental data points.
- 0.01
0 0.02 0.04 0.06
1/d

FIGURE 10. Plot of reciprocal pore diameter (1/d) versus DTp/Tb for the measured
TABLE 8 methane hydrate dissociation conditions in the presence of 0.035 mass fraction of
The absolute average relative deviations of the predicted results from the hydrate methanol aqueous solution. DTp, depression of the dissociation temperature of
dissociation data for the (methane + methanol aqueous solution) systems from clathrate hydrate; Tb, the dissociation temperature of the clathrate hydrate in the
Østergaard et al. [44]. bulk phase; s, experimental data (this work); solid line, fitting curve.

Pore diameter/nm AARD% Methanol concentration/mass

fraction in aqueous solution As can be observed in figure 9, there are two distinct regions in
9.2 (13.5) 0.1 0.035 dissociation patterns of the investigated gas hydrates. Intersection
15 (21.5) 0.1 0.035 between heating curve and unconfined (bulk) hydrate phase equi-
15 (21.5) 0.1 0.036
librium curve at the point of second change of slop of heating curve
30 (52.5) 0.1 0.035
30 (52.5) 0.05 0.039 demonstrates that this point is related to the dissociation of gas
hydrates crystals outside the pores (unconfined clathrates).

4.2. Calculation of interfacial energy

6.5 * **
1 The interfacial energy between methane clathrate hydrate and
0.035 mass fraction of methanol aqueous solution is generally
2
6 determined using the Gibbs–Thomson equation. According to this
method, depression of the dissociation temperature of clathrate
P/MPa

5.5 hydrate, DTp, can be calculated as a function of pore diameter d

as follows [18,20,22]:
3
5 DT p F rHW
4 ¼ : ð16Þ
Tb qh DHh;d d
4.5
265.15 269.15 273.15 277.15 281.15 285.15 In the preceding equation, Tb is the dissociation temperature of the
T/K clathrate hydrate in the bulk phase, rHW denotes the interfacial en-
ergy between the hydrate and aqueous solution phases, qh is the
FIGURE 9. Primary heating and cooling curve for methane hydrate in 15 nm silica
density of hydrate with the value of 914 kg  m3 taken from [1],
gel in the presence of 0.035 mass fraction of methanol aqueous solution. Solid line,
cooling-heating curve; dashed line, model predictions: ⁄, represents the prediction DHh,d stands for the enthalpy of dissociation of hydrate with the va-
for max pore in the presence of 0.035 mass fraction MeOH; ⁄⁄, represents the lue of 57.7 kJ  mol1 taken from [46], and F is the shape factor that
prediction for bulk phase in the presence of 0.035 mass fraction MeOH; (1) bulk for cylindrical pores with the values of 2. The values of
dissociation; (2) bulk formation; (3) pore formation; (4) pore dissociation. (34 ± 4) mJ  m2 for hydrate-methanol aqueous solution interfacial
tension was obtained using the value of the slop of a linear fit of
DTp/Tb to reciprocal pore diameter (1/d), as sketched in figure 10,
Table 6 also reports the obtained bulk hydrate dissociation data
for methane clathrate hydrate in silica gel pores containing 0.035
in the presence of 0.035 mass fraction of methanol aqueous solu-
mass fraction of methanol aqueous solution. In comparison with
tion. Figures 5 and 6 show the predicted gas hydrate phase equilib-
the values obtained by Anderson et al. [43] (35 ± 1) mJ  m2 for
ria compared with the corresponding measured hydrate
pure water, the change in interfacial tension between methane
dissociation data. The experimental dissociation values reported
clathrate hydrate and methanol aqueous solution is slight.
by Østergaard et al. [44] are also included in these figures. Accept-
able agreement has been found between the predicted results
using the proposed model and the measured/existing experimental 5. Conclusions
gas hydrate dissociation data. Figures 7 and 8 depict the experi-
mental data of Østergaard and coworkers [44] for silica gel with In this communication, experimental methane hydrate dissoci-
nominal pore diameters of (9.2 and 30.6) nm along with the ap- ation conditions in the presence of 0.035 mass fraction of methanol
plied model predictions. Furthermore, the predictions of the re- aqueous solutions and silica gel samples with (10 and 15) nm pore
cently proposed correlation by Mohammadi and co-workers [45] sizes were measured. The dissociation conditions were measured
have been included in these figures for comparison purpose. The pursuing the isochoric pressure search method [26–30].
average absolute relative deviations of model predictions from A thermodynamic model was also proposed for prediction of
the investigated experimental hydrate dissociation values are re- hydrate dissociation conditions in the presence of methanol in por-
ported in tables 7 and 8. ous media. The VPT-EoS [37] and the NDD mixing rules [38] were
 H. Hashemi et al. / J. Chem. Thermodynamics 49 (2012) 7–13 13

