Chemical Engineering Science 117 (2014) 342–351

Contents lists available at ScienceDirect

Chemical Engineering Science

journal homepage: www.elsevier.com/locate/ces

Unusual behavior of propane as a co-guest during hydrate formation

in silica sand: Potential application to seawater desalination
and carbon dioxide capture
Ponnivalavan Babu a, Rajnish Kumar b, Praveen Linga a,n
a
Department of Chemical and Biomolecular Engineering, National University of Singapore, Singapore 117576, Singapore
b
Chemical Engineering and Process Development Division, National Chemical Laboratory, Pune, India

H I G H L I G H T S

Unusual behavior of hydrate formation in presence of propane as co-guest is reported.

Mechanism of enhanced kinetics of hydrate formation in sand with propane as co-guest is presented.

Conceptual process flow sheet applying the clahtrate process for desalination is proposed.

An innovative approach is proposed for simultaneous CO2 capture and desalination.

art ic l e i nf o a b s t r a c t

Article history: We report an unusual behavior of hydrate formation in silica sand with gas mixtures containing propane
Received 16 May 2014 as a co-guest. Based on morphology study we observed that propane as a co-guest has the ability to draw
Received in revised form water dispersed in silica sand to the hydrate formation region and showed a tendency to result in drastic
20 June 2014
hydrate growth due to the migration of water molecules to the gas phase region. Hydrate nucleation
Accepted 29 June 2014
occurred in the interstitial pore space between the silica sand particles and hydrate growth occurred in
Available online 11 July 2014
the gas phase above the silica sand bed and to sustain the hydrate growth, dispersed water was drawn
Keywords: towards the hydrate growth front. In addition, we elucidated the effect of sand bed height to maximize
Gas hydrates the growth rates utilizing this behavior that results in enhanced kinetics. We propose conceptual designs
Propane hydrate: pre-combustion capture
for utilizing this behavior of propane as a co-guest in sand for seawater desalination and an innovative
Carbon dioxide capture
approach to simultaneously capture carbon dioxide and desalinate seawater.
Enhanced kinetics
Desalination & 2014 Elsevier Ltd. All rights reserved.

1. Introduction
Gas hydrates are crystalline compounds in which water molecules
through hydrogen bonding form cavities that enclose small guest
molecules like methane, carbon dioxide, hydrogen (Englezos, 1993;
Sloan and Koh, 2008). Due to the high gas storage capacity, non-
explosive nature and selective fractionation, gas hydrates have a wide
range of applications including gas storage, gas separation, desalina-
tion, natural gas storage and transport and CO2 capture and seques-
tration (Chatti et al., 2005; Eslamimanesh et al., 2012; Lee et al.,
2005; Linga et al., 2007; Park et al., 2011; Rufford et al., 2012;
Veluswamy et al., 2014b; Xu and Li, 2014). Among the applications,
CO2 capture has received significant attention as water is used a
solvent to capture CO2 from flue and fuel gas streams employing the
n
Corresponding author. Tel.: þ 65 6601 1487; fax: þ 65 6779 1936.
E-mail address: praveen.linga@nus.edu.sg (P. Linga).
clathrate process (Babu et al., 2014; Duc et al., 2007; Gholinezhad
et al., 2011; Kang and Lee, 2000; Kumar et al., 2013; Li et al., 2010,
2011a, 2011b; Linga et al., 2010; Park et al., 2013; Tajima et al., 2004;
Xu et al., 2013; Yang et al., 2012).
Linga et al. (2012) employed several gas/gas mixtures and
presented kinetic data in a silica sand bed dispersed with water
and reported enhanced kinetics in a sand bed compared to a
stirred tank reactor. Significantly higher formation rates for the gas
mixtures containing propane as a co-guest was also observed.
Recently, Babu et al. (2013a) employed a fixed bed column with
silica sand bed for pre-combustion CO2 capture and reported 38%
water conversion to hydrate in presence of 2.5 mol% propane in
fuel gas mixture. They concluded that propane not only reduces
the operating pressure from 9.0 MPa to 6.0 MPa but also signifi-
cantly enhances the kinetics of hydrate formation. Fig. S1
(Supporting Information) shows literature data for enhanced
water conversion to hydrate in presence of propane as a co-
guest in silica sand bed (Babu et al., 2013a, 2013b). The presence

