Applied Energy 158 (2015) 133–141

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Evaluation of CO2 removal from a CO2 + CH4 gas mixture using gas
hydrate formation in liquid water and THF solutions
Dong-Liang Zhong a,b,⇑, Zheng Li a, Yi-Yu Lu b, Jia-Le Wang a, Jin Yan b
a
Key Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education of China, Chongqing University, Chongqing 400044, China
b
State Key Laboratory of Coal Mine Disaster Dynamics and Control, Chongqing University, Chongqing 400044, China

h i g h l i g h t s

Gas hydrate formation was used for CO2 removal from a simulated shale gas.

The CO2 selectivity was reduced as the driving force increases.

The CO2 selectivity was compromised in the presence of THF and SDS.

Higher CO2 recovery and separation factor was obtained in liquid water.

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 performance of hydrate formation for CO2 removal from a simulated recovered shale gas
Received 5 May 2015 (40 mol% CO2 and 60 mol% CH4) was evaluated in liquid water, and THF and THF/SDS solutions.
Received in revised form 27 July 2015 Experiments were carried out at a fixed temperature of 277.15 K and in the pressure range of (2.8–6.7)
Accepted 15 August 2015
MPa. The impact of driving force (overpressure) on hydrate growth and CO2 selectivity was studied both
Available online 28 August 2015
in liquid water and in THF solutions. It was found that higher driving force resulted in a reduction of the
final gas uptake as well as a significant decrease of the CO2 recovery and separation factor. The results
Keywords:
indicated that mass transfer for hydrate growth was greatly hindered at high driving force, and the com-
Gas hydrates
Shale gas
petition between CH4 and CO2 molecules for the occupancy of hydrate cavities became stronger with the
Carbon dioxide capture increase of driving force. Although the kinetics of hydrate formation with the CO2/CH4 gas mixture was
Gas separation improved by adding THF and SDS into liquid water, the selectivity for CO2 removal from the CO2/CH4 gas
Kinetics mixture was compromised as compared to that in liquid water. The highest CO2 recovery (52%) and sep-
aration factor (8.8) were obtained at a low driving force (2.5 MPa) in liquid water, which were much
higher than those obtained in THF and THF/SDS solutions. As a result, enhancing the kinetics of hydrate
formation without reducing CO2 selectivity is the key factor to improve the hydrate based gas separation
process for CO2 removal from the CO2/CH4 gas mixture.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction 18% C2H6; 2% CO2, 80% CH4, 12% C2H6 and 6% C3H8). However, the
shale gas typically undergoes a number of stages at the wellhead or
Shale gas is an unconventional natural gas and has been recog- at the processing plant to remove acid gases, water, nitrogen, etc.
nized as a potential energy resource due to its huge reserves on the Martin et al. [6] considered shale gas as pure methane and pre-
globe [1,2]. The composition of shale gas (mainly methane, ethane, sented a superstructure optimization approach for the production
propane, CO2 and N2) varies significantly from site to site of the of hydrogen from the shale gas based on the above fact. A major
actual location of shale reservoirs [3,4]. Li et al. [5] used a compo- concern in the production of shale gas is to explore an environment
sitional model to study the effect of apparent permeability for dif- friendly recovery method [7,8]. Recently, a novel technique that
ferent components of two model shale gases (2% CO2, 80% CH4, and has been proposed to enhance the recovery of shale gas is to inject
supercritical CO2 (SC-CO2) into the shale gas reservoir instead of
using aqueous hydraulic fracturing [9,10]. It was found that the
⇑ Corresponding author at: College of Power Engineering, Chongqing University, permeability of shale could be improved and CH4 stored in the
Chongqing 400044, China. Tel./fax: +86 23 65102473. shale rocks can be replaced by the injected CO2. This new method
E-mail address: zhongdl@cqu.edu.cn (D.-L. Zhong).

http://dx.doi.org/10.1016/j.apenergy.2015.08.058
0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.
134 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141