applied to evaluate the fugacity coefficients of the species present [15] D.H. Smith, J.W. Wilder, K. Seshadri, AIChE J. 48 (2002) 393–400.
[16] W. Zhang, J.W. Wilder, D.H. Smith, AIChE J. 48 (2002) 2324–2331.
and a previously developed activity model [31] was used to calcu-
[17] S.P. Kang, J.W. Lee, H.J. Ryu, Fluid. Phase Equilibr. 274 (2008) 67–71.
late the activity of water in methanol aqueous solution. To model [18] M.B. Clennell, M. Hovland, J.S. Booth, P. Henry, W.J. Winters, J. Geophys. Res.
the hydrate phase, the solid solution theory of vdW-P [33] was em- 104 (1999) 22985–23003.
ployed. It can be concluded that the deviations between the pre- [19] P. Henry, M. Thomas, M.B. Clennell, J. Geophys. Res. 104 (1999) 23005–23022.
[20] M.A. Clarke, M. Pooladi-Darvish, P.R. Bishnoi, Ind. Eng. Chem. Res. 38 (1999)
dicted phase equilibria and experimental dissociation values are 2485–2490.
reasonable. The predicted results were also successfully compared [21] J.B. Klauda, S.I. Sandler, Ind. Eng. Chem. Res. 40 (2001) 4197–4208.
to a recently proposed correlation [45]. [22] M. Lameudo, R. Anderson, B. Tohidi, R.W. Burgass, J. Phys. Chem. B 107 (2003)
3507–3514.
Applying the obtained experimental data, interfacial energy be- [23] X.S. Li, Y. Zhang, G. Li, Z.Y. Chen, K.F. Yan, Q.P. Li, J. Chem. Thermodyn. 40
tween methane clathrate hydrate and the methanol aqueous solu- (2008) 1464–1474.
tions investigated was determined. It was found that the interfacial [24] E.Y. Aladko, Y.A. Dyadin, V.B. Fenelonov, E.G. Larionov, A.Yu. Manakov, F.V.
Zhurko, M.S. Melgunov, J. Phys. Chem. B 110 (2006) 19717–19725.
energy between methane hydrate and the aqueous solution was [25] R. Sun, Z. Duan, J. Chem. Geol. 244 (2007) 248–262.
changed very slightly due to the presence of methanol in water [26] R. Ohmura, S. Takeya, T. Uchida, T. Ebinuma, Ind. Eng. Chem. Res. 43 (2004)
and, consequently, can be neglected. Furthermore, it was con- 4964–4966.
[27] A.H. Mohammadi, W. Afzal, D. Richon, J. Chem. Eng. Data 53 (2008) 683–686.
cluded that the presence of the investigated porous media along [28] B. Tohidi, R.W. Burgass, A. Danesh, K.K. Østergaard, A.C. Todd, Ann. N.Y. Acad.
with methanol aqueous solution results in shifting the (liquid Sci. 912 (2000) 924–931.
water + hydrate + vapour phase) equilibrium boundary curves to [29] V. Belandria, A. Eslamimanesh, A.H. Mohammadi, P. Théveneau, H. Legendre,
D. Richon, Ind. Eng. Chem. Res. 50 (2011) 5783–5794.
the left i.e. low temperature and high pressure conditions.
[30] V. Belandria, A. Eslamimanesh, A.H. Mohammadi, D. Richon, Ind. Eng. Chem.
Res. 50 (2011) 4722–4730.
References [31] J. Javanmardi, M. Moshfeghian, R.N. Maddox, Can. J. Chem. Eng. 79 (2001) 367–
373.