http://dx.doi.org/10.1016/j.ces.2014.06.044
0009-2509/& 2014 Elsevier Ltd. All rights reserved.
 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
of 2.5 mol% propane allows the fuel gas mixture to form structure
sII instead of sI, however, the significant increase in gas uptake and
water conversion to hydrate cannot be attributed to the structure
change alone as the gas uptake measurements were at comparable
driving force of 3.4 and 3.7 MPa for the fuel gas mixture without
(Babu et al., 2013b) and with propane (Babu et al., 2013a) as a co-
guest. At the molecular level, it has been shown that in presence of
propane, the resulting sII hydrate contains almost similar amount
of CO2 as in sI CO2/H2 hydrate (Kumar et al., 2009a, 2009b). While
kinetic data in the literature and our recent works present
evidence of significantly enhanced rate if propane is present as a
co-guest in water dispersed in silica sand, the reason for this
peculiar behavior is not yet known.
The objective of this work is to elucidate the behavior of propane
as a co-guest in hydrate formation in silica sand. Based on our
morphology and kinetic observations we have proposed innovative
conceptual ideas for the application of gas hydrate technology.
2. Experimental
The gas mixtures employed in this study were supplied by
Soxal Private Limited. Silica sand was supplied by Sigma-Aldrich
(Catalog number 274739). The average particle size of the silica
sand was 329 mm (ranges from 150 to 630 mm). The pore volume
and porosity of the sand used was 0.217 cm3/g and 0.35 respec-
tively. Silica gel of particle size distribution of 75–200 mm and
100 nm pore diameter supplied by Silicycle was employed. Deio-
nized and distilled water was used in all experiments.
2.1. Morphology experiments
Morphology experiments were conducted by employing a mor-
phology apparatus described in detail in our previous work (Lim
et al., 2013). Briefly, the morphology apparatus consists of a crystal-
lizer immersed in a water bath. The crystallizer has two stainless
steel lid and central transparent acrylic column. A thermocouple was
inserted in the top port of the crystallizer to measure the gas phase
temperature. A pressure transducer and an analog pressure gauge
were used to measure the pressure of the crystallizer. An external
refrigerator was used to maintain the temperature of the setup.
A microscope coupled with digital camera was used to observe and
capture the images of the crystallizer during the hydrate formation
experiment. All experiments were conducted in batch manner.
18.43 g of silica sand was placed inside the crystallizer. 4.0 ml
of water was then dispersed into the interstitial pore space of the
silica sand particles to achieve 100% water saturation. The crystal-
lizer was closed and the thermocouple was connected. The
crystallizer was placed inside the water bath and then cooled to
274.2 K (experimental temperature). The crystallizer was flushed
three times with the gas mixture and then pressurized to experi-
mental pressure. Microscopic images are recorded for every 20 s
and the hydrate crystal growth is observed using the microscope.
2.2. Kinetic experiments
The detailed description of the apparatus is available in the
literature (Babu et al., 2013b). Briefly, the apparatus consists of a
crystallizer of volume 1.2 L and an inner diameter of 10.16 cm. The
crystallizer was immersed in a temperature controlled water bath.
A pressure transmitter is used to measure the pressure of the
crystallizer. A thermocouple was placed at the height of 1.2 cm from
the bottom of the bed to measure the temperature of the silica sand
bed in the crystallizer. An external refrigerator was used to maintain
the temperature of the setup. A DAQ coupled with computer was used
to record the data. All experiments were conducted in batch manner.
343
The detailed description of the preparation of silica sand bed and
hydrate formation procedure is given in detail elsewhere (Babu et al.,
2013b). Briefly, 645.16 g of silica sand was placed inside the crystal-
lizer for bed height of 5 cm. 140 ml of water was then dispersed into
the interstitial pore space of the silica sand particles. All the
experiments were conducted with 100% water saturation in silica
sand. The crystallizer was closed and the thermocouples were
connected. The crystallizer was placed inside the water bath and
then cooled to experimental temperature. The crystallizer was
flushed three times with the gas mixture and then pressurized to
experimental pressure. The data was recorded for every 20 s. Time
zero corresponds to the time at which the pressure and temperature
of the crystallizer reaches the experimental conditions. The crystal-
lizer pressure drops as a result of gas dissolution and hydrate
formation. When no significant pressure drop was observed, the
experiment was terminated. The crystallizer was depressurized to
atmospheric pressure and the system was heated to room tempera-
ture for decomposing of the hydrate crystals. After 2 h of standby
time, the formation experiment for cycle 2 was repeated. Memory
water refers to the water that has undergone hydrate formation.
The procedure for formation experiments were the same for silica
sand bed heights of 1.5 and 3.5 cm except the amount of silica sand
and water taken. For 1.5 cm bed height, 184. 33 g of silica sand and
40 ml of water was taken. For 3.5 cm bed height, 96.8 ml of water
was dispersed in 445.93 g of silica sand. The procedure to calculate
gas uptake, water conversion and normalized rate (NR30) of hydrate
formation are given below.
Calculation of normalized gas uptake:
Normalized gas uptake (NGt) at any given time t can be
calculated using the following equation
Δnt VðP=zRTÞ0  VðP=zRTÞt
NGt ¼ ¼ ð1Þ
nw nw
where Δnt is the moles of gas consumed at any time t. P, V, T are
the crystallizer pressure, volume of gas phase and temperature
of the liquid phase in the crystallizer. R is the ideal gas constant,
Z is the compressibility calculated by Pitzer's correlation and nw
is the number of moles of water employed in the experiment.
The following equation is used to further represent the normal-
ized gas uptake in a volume free basis
Δnt  22; 414
NGt ðv=vÞ ¼
Vw
Vw is the volume of water taken for the experiment in cm3.
Calculation of Water conversion to hydrate:
Conversion of water to hydrate is determined by using the
following equation
ðΔnH;↓ Þt
Conversion of water to hydrates ð%Þ ¼
nH 2 O
hydration number  100 ð2Þ
where ðΔnH;↓ Þt is the number of moles of gas consumed for
hydrate formation at the end of the experiment.
Calculation of the normalized rate of hydrate formation (NR30):
Normalized rate of hydrate formation can be calculated by
using the following equation
ðR30 Þ
NR30 ¼ ðmole of gas min  1 m  3 Þ ð3Þ
VW
where R30 is the rate (mol of gas/min) of hydrate growth
calculated by fitting the gas uptake data for the first 30 min
of hydrate growth with time and Vw is the volume of water
taken for the experiment in m3.
344 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351

3. Results and discussion
Morphology experiments were conducted for CO2/H2/C3H8 (38.1/
59.4/2.5%) gas mixture at 274.2 K and experimental pressures of
5.5 and 6.0 MPa respectively. The chosen experimental conditions
were similar to the one presented in our previous kinetic work (Babu
et al., 2013a) since the focus is on understanding the enhanced
kinetics with propane as a co-guest. The equilibrium pressure for the
gas mixture employed in this study at 274.2 K is 2.3 MPa (Babu et al.,
2013c).
Fig. 1 shows formation and growth of hydrate crystals in the
crystallizer at 6.0 MPa and 274.2 K. The time zero image (Fig. 1a)