not only promotes the production of shale gas but also offers an tified mass purity of 99.9% were purchased from Chongqing Orien-
option of geological storage for the greenhouse gas CO2. In case tal Chemical Co., Ltd. The uncertainty in the mass fraction of THF
of real applications, the recovered shale gas by SC-CO2 might be and SDS was 0.1%. Deionized water was used in all experimental
a CO2/CH4 mixture containing impurities such as C2H6 and C3H8. runs.
For the sake of simplification, we used pure CH4 representing shale
gas in the preliminary lab-scale experiments as reported in the lit- 2.2. Apparatus
erature [6] and considered the recovered shale gas by SC-CO2 is a
CO2/CH4 mixture. Therefore, it is required to remove CO2 from A detailed description of the experimental apparatus was pre-
the CO2/CH4 gas mixture so that the recovered shale gas can be sented elsewhere [33]. Briefly, it consists of a high pressure vessel
concentrated and the captured CO2 can be reused for the CO2– (316 stainless steel) with the maximum operating pressure of
CH4 replacement process. 10 MPa. The volume of the vessel was 375 cm3. A magnetic stir
Gas hydrates are icelike crystalline inclusion compounds bar was employed to agitate the vessel contents, which was cou-
formed by water and small sized gas molecules (CH4, C2H6, CO2, pled with an electromagnetic plate beneath the vessel. Tempera-
etc.) under suitable temperature and pressure conditions [11– ture in the vessel was controlled by an external refrigerator. Two
13]. Gas hydrates formation/decomposition has been identified as copper–constantan thermocouples (Omega Engineering Corpora-
a promising technology for various industrial purposes such as tion, USA) with the uncertainty of ±0.1 K were inserted into the
gas storage and transportation [14,15], desalination [16,17], refrig- vessel to measure the liquid and gas temperature. A pressure trans-
eration and air conditioning [18,19], and gas separation [20–22], ducer (EJX430A, Yokogawa Electric Corporation, Tokyo, Japan) with
particularly CO2 capture from flue, fuel and industrial gases [23– the uncertainty of 0.04% in the range of 0–16 MPa was used to
28]. measure the pressure inside the vessel. The temperature and pres-
The separation of CO2 from the CO2/CH4 gas mixture using gas sure data were collected by a data acquisition unit (GE Automation
hydrate formation has already attracted attention. By using gas Corporation, USA) and recorded in a computer every 10 s. A gas
chromatography and Raman spectroscopy it was found that CO2 chromatograph (GC-2014, Shimadzu Corporation, Kyoto, Japan)
was incorporated into the hydrate cages preferentially during with the uncertainty of 0.1 mol% was used to analyze the composi-
CO2/CH4 hydrate formation [29]. In another study it was found tion of the gas mixture remaining in the vessel as well as that
that the addition of catalyst cetyl trimethyl ammonium bromide released from the hydrates at the end of the experiment.
(CTAB) to the hydrate formation system performed better for CO2
removal than the promoter tetrahydrofuran (THF) [30]. Semi- 2.3. Procedures
clathrate hydrates were also employed to capture CO2 from raw
natural gas or biogas (CO2/CH4) [31]. Recent phase equilibrium The experiments were carried out in a batch mode and the pro-
data of semiclathrate hydrates formed with the CO2/CH4 mixture cedure was given as follows. Prior to the experiments, the vessel
in the presence of quaternary ammonium salts (TBAB, TBAC, and was cleaned with deionized water and dried. Then 140 cm3 aque-
TBAF) were reported. Ricaurte et al. [32] studied the kinetics of ous solution (deionized water, 1 mol% THF solution, or 1 mol%
CO2 separation from a CO2/CH4 gas mixture by hydrate formation THF solution containing 500 ppm SDS) was filled in the vessel.
in the presence of THF. Since phase equilibrium data for the CO2– The vessel and tubing were purged thrice using the CO2/CH4 gas
CH4–THF mixed hydrates were not measured driving forces for mixture to remove the air remaining in the system. Once the
hydrate growth were not determined quantitatively. Recently, experimental temperature and pressure in the vessel reached
we reported the phase equilibrium data of gas hydrates formed desired values, the inlet and outlet valves of the vessel were closed
from a CO2/CH4 mixture in the presence of THF. Based on the to isolate the vessel from the gas cylinder. The stirrer was started at
phase equilibrium data we obtained the heat of hydrate dissoci- a constant speed of 150 rpm. This was considered to be time zero
ation and confirmed that the CO2/CH4 mixed gas hydrates formed for hydrate formation experiments. Meantime, temperature and
in the presence of THF were structure II hydrates [33]. Although pressure data were collected by the data acquisition unit and
CO2 removal from the CO2/CH4 gas mixture employing the logged into the computer every 10 s. When the temperature
hydrate based gas separation (HBGS) process have been studied restored to about 274.15 K and the pressure drop in the vessel
intensively, the kinetics of CO2 separation from the CO2/CH4 gas was not observed for at least 2 h, it was considered the end of
mixture in the absence/presence of promoters is not well known. the experiment. The composition of gas mixtures remaining in
In this work, the impact of driving force (overpressure) on the vessel and recovered from the decomposed hydrates was ana-
hydrate growth and CO2 removal from the CO2/CH4 gas mixture lyzed by the gas chromatograph (GC). To obtain reproducible
was investigated in liquid water and in the presence of THF. results two sets of experiments were carried out under the same
The CO2 selectivity of gas hydrates formed from the CO2/CH4 experimental condition following the above procedure. It should
mixture was also evaluated in liquid water and in the presence be noted that one set of experiment includes a ‘fresh’ run and a
of THF and SDS. The kinetic data reported in this study will ‘memory’ run. The ‘memory’ run was started 2 h after hydrate dis-
improve our understanding of how to increase the efficiency of sociation and the solution used in the ‘memory’ run has experi-
CO2 separation from the CO2/CH4 gas mixture by gas hydrate enced hydrate formation in the ‘fresh’ run.
formation.
2.4. Calculation of the amount of gas mixture consumed
2. Experimental methods
Based on the gas compositions measured at the beginning and
2.1. Materials the end of the experiments, the number of moles of the gas mix-
ture incorporated into the hydrate phase (gas uptake) was calcu-
The CO2/CH4 gas mixture containing 40 mol% CO2 and 60 mol% lated by the following equation:
CH4 was supplied by Chongqing Rising Gas with a composition
uncertainty of ±0.05 mol%, which was selected as a model shale