[1] E.D. Sloan, C.A. Koh, Clathrate Hydrates of Natural Gases, third ed., CRC Press, [32] G.D. Holder, G. Corbin, K.D. Papadopoulos, Ind. Eng. Chem. Fund. 19 (1980)
New York, 2007. 282–286.
[2] B.C. Gbaruko, J.C. Igwe, P.N. Gbaruko, R.C. Nwokeoma, J. Petrol. Sci. Eng. 56 [33] J.H. van der Waals, J.C. Platteeuw, Adv. Chem. Phys. 2 (1959) 1–57.
(2007) 192–198. [34] A.H. Mohammadi, R. Anderson, B. Tohidi, AIChE J. 51 (2005) 2825–2833.
[3] Y.T. Seo, I.L. Moudrakovski, J.A. Ripmeester, J.W. Lee, H. Lee, Environ. Sci. [35] T. Kihara, J. Phys. Soc. Jpn. 8 (1953) 686–687.
Technol. 39 (2005) 2315–2319. [36] V. McKoy, O. Sinanoglu, J. Chem. Phys. 38 (1963) 2946–2956.
[4] J.Z. Zhao, Y.S. Zhao, D.X. Shi, Meitan Xuebao/J. China Coal Soc. 33 (2008) 1419– [37] J.O. Valderrama, J. Chem. Eng. Jpn. 23 (1990) 87–91.
1424. [38] D. Avlonitis, A. Danesh, A.C. Todd, Fluid Phase Equilibr. 94 (1994) 181–216.
[5] P. Englezos, J.D. Lee, Korean J. Chem. Eng. 22 (2005) 671–681. [39] M. Margules, Sitz. Akad. Wiss. Wien Math. Nat. Kl. IIa 104 (1895) 1243.
[6] M.C. Boatman, J. Peterson, Oceanic Gas Hydrate Research and Activities [40] A. Eslamimanesh, A.H. Mohammadi, D. Richon, AIChE J. 57 (2011) 2566–2573.
Review, U.S., Department of the Interior Minerals Management Service Gulf of [41] D.H. Smith, J.W. Wilder, K. Seshadri, Environ. Sci. Technol. 36 (2002) 5192–
Mexico OCS, Region, February 2000. 5198.
[7] International Panel on Climate Control (IPCC), 2005. Carbon dioxide capture [42] S. Lee, Y. Seo, Langmuir 26 (2010) 9742–9748.
and storage. Special report. International panel on climate (IPCC), February 5, [43] R. Anderson, M. Llamedo, B. Tohidi, R.W. Burgass, J. Phys. Chem. B 107 (2003)
2007. Climate Change 2007: The Physical Science Basis – Summary for Policy 3500–3506.
Makers. http://www.ipcc.ch/SPM2feb07.pdf. [44] K. Østergaard, R. Anderson, M. Llamedo, B. Tohidi, Terra Nova 14 (2002) 307–
[8] A. Eslamimanesh, A.H. Mohammadi, D. Richon, P. Naidoo, D. Ramjugernath, J. 312.
Chem. Thermodyn. 46 (2012) 62–71. [45] A.H. Mohammadi, A. Eslamimanesh, D. Richon, F. Gharagheizi, M. Yazdizadeh,
[9] Y.P. Handa, D.J. Stupin, Phys. Chem. 96 (1992) 8599–8603. J. Javanmardi, H. Hashemi, M. Zarifi, S. Babaee, Ind. Eng. Chem. Res. 51 (2012)
[10] T. Uchida, T. Ebinuma, S. Takeya, J. Nagao, H. Narita, J. Phys. Chem. B 103 1062–1072.
(1999) 3659–3662. [46] J.S. Lievois, R. Perkins, R.J. Martin, R. Kobayashi, Fluid Phase Equilibr. 59 (1990)
[11] T. Uchida, T. Ebinuma, S. Takeya, J. Nagao, H. Narita, J. Phys. Chem. B 106 73–97.
(2002) 820–826.
[12] J.W. Wilder, K. Seshadri, D.H. Smith, Langmuir 17 (2001) 6729–6735.
[13] J.W. Wilder, K. Seshadri, D.H. Smith, J. Phys. Chem. B 105 (2001) 9970–9972.
[14] K. Seshadri, J.W. Wilder, D.H. Smith, J. Phys. Chem. B 105 (2001) 2627–2631. JCT 11/580