Fig. 1. Sequential images of hydrate formation and growth in silica sand bed at 6.0 MPa and 274.2 K. Time zero corresponds to the nucleation of hydrate crystal. (a) Before
nucleation (IP), (b) at nucleation (IP), (c) 1 min from IP, (d) 4 min from IP, (e) 8 min from IP and (g) 17 min from IP.
 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
shows a clear gas phase and water dispersed in the interstitial
pore space between the silica sand particles. Hydrate crystals
nucleated at about 1.0 min from the start of the experiment shown
in Fig. 1b. Hydrate nucleation occurred in the interstitial pore
space between the silica sand particles. Nucleation was followed
by hydrate crystal growth in the upward direction towards the gas
phase. Fig. 1c shows the hydrate growth above the silica sand bed.
Water dispersed in the interstitial pores space between the silica
sand particles was drawn to the gas phase for hydrate growth. As
the time progresses, more water was drawn from the silica sand
bed and the hydrate front grows thicker and bigger. The transport
of water toward the hydrate growth front can be seen evidently
(Fig. 1d–g) from the change of the color of the silica sand particles.
At the end of the experiment, we could see a thick hydrate mass
above the silica sand bed. For a better visualization of the
morphology results, a video of the morphology experiment pre-
sented in Fig. 1 is given in Supporting Information (Supporting
Video_1). It has to be noted that the hydrate crystal growth was
not only on the walls of the crystallizer. Fig. S2 shows the
sequential images of the hydrate formation and growth in silica
sand bed at 5.5 MPa and 274.2 K. Hydrate crystals and its growth
above the silica sand bed at the center of the crystallizer can
clearly be seen. This ability to draw water above the silica sand bed
for hydrate growth resulted in enhanced kinetics of hydrate
formation and hence higher water conversion to hydrate. A video
of morphology experiment presented in Fig. S2 is available in
Supporting Information (Supporting Video_2). Another morphol-
ogy video of an experiment conducted at 6.0 MPa and 274.2 K is
presented as Supporting Information (Supporting Video_3).
Supplementary material related to this article can be found
online at doi:10.1016/j.ces.2014.06.044.
Based on the morphology experiments, it was evident that
water is being drawn outside the silica sand bed to the gas phase
for hydrate growth. This is possible due to the fact that in silica
sand, water is dispersed between the interstitial pore spaces of the
sand particles. Given this behavior, the height of silica sand bed
will play an important role and the depth until which such
migration of water due to capillary movement can happen from
the bed to the hydrate growth front in the gas phase needs to be
understood. In order to evaluate the effect of silica sand bed
height, kinetic experiments were carried out at 6.0 MPa and
274.2 K with three different silica sand bed heights of 1.5,
3.5 and 5.0 cm that were fully saturated with water.
Summary of the experimental conditions and results like
induction time, normalized gas uptake for 3 h after nucleation,
water conversion to hydrate and normalized rate of hydrate
Table 1
Summary of experimental conditions and results for fuel gas mixture with 2.5% propane. All experiments were conducted at 274.2 K and a starting pressure of 6.0 MPa.
Exp. no System Sand bed Solution stateb
height (cm)
1 CO2 (38.1)/H2 (59.4)/C3H8 (2.5)/water/silica sand 1.5 C1
2 C2
3 C1
4 C2
5 3.5 C1
6 C2
7 C1
8 C2
9 5.0 C1
10 C2
11 C1
12 C2
a
Hydration number of 10.05 (Kumar et al., 2009a, 2009b) for CO2/H2/C3H8 was used for calculation.
b
C1: Cycle 1; C2: Cycle 2.
345
formation are given in Table 1. Hydrate nucleation is stochastic
in nature and it depends on many factors like experimental
conditions, reactor configuration, solution state, gas saturation in
the solution etc. Due to the stochastic nature of induction time,
although the experimental condition and reactor configuration in
this study were same, different induction time was observed for
different silica sand bed height as can be seen in Table 1.
Effect of bed height on the hydrate growth and water conver-
sion to hydrate at 6.0 MPa and 274.2 K in presence of 2.5 mol%
propane is shown in Fig. 2a. Time zero corresponds to the hydrate
nucleation time (induction time). The average gas uptake for
hydrate growth and average water conversion to hydrate after
nucleation along with standard deviation are presented in the
figure. For experiment conducted with 1.5 cm silica sand bed
height, the average gas uptake was 2.53 and 3.74 times higher
than that of experiments conducted with 3.5 and 5.0 cm silica
sand bed height. Water conversion of up to 73.60 (73.61), 29.02
(72.77) and 19.42 (73.54) % was achieved in 3 h after hydrate
nucleation. As can be seen from the figure, the water conversion to
hydrate decreases as the bed height increases. This clearly shows
that the depth until which water is able to move up to the hydrate
growth front in the gas phase is important. The amount of water
converted to hydrates for the three bed heights are 29.44 (7 1.44),
28.09 (7 2.68) and 27.19 (74.96) ml for 1.5, 3.5 and 5.0 cm bed
heights respectively. This suggests that water deep inside the bed
beyond 1.5 cm is not participating in hydrate formation and
remains occluded in the sand and hence the overall conversion
of water to hydrates is higher for the 1.5 cm sand bed. We did not
choose to perform experiments for bed height less than 1.5 cm
due to the fact that the bed may require less than 40 ml of water
making the bed very small. We also chose not to perform
experiments with bed height greater than 5 cm as can be seen in
Fig. 2, it is expected that with a higher bed height, the gas uptake
will be less. Effect of silica sand bed height on the average
normalized rate of hydrate formation is shown in Fig. 2b. As can
be seen from the figure, increase in bed height resulted in decrease
in normalized rate of hydrate formation. Experiments conducted