PV PV
gas recovered by supercritical CO2 injection. Tetrahydrofuran DnH ¼ ng;0  ng;t ¼  ð1Þ
zRT 0 zRT t
(THF) and the surfactant sodium dodecyl sulfate (SDS) with a cer-
 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141 135

where ng is the number of moles of the gas mixture in the vessel at nHCO2
time 0 and time t, P is the pressure in the reactor, T is the temper- R¼  100% ð6Þ
nfeed
CO2
ature of the gas phase, V is the volume of the gas phase, and z is the
gas compressibility calculated by the Pitzer correlation for the sec- where nfeed H
CO2 is the moles of CO2 supplied to the reactor and nCO2 is
ond Virial coefficient [34] the moles of CO2 incorporated into the hydrate phase at the end
Pr Pr of the experiments.
z ¼ 1 þ B0 þ xB1 ð2Þ Separation factor (S) represents the hydrate selectivity of CO2
Tr Tr
from the gas mixture (CO2/CH4) and was calculated by the follow-
where the equations of Abbott were used for B0 and B1. ing equation.
The gas consumption (DnH) was normalized by considering the
amount of water used in each experiment. The normalized gas nHCO2  ngas
CH4
S¼ ð7Þ
uptake was determined by Eq. (3). ngas
CO2  nCH4
H

DnH where ngas gas

DnH;norm ¼ ð3Þ CO2 and nCH4 are the numbers of moles of CO2 and CH4 in the
nH2 O
gas phase at the end of the experiments. nHCH4 is the number of moles
where DnH is number of moles of the gas mixture consumed for of CH4 incorporated into the hydrate crystals at the end of the
hydrate formation at the end of the experiment and nH2O is number experiments.
of moles of water used in the experiment.
The rate of hydrate growth was customarily considered to be 3. Results and discussion
the rate of gas uptake and was calculated using the forward differ-
ence method as follows Table 1 shows the experimental conditions and results for CO2


dDnH;norm DnH;norm;tþDt  DnH;norm;t removal from the CO2/CH4 gas mixture using hydrate formation
¼ ; Dt ¼ 10 s ð4Þ in three different systems (liquid water, THF solution, and THF/
dt t Dt
SDS solution). The experimental temperature was fixed at
The average value of these rates (Rav) was calculated every 277.15 K. Overpressure (DP) was employed as driving force for
30 min as follows hydrate formation and was defined as the difference between the
2dDn  
dDnH;norm
 
dDnH;norm
 3 initial experimental pressure (Pexp) and the phase equilibrium
H;norm
dt
þ dt
þ  þ dt pressure (Peq) at the experimental temperature (DP = Pexp  Peq).
Rav ¼ 4 1 2 m5
; m ¼ 180
m The concentration of THF and SDS was fixed at 1.0 mol% and
500 ppm, respectively. The phase equilibrium pressure (Peq) of
ð5Þ gas hydrates formed from the 40 mol% CO2/CH4 gas mixture was
2.69 MPa at 277.15 K in liquid water [12], and was 0.3 MPa at
277.15 K in the presence of 1.0 mol% THF [33]. The results reported
2.5. Calculation of CO2 recovery and separation factor
in Table 1 include induction time (tind), final gas uptake (DnH, norm),
CO2 recovery or split fraction (R) indicates the recovery effi- CO2 content of the gas mixture (xgas
CO2 ) remaining in the vessel at the

ciency of CO2 from the total CO2 supplied to the reactor and was end of the experiment, CO2 content in the hydrate phase (xH
CO2 ), CO2
calculated by the following equation [35]. recovery (R), and separation factor (S).