with 1.5 cm bed height resulted in an average normalized rate of
75.00 mol min  1 m  3 and it was 3.18 and 3.86 times higher than
the experiments conducted with bed height of 3.5 and 5 cm
respectively. The extensive hydrate formation on top of the bed
for experiment 4 (given in Table 1) can be seen in Fig. S3. This
image was captured by quickly venting the residual gas phase after
the completion of the formation experiment and opening the
reactor. The reason for higher normalized gas uptake and water
conversion for smaller bed height could be that since water is
Induction Gas consumed after 3 h Water conversiona
time IT (min) from IT (mol/mol of H2O) (mol%) [7 stdev]
59.0 0.0706
1.3 0.0698 73.60
263.0 0.0772 [7 3.61]
1.0 0.0745
2.0 0.0300
3.3 0.0248 29.02
2.0 0.0308 [7 2.77]
3.0 0.0299
250.3 0.0203
0.7 0.0150 19.42
13.67 0.0193 [7 3.54]
0.33 0.0236
346 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
Fig. 2. Comparison of hydrate growth and water conversion (a) and normalized rate NR30 (b) between silica sand bed height of 1.5, 3.5 and 5.0 cm at 6.0 MPa and 274.2 K.
The number in the parenthesis in (a) refers to the number of experiments.
drawn towards the gas phase to the hydrate growth front,
resulting in drastic hydrate formation on top of the bed. For the
case of 3.5 and 5.0 cm bed heights, since the depth of bed is higher,
water at lower depths is not able to participate in hydrate
formation which could be due to the drastic hydrate formation
above bed which limits further capillary movement of water
present deeper in the bed to come out of the bed. We also
believe that the transport of gas through the hydrate mass (See
Fig. S3 in Supporting Information) to reach the depths where the
unreacted water is present seems to be limited by mass transfer
resistance. A cartoon image showing the initial and final state of
the different bed heights is presented in Fig. S4 (Supporting
Information).
In our experiments, we observed hydrate growth in the gas
phase where water was drawn outside the silica sand bed for
hydrate growth for the fuel gas mixture with 2.5 mol% propane as
a co-guest. We believe that the ability of the gas mixture to draw
the dispersed water above the silica sand for hydrate growth is
only due to the presence of propane as a co-guest. We performed
an experiment with CO2/H2 (40/60 mol%) system without propane
in the same morphology reactor at driving force of 2.9 MPa
(experimental conditions 8.5 MPa and 274.2 K). The sequential
images of the reactor contents are presented as Fig. S5. At the start
of the experiment, a clear gas phase and water dispersed in silica
sand was observed in the crystallizer. A small temperature spike
at about 33 min indicates the hydrate nucleation. It is noted that
the thermocouple for the morphology experiments was located in
the gas phase and hence a small heat increase is picked up by the
thermocouple. The change of color of the silica sand layer at the
gas–sand interface indicates the nucleation at the gas–sand inter-
face. Hydrates were observed at the sand–gas interface as indi-
cated in Fig. S5. We did not observe hydrate growth front towards
the gas phase. But the gas uptake measurement curve shows the gas
uptake for hydrate growth. A gas uptake of 0.0201 mol of gas/mol of
water was achieved after three hours of hydrate growth. It is noted
that we never observed hydrates significantly growing in the gas
phase for the CO2/H2 (40/60 mol%) system without propane. We
performed five morphology experiments (2 experiments at 8.5 MPa
and 3 experiments at 9.0 MPa) and our morphological observations
were consistent for the other four experiments as well but are not
presented. This observation also supports the macroscopic kinetic
results presented by Babu et al. (2013b) for CO2/H2 (40/60 mol%)
system at 8.5 and 9.0 MPa. The average gas uptake for the morphol-
ogy experiments were 0.0193 (70.001) and 0.0261 (70.012) mol/
mol of water after three hours of hydrate growth for experiments at
274.2 K and 8.5 and 9.0 MPa respectively.
It is noted that propane on its own forms structure II (sII). There
are not many kinetic studies on pure propane hydrates due to the
fact that there is very little driving force (ΔP) to perform kinetic
measurements of hydrate formation with propane gas (Bergeron
and Servio, 2008; Giavarini et al., 2003; Jensen et al., 2008; Rivera
and Janda, 2012). On the other hand, macroscopic kinetic data for
methane, methane/propane and methane/ethane gas/gas mixtures
in silica sand are available in the literature. A comparison of
literature kinetic data for pure methane gas (Linga et al., 2012;
Mekala et al., 2014), methane/ethane (91/9 mol%) (Linga et al.,
2012), and methane/propane (90.5/9.5 mol%) (Daraboina and
Linga, 2013) gas mixtures are presented in Fig. S6 (Supporting
Information). As can be seen in Fig. S6, the presence of propane as
a co-guest results in a significantly higher hydrate formation rate
in silica sand in spite of having the lowest driving force compared
to methane and methane/propane. It is noted that all the data
presented in Fig. S6 were conducted with 100% water saturation.
Bagherzadeh et al. (2011) reported based on magnetic resonance
imaging technique that methane hydrate formation in 100% water
saturated silica sand appears inside the sand bed at varying rates
and nucleation appeared at different locations inside the sand bed.
Based on a morphology study, Babu et al. (2013d) reported that
methane hydrate formation and growth appeared inside the
interstitial pore space between the silica sand particles. Recently,
Mekala et al.(2014) reported that for methane/silica sand/water
with 100% water saturation, the water conversions were about
7.71% (71.65) after 3 h of hydrate growth (see Fig. S6). Based on
the morphology study (Figs. 1, S2 and S5 along with the videos
presented in Supporting Information) and the kinetic data (pre-
sented in Table 1 and Fig. 2) in this work, significantly higher rate
of hydrate formation can be attributed to the presence of propane
as a co-guest which is able to draw water that is dispersed in the
interstitial pore spaces between the silica sand particles towards
to the gas phase for hydrate formation. This behavior of hydrate
formation in presence of propane as a co-guest is therefore
“unusual” and not predictable since this was not observed for all
the other guest gas/gas mixtures without propane, given that
water was dispersed in sand for all the gas/gas mixtures (CO2/H2;
CH4, CH4/C2H6).
In order to understand if this behavior of propane as a co-guest
can be observed in other contact modes, we performed experi-
ments at the same experimental conditions of 6.0 MPa and 274.2 K
with silica gel (water is dispersed inside the pore of the gels) and a
traditional stirred tank reactor (bulk gas liquid contact mode).
A silica gel of 100 nm pore diameter and 0.83 ml/g pore volume
supplied by Silicylce was chosen as it was reported to be the best
 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
Fig. 3. Comparison of hydrate growth and water conversion (a) and normalized rate NR30 (b) between silica sand, silica gel and stirred tank reactor at comparable
experimental conditions (refer to Table S1 for silica gel and STR data). The number in the parenthesis in (a) refers to the number of experiments.
Fig. 4. Effect of Silica sand bed height on hydrate growth and water conversion
to hydrate from natural gas (CH4/C2H6/C3H8) mixture at 5.5 MPa and 277.2 K.
The average data along with standard deviation are presented. The number in
parenthesis in the legend indicates the number of experiments.
in the literature (Adeyemo et al., 2010; Seo and Kang, 2010). The
detailed description of experimental procedure for dispersing
water into the silica gel is available in our previous work (Babu
et al., 2013b) but a brief description of the procedure is given in
Supporting Information. Similarly, the experimental apparatus and
procedure for our mechanically agitated stirred tank reactor is
available in detail in our previous work (Veluswamy et al., 2014a)
and a brief description is given in supporting information. Fig. 3
compares the hydrate growth curves for experiments conducted
between sand, silica gel and stirred tank reactor. It is interesting to
note that in a stirred tank reactor and in silica gel such a behavior
of propane as a co-guest is not observed. The experimental data
like the induction time, and experimental conditions are reported
in Table S1. For the case of silica gel, the water is confined inside
the pore of the gels whereas for the case of silica sand the water is
dispersed in the interstitial space between the sand particles thus
facilitating the migration and capillary movement of water in
sand. Where as in the case of stirred tank reactor crystal agglom-
eration at the gas/liquid interface limits the hydrate growth rate
after initial period of hydrate growth.
We also evaluated the effect of bed height by independently
choosing a three component natural gas mixture of CH4/C2H6/C3H8
(93.0%/5.0%/2.0%) with propane as a co-guest. For this gas mixture,
we only choose to test 1.5 and 5 cm bed heights. Experiments were
performed at 5.5 MPa and 277.2 K (driving force of 4.1 MPa,
347
equilibrium pressure calculated using CSMGem of Sloan and Koh
(2008)) and the results are summarized in Table 2. The hydrate
growth curves for the bed heights of 1.5 and 5 cm for the natural
gas mixture are presented in Fig. 4. As can be seen from the figure,
silica sand bed height of 1.5 cm resulted in a gas uptake of 0.1178
(70.0032) mol of gas/mol of water and water conversion of
81.53% (72.12) within one hour after nucleation. While the gas
uptake and water conversion to hydrate in an hour after nuclea-
tion in silica sand bed height of 5 cm were 0.0683 ( 70.0172) mol
of gas/mol of water and 47.19% (711.87) respectively. As can be
seen in Table 2, the normalized rate is 3.3 times higher for the
1.5 cm bed compared to the 5.5 cm bed height.
Apart from the silica sand bed height, the particle size of the silica
sand (as it affects the porosity), driving force and composition of
propane as a co-guest in the gas mixture may also affect the extent of
water migration from the silica sand for hydrate formation. In order
to understand the effect of experimental pressure or driving force
(ΔP) on this behavior of propane as a co-guest, we performed kinetic
experiments at 4.0 MPa for the natural gas mixture and the results
are also presented in Table 2. The effect of experimental pressure or
driving force on the gas uptake (1 h from Induction Point) and
normalized rate of hydrate formation for first 30 min after nucleation
for natural gas mixture at comparable silica sand bed height is
presented in Fig. 5. The experimental pressure employed in this
study and Linga et al. (2012) were 4.0 MPa (ΔP¼2.6 MPa), 5.5 MPa
(ΔP¼4.1 MPa) and 8.0 MPa (ΔP¼ 6.6 MPa) respectively. We can infer
from Fig. 5 that the ability of propane as a co-guest in the natural gas
mixture to draw water towards the gas phase for hydrate formation
appears not be influenced significantly between 4.0, 5.5 and 8.0 MPa
starting pressure respectively. Interestingly, Daraboina et al. (2011)
employed a small reactor with rigorous mixing in an stirred auto-
clave at 8.0 MPa and 275.2 K using the same natural gas mixture
composition and reported gas uptake of 0.0173 mol of gas which
translates to 38.78 v/v after 10 h of hydrate formation.
It is also interesting to note that if we compare the water
conversion (Tables 1 and 2), the normalized rate (Tables 1 and 2)
and the hydrate growth profile patterns between the two gas
mixtures employed in this work, they are different even though
the propane composition was relatively in the same range
of 2–2.5 mol% for each gas mixture. For instance, the water
conversion for 1.5 cm bed height for the two gas mixtures were
about 73.6% and 85.16% for CO2/H2/C3H8 (38.1/59.4/2.5%) and CH4/
C2H6/C3H8 (93.0%/5.0%/2.0%) respectively. Similarly, the normal-
ized gas uptake rate for 1.5 cm bed height was 75.00 and
185.19 mol of gas min  1 m  3 for CO2/H2/C3H8 (38.1/59.4/2.5%)
348 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351