Table 1
Experimental conditions and results for CO2 removal from the CO2/CH4 mixture by hydrate formation.a

Exp. no. System Exp. state T (K) DPb (MPa) tind (min) Final gas uptake xgas
CO2 (mol%) xH
CO2 (mol%)
R (%) S
(mol of gas/mol of water)
1 CO2/CH4/H2O Fresh 277.15 2.5 15.6 0.0187 28.3 63.1 48.7 11.2
2 Memory 6.7 0.0217 27.1 58.1 53.6 9.5
3 Fresh 5.2 0.0224 27.5 60.7 53.4 8.0
4 Memory 5.0 0.0233 28.0 63.3 53.4 6.5
5 Fresh 4.0 5.0 0.0187 38.0 47.3 23.8 1.9
6 Memory 0.9 0.0201 38.1 53.0 24.7 1.7
7 Fresh 2.2 0.0168 38.2 53.5 21.4 1.8
8 Memory 0.3 0.0193 39.1 48.7 22.1 1.4
9 CO2/CH4/THF/H2O Fresh 277.15 2.5 1.5 0.0129 29.8 52.1 50.7 4.0
10 Memory 1.1 0.0125 30.0 54.1 48.4 3.5
11 Fresh 2.0 0.0131 29.1 53.8 53.5 4.0
12 Memory 0.5 0.0129 32.9 55.6 46.8 2.6
13 Fresh 4.0 0.2 0.0112 37.4 43.0 24.4 2.0
14 Memory 0.2 0.0113 37.7 44.4 23.8 1.8
15 Fresh 0.3 0.0117 37.3 41.6 25.3 2.0
16 Memory 0.4 0.0118 37.5 44.2 25.1 1.9
17 CO2/CH4/THF/SDS/H2O Fresh 277.15 2.5 0.7 0.0115 28.5 60.0 51.4 5.5
18 Memory 0.5 0.0097 27.8 69.4 49.3 9.4
19 Fresh 3.1 0.0118 28.8 64.8 51.5 5.0
20 Memory 2.4 0.0108 29.8 73.4 47.7 5.0
a
The composition of the gas mixture was 40 mol% CO2 and 60 mol% CH4. Constant conditions: liquid volume = 140 cm3, THF concentration = 1 mol%, and SDS
concentration = 500 ppm.
b
Driving force DP = Pexp  Peq. Peq at 277.15 K was 0.3 MPa in the presence of 1 mol% THF.
136 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141

280 0.025 5.4

(a)
5.2

Gas uptake (mol of gas/mol of water)

Tliquid experiment 3 (liquid water) experiment 3
279 0.020
Gas uptake 5.0
Pressure at 277.15 K
60 min 4.8
Temperature (K)

Pressure (MPa)
278 tind= 5.2 min 0.015
4.6

4.4
277 0.010
4.2

276 0.005 4.0

3.8

275 0.000 3.6

0 10 20 30 40 50 60 70 80 90 100 0 100 200 300 400 500 600 700 800 900
Time (min) Time (min)

Fig. 2. Pressure evolution for gas hydrate formation from the CO2/CH4 gas mixture
280 0.025
in liquid water with the initial pressure fixed at 5.2 MPa.
(b)
Tliquid
279 Gas uptake 0.020 Gas uptake (mol of gas/mol of water)
Fig. 2 shows the pressure evolution for the experiment per-
experiment 3 (liquid water) formed in liquid water (experiment 3 in Fig. 1). It was captured
Temperature (K)

278 0.015 during the process of hydrate formation from the CO2/CH4 gas mix-
ture. As seen in Fig. 2, pressure in the vessel was 5.2 MPa at the
beginning of the experiment and decreased to 3.8 MPa at the end
277 0.010
of the experiment. This was caused due to gas dissolution in the
liquid phase and gas consumption for the subsequent hydrate
60 min growth. The composition of the gas mixture remaining in the ves-
276 0.005
hydrate nucleation sel was measured at the end of the experiment. It was composed of
tind=5.2 min 27.5 mol% CO2 and 72.5 mol% CH4 (Table 1), indicating that CO2
275 0.000 molecules were preferentially incorporated in the hydrate phase
0 100 200 300 400 500 600 700 800 900
as compared to the initial CO2 content in the gas mixture (30 mol
Time (min)
%). Note that the hydrate phase equilibrium pressure (Peq) for the
Fig. 1. Temperature and pressure profiles for hydrate formation from the CO2/CH4 27.5 mol% CO2/CH4 gas mixture was 2.95 MPa at 277.15 K, so the
gas mixture in liquid water at 277.15 K. (a) Temperature and pressure curves for driving force (overpressure) at the end of the experiment was
hydrate formation during the first 100 min of the experiment; and (b) temperature 0.85 MPa (DP = Pexp  Peq = 3.8–2.95 = 0.85 MPa). This means that
and pressure curves for hydrate formation between 0 and 900 min. hydrate growth was interrupted before the phase equilibrium state
was reached. This is probably attributed to the arrangement of the
stirred tank reactor where gas hydrates primarily nucleated at the
gas/bulk liquid interface at