Table 2
Summary of Experimental conditions and results for the natural gas mixture. All experiments were conducted at 277.2 K.

Exp. System Pressure Bed Solution Induction Gas consumed after 3 h from Water conversiona NR30
no (MPa) height stateb time IT (min) IT (mol/mol of H2O) (mol %) [7 stdev] (mol min  1 m  3)
(cm) [ 7stdev]

13 CH4 (93.0)/C2H6 (5.0)/C3H8 5.5 1.5 C1 1.7 0.1241 85.16 185.19

14 (2.0)/Water/Silica Sand C2 1.7 0.1225 [7 0.51] [ 7 55.65]
15 C1 0.7 0.1234

16 5.0 C1 0.7 0.1090

17 C2 0.7 0.1096 75.78 55.56
18 C1 2.0 0.1104 [7 0.46] [ 7 28.87]

19 4.0 1.5 C1 0.7 0.1288

20 C2 0.7 0.1313 89.30 109.26
21 C1 0.7 0.1275 [7 1.34] [ 7 3.21]

a
Hydration number of 6.91 (calculated at equilibrium by CSMGem of Sloan and Koh (2008)) for CH4/C2H6/C3H8 was used for calculation.
b
C1: Cycle 1; C2: Cycle 2.

3.1.1. Seawater desalination

Fig. 5. Effect of experimental pressure (or driving force) on gas uptake (1 h of
hydrate growth) and normalized rate of hydrate formation for the natural gas (CH4/
C2H6/C3H8) mixture at 4.0 MPa (this work), 5.5 MPa (this work) and 8.0 MPa (Linga
et al., 2012) at comparable silica sand bed heights of 1.5 (this work) and 1.8 cm
(Linga et al., 2012) respectively.
and CH4/C2H6/C3H8 (93.0%/5.0%/2.0%) respectively. Linga et al.
(2012) reported after evaluating different gas/gas mixtures in both
stirred vessel and fixed bed with sand that the pattern of hydrate
formation growth profiles were different for each gas/gas mixture
for any given contact mode. They reported that the possible
reasons for this could be due to several factors, like the driving
force at the start of the experiment, the fractionation effect in the
gas phase during hydrate formation that can dynamically impact the
hydrate formation kinetics for each gas/gas mixture, the relative
rate at which each guest gas is entering into the cages at the
molecular level and so on. Hence, it is not surprising that even in
this work the two gas mixtures had different final water conversions,
different normalized rates and different growth profile patterns
between them.
Hydrate based desalination (HBD) process was proposed as one
of the freezing approaches to desalinate seawater several decades
ago. Major challenges for the HBD process to being commercia-
lized were slow kinetics of hydrate formation and the separation
of hydrate crystals from the brine. The ability of propane as a
co-guest to draw water from the silica sand bed can be effectively
used for the HBD process addressing the slow kinetics and effective
separation of the hydrate crystals. A schematic of the HBD process is
shown in Fig. 6. Initially the sea water should be pumped into the
hydrate crystallizer to fill the interstitial pore space available between
the silica sand particles placed in the crystallizer. Crystallizer will
then be cooled to the experimental temperature suitable for hydrate
formation. This can be done by utilizing LNG cold energy during
re-gasification or a traditional ammonia refrigeration system. When
exposed to a gas mixture containing propane as a co-guest, water
will be drawn out from the interstitial pore space to form hydrates
above the silica sand layer which we call the hydrate zone. The salts
would preferably get occluded in the silica sand and in the left over
brine solution below the retainer as shown in the figure. It must be
emphasized that some salts might end up adsorbed to the hydrates
by moving along with the water. Literature studies have shown that
even with a well-mixed agitated system it is possible to achieve up to
80% removal of dissolved minerals (Park et al., 2011). One can expect
that the % removal of dissolved minerals to be higher than 80% with
our approach but this needs to be investigated further. After the
completion of the hydrate formation, the hydrate crystals can be
separated from the hydrate zone and decomposed to yield salt free
water. The ability to remove the crystals from the hydrate zone will
be an operational challenge but can be overcome with novel reactor
designs. Water conversion can be related to water recovery employed
in desalination. It is possible to achieve significantly higher water
recovery in the range of 60–80% in 1 h operating at less than 60 bars
(6 MPa). Depending on the concentration of propane as a co-guest
and the type of other constituents of the gas mixture there is room to
achieve higher conversions and also to minimize the operating
pressure for the HBD process. This conceptual process is currently
under investigation at a bench scale.

3.1. Potential applications
In order to exploit this behavior of propane as a co-guest in
sand, we propose conceptual flow sheets for applying the clathrate
process for a standalone seawater desalination method and an
innovative approach to simultaneously capture carbon dioxide and
desalinate seawater in the following section.
3.1.2. Simultaneous CO2 capture and seawater desalination
The ability to draw water and enhanced kinetics in presence of
propane in fuel gas mixture can be employed for simultaneous CO2
capture and desalination. Schematic of the conceptual flow sheet
for the simultaneous pre-combustion CO2 capture along with
desalination is shown in Fig. 7. In the first step, hydrates are
crystallized from seawater and fuel gas mixture with 2.5% propane
 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351 349

Fig. 6. Schematic concept of the HBD process for seawater desalination.

Fig. 7. Schematic flow sheet of a single stage simultaneous CO2 capture and Desalination operation of HBD process.

at suitable pressure and temperature in the hydrate crystallizer.
The salts in the seawater do not participate in hydrate formation
and hence get preferably occluded in the silica sand. At the end of
the hydrate crystallization step, the hydrogen rich residual gas
phase can be sent to gas turbine for energy generation, solid
hydrate crystals above the silica sand layer can be recovered, and
the decomposed CO2 gas can be sent for further purification or
sequestration/storage or for utilization. The concentrated brine in
the hydrate crystallizer can be flushed out by washing the silica
bed with sea water. This innovative process can run with sig-
nificant cost advantage as it combines two energy intensive
processes into one and can be very attractive from an energy
and environment perspective. While the concept of simultaneous
CO2 capture and seawater desalination is attractive, it must be
noted that there may be several engineering/process challenges
and scale up issues that may have to be overcome.
3.1.3. Natural gas storage and transport
Apart from CO2 capture and desalination, hydrates are considered
as potential medium for storage and transport of gases especially
natural gas (Gudmundsson et al., 1998; Gudmundsson and
Borrehaug, 1996; Gudmundsson et al., 1995) Mitsui Engineering has
pioneered the hydrate technology for the past 10–15 years and have
also considered natural gas hydrates for stationery storage and
domestic transport in addition to the large scale storage and
transport from exploration sites to Japan (Kanda et al., 2005;
Nogami et al., 2008; Takahashi et al., 2008; Watanabe et al., 2008).
Advantages of storing and transporting natural gas as hydrates are
safety, lower process volume and lower capital cost. It is noted that
natural gas consists of about 0.5–3.0% propane depending on the
source from where it is produced. Employing a fixed bed column
with silica sand as porous media can enhance the kinetics of hydrate
formation of natural gas. The hydrate formed above the silica sand
layer can be separated, pelletized and can be stored/transported. As
can be seen in Fig. 5, in 1 h, we can achieve a natural gas storage of
137.9 (79.36) v/v at a very high production rate of 2.4474 min  1 or
109.26 mol of gas m  3 min  1 just at 40 bars. For a stirred tank
reactor, Linga et al. (2012) reported a gas uptake of 36.48 v/v for 1 h
of hydrate formation for the same gas mixture composition at a high
pressure of 80 bars and 277.2 K (calculated by us based on the data of
Linga et al. (2012)). Such an approach can eliminate the need of
adding surfactants like sodium dodecyl sulfate (SDS) to improve the
rate of hydrate formation for NGH storage and transport. A more
pertinent issue for the usage of SDS is its toxicity when employed in a
large scale.
350 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
It is noted that while the ability of propane as a co-guest to
draw water from porous media was demonstrated in this work
along with potential applications to exploit this behavior, this
effect can be a affected/influenced by bed porosity and the
composition of propane as a co-guest, which needs to be further
investigated in detail and is currently under progress.
4. Conclusion
Unusual behavior of hydrate formation from gas mixture
containing propane in silica sand bed was reported. We observed
that propane as the ability to draw water dispersed in the silica
sand bed to the hydrate formation zone and migration of water to
the gas phase resulted in enhanced hydrate formation. We also
investigated the effect of silica sand bed height on the migration of
water to sustain hydrate growth. Based on the mechanism of
hydrate formation observed in our study, we have proposed
conceptual design for seawater desalination and an innovative
approach for simultaneous CO2 capture and desalination.
Supporting information
Sequential images of hydrate formation and growth at 5.5 MPa
and 274.2 K in silica sand bed from CO2/H2/C3H8 (38.1%/59.4%/
2.5%) gas mixture. Conversion of water to hydrates during hydrate
growth phase of CH4, CH4/C3H8, CH4/C2H6 and CH4/C2H6/C3H8
(93.0%/5.0%/2.0%) gas mixtures were given. This information is
available free of charge via the Internet at www.elseiver.com.
Acknowledgment
The financial support from the Ministry of Education's AcRF
Tier 1 (R-279-000-386-112 and R-279-000-420–750) is greatly
appreciated. Rajnish Kumar appreciates the financial support from
the Department of Science and Technology in India (GAP-294026).
Appendix. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.ces.2014.06.044.
References
Adeyemo, A., Kumar, R., Linga, P., Ripmeester, J., Englezos, P., 2010. Capture of
carbon dioxide from flue or fuel gas mixtures by clathrate crystallization in a
silica gel column. Int. J. Greenh. Gas Control 4, 478–485.
Babu, P., Ho, C.Y., Kumar, R., Linga, P., 2014. Enhanced kinetics for the clathrate
process in a fixed bed reactor in the presence of liquid promoters for pre-
combustion carbon dioxide capture. Energy 70, 664–673.
Babu, P., Kumar, R., Linga, P., 2013a. Medium pressure hydrate based gas separation
(HBGS) process for pre-combustion capture of carbon dioxide employing a
novel fixed bed reactor. Int. J. Greenh. Gas Control 17, 206–214.
Babu, P., Kumar, R., Linga, P., 2013b. Pre-combustion capture of carbon dioxide in a
fixed bed reactor using the clathrate hydrate process. Energy 50, 364–373.
Babu, P., Yang, T., Veluswamy, H.P., Kumar, R., Linga, P., 2013c. Hydrate phase
equilibrium of ternary gas mixtures containing carbon dioxide, hydrogen and
propane. J. Chem. Thermodyn. 61, 58–63.
Babu, P., Yee, D., Linga, P., Palmer, A., Khoo, B.C., Tan, T.S., Rangsunvigit, P., 2013d.
Morphology of methane hydrate formation in porous media. Energy Fuels 27,
3364–3372.
Bagherzadeh, S.A., Moudrakovski, I.L., Ripmeester, J.A., Englezos, P., 2011. Magnetic
resonance imaging of gas hydrate formation in a bed of silica sand particles.
Energy Fuels 25, 3083–3092.
Bergeron, S., Servio, P., 2008. Reaction rate constant of propane hydrate formation.
Fluid Phase Equilibria 265, 30–36.
Chatti, I., Delahaye, A., Fournaison, L., Petitet, J.P., 2005. Benefits and drawbacks of
clathrate hydrates: a review of their areas of interest. Energy Convers. Manage.
46, 1333–1343.
Daraboina, N., Linga, P., 2013. Experimental investigation of the effect of poly-N-
vinyl pyrrolidone (PVP) on methane/propane clathrates using a new contact
mode. Chem. Eng. Sci. 93, 387–394.
Daraboina, N., Linga, P., Ripmeester, J., Walker, V.K., Englezos, P., 2011. Natural gas
hydrate formation and decomposition in the presence of kinetic inhibitors. 2.
Stirred reactor experiments. Energy Fuels 25, 4384–4391.
Duc, N.H., Chauvy, F., Herri, J.M., 2007. CO2 capture by hydrate crystallization – a
potential solution for gas emission of steelmaking industry. Energy Convers.
Manage. 48, 1313–1322.
Englezos, P., 1993. Clathrate hydrates. Ind. Eng. Chem. Res. 32, 1251–1274.
Eslamimanesh, A., Mohammadi, A.H., Richon, D., Naidoo, P., Ramjugernath, D., 2012.
Application of gas hydrate formation in separation processes: a review of
experimental studies. J. Chem. Thermodyn. 46, 62–71.
Gholinezhad, J., Chapoy, A., Tohidi, B., 2011. Separation and capture of carbon
dioxide from CO2/H2 syngas mixture using semi-clathrate hydrates. Chem. Eng.
Res. Des. 89, 1747–1751.
Giavarini, C., Maccioni, F., Santarelli, M.L., 2003. Formation kinetics of propane
hydrates. Ind. Eng. Chem. Res. 42, 1517–1521.
Gudmundsson, J., Andersson, V., Levik, O., Parlaktuna, M., 1998. Hydrate concept for
capturing associated gas. In: Proceedings of the SPE European Petroleum
Conference, the Hague, Netherlands, 20 October.
Gudmundsson, J., Borrehaug, A., 1996. Natural gas hydrate-an alternative to
liquefied natural gas? Pet. Rev. 50, 232–235.
Gudmundsson, J., Hveding, F., Borrehaug, A., 1995. Transport of natural gas as frozen
hydrate. In: Proceedigns of the 5th International Offshore and Polar Engineer-
ing Conference, The Hague, The Netherlands, pp. 282–288.
Jensen, L., Thomsen, K., von Solms, N., 2008. Propane hydrate nucleation: experi-
mental investigation and correlation. Chem. Eng. Sci. 63, 3069–3080.
Kanda, H., Uchida, K., Nakamura, K., Suzuki, T., 2005. Economics and energy
requirements on natural gas ocean transportation in form of natural gas
hydrate (NGH) pellets. In: Proceedings of 5th International Conference on Gas
Hydrate, Trondheim, Norway.
Kang, S.P., Lee, H., 2000. Recovery of CO2 from flue gas using gas hydrate:
thermodynamic verification through phase equilibrium measurements.
Environ. Sci. Technol. 34, 4397–4400.
Kumar, A., Sakpal, T., Linga, P., Kumar, R., 2013. Influence of contact medium
and surfactants on carbon dioxide clathrate hydrate kinetics. Fuel 105,
664–671.
Kumar, R., Englezos, P., Moudrakovski, I., Ripmeester, J.A., 2009a. Structure and
composition of CO2/H2 and CO2/H2/C3H8 hydrate in relation to simultaneous
CO2 capture and H2 production. AIChE J. 55, 1584–1594.
Kumar, R., Lang, S., Englezos, P., Ripmeester, J., 2009b. Application of the ATR-IR
spectroscopic technique to the characterization of hydrates formed by CO2,CO2/
H2 and CO2/H2/C3H8. J. Phys. Chem. A 113, 6308–6313.
Lee, H., Lee, J.W., Kim, D.Y., Park, J., Seo, Y.T., Zeng, H., Moudrakovskr, I.L., Ratcliffe, C.
I., Ripmeester, J.A., 2005. Tuning clathrate hydrates for hydrogen storage.
Nature 434, 743–746.
Li, X.-S., Xia, Z.-M., Chen, Z.-Y., Wu, H.-J., 2011a. Precombustion capture of carbon
dioxide and hydrogen with a one-stage hydrate/membrane process in the
presence of tetra-n-butylammonium bromide (TBAB). Energy Fuels 25,
1302–1309.
Li, X.-S., Xu, C.-G., Chen, Z.-Y., Wu, H.-J., 2010. Tetra-n-butyl ammonium bromide
semi-clathrate hydrate process for post-combustion capture of carbon dioxide
in the presence of dodecyl trimethyl ammonium chloride. Energy 35,
3902–3908.
Li, X.-S., Xu, C.-G., Chen, Z.-Y., Wu, H.-J., 2011b. Hydrate-based pre-combustion
carbon dioxide capture process in the system with tetra-n-butyl ammonium
bromide solution in the presence of cyclopentane. Energy 36, 1394–1403.
Lim, Y.-A., Babu, P., Kumar, R., Linga, P., 2013. Morphology of carbon dioxide–
hydrogen–cyclopentane hydrates with or without sodium dodecyl sulfate.
Cryst. Growth Des. 13, 2047–2059.
Linga, P., Daraboina, N., Ripmeester, J.A., Englezos, P., 2012. Enhanced rate of gas
hydrate formation in a fixed bed column filled with sand compared to a stirred
vessel. Chem. Eng. Sci. 68, 617–623.
Linga, P., Kumar, R., Englezos, P., 2007. The clathrate hydrate process for post and
pre-combustion capture of carbon dioxide. J. Hazard. Mater. 149, 625–629.
Linga, P., Kumar, R., Lee, J.D., Ripmeester, J., Englezos, P., 2010. A new apparatus to
enhance the rate of gas hydrate formation: application to capture of carbon
dioxide. Int. J. Greenh. Gas Control 4, 630–637.
Mekala, P., Babu, P., Sangwai, J.S., Linga, P., 2014. Formation and dissociation kinetics
of methane hydrates in seawater and silica sand. Energy Fuels 28, 2708–2716.
Nogami, T., Oya, N., Ishida, H., Matsumoto, H., 2008. Development of natural gas
ocean transportation chain by means of natural gas hydrate (NGH). In:
Proceedings of the 6th International Conference on Gas Hydrates.
Park, K.N., Hong, S.Y., Lee, J.W., Kang, K.C., Lee, Y.C., Ha, M.G., Lee, J.D., 2011. A new
apparatus for seawater desalination by gas hydrate process and removal
characteristics of dissolved minerals (Naþ , Mg2 þ, Ca2þ , Kþ , B3 þ ). Desalina-
tion 274, 91–96.
Park, S., Lee, S., Lee, Y., Lee, Y., Seo, Y., 2013. Hydrate-based pre-combustion capture
of carbon dioxide in the presence of a thermodynamic promoter and porous
silica gels. Int. J. Greenh. Gas Control 14, 193–199.
Rivera, J.J., Janda, K.C., 2012. Ice particle size and temperature dependence of the
kinetics of propane clathrate hydrate formation. J. Phys. Chem. C 116,
19062–19072.
Rufford, T.E., Smart, S., Watson, G.C.Y., Graham, B.F., Boxall, J., Diniz da Costa, J.C.,
May, E.F., 2012. The removal of CO2 and N2 from natural gas: a review of
 P. Babu et al. / Chemical Engineering Science 117 (2014) 342–351
conventional and emerging process technologies. J. Pet. Sci. Eng. 94–95,
123–154.
Seo, Y., Kang, S.P., 2010. Enhancing CO2 separation for pre-combustion capture with
hydrate formation in silica gel pore structure. Chem. Eng. J. 161, 308–312.
Sloan, E.D., Koh, C.A., 2008. Clathrate Hydrates of Natural Gases. CRC Press.
Tajima, H., Yamasaki, A., Kiyono, F., 2004. Energy consumption estimation for
greenhouse gas separation processes by clathrate hydrate formation. Energy 29,
1713–1729.
Takahashi, M., Moriya, H., Katoh, Y., Iwasaki, T., 2008. Development of natural gas
hydrate (NGH) pellet production system by bench scale unit for transportation
and storage of NGH pellet. In: Proceedings of 6th International Conference on
Gas Hydrate, Vancouver, Canada.
Veluswamy, H.P., Chin, W.I., Linga, P., 2014a. Clathrate hydrates for hydrogen
storage: the impact of tetrahydrofuran, tetra-n-butylammonium bromide and
cyclopentane as promoters on the macroscopic kinetics. Int. J. Hydrogen Energy ,
http://dx.doi.org/10.1016/j.ijhydene.2014.01.054.
351
Veluswamy, H.P., Kumar, R., Linga, P., 2014b. Hydrogen storage in clathrate
hydrates: current state of the art and future directions. Appl. Energy 122,
112–132.
Watanabe, S., Takahashi, S., Mizubayashi, H., Murata, S., Murakami, H., 2008. A
demonstration project of NGH land transportation system. In: Proceedings of
the Sixth International Conference on Gas Hydrates, Vancouver, Canada, July,
pp. 6–10.
Xu, C.-G., Li, X.-S., 2014. Research progress of hydrate-based CO2 separation and
capture from gas mixtures. RSC Adv. 4, 18301–18316.
Xu, C.-G., Zhang, S.-H., Cai, J., Chen, Z.-Y., Li, X.-S., 2013. CO2 (carbon dioxide)
separation from CO2–H2 (hydrogen) gas mixtures by gas hydrates in TBAB
(tetra-n-butyl ammonium bromide) solution and Raman spectroscopic analy-
sis. Energy 59, 719–725.
Yang, M., Song, Y., Ruan, X., Liu, Y., Zhao, J., Li, Q., 2012. Characteristics of CO2
hydrate formation and dissociation in glass beads and silica gel. Energies 5,
925–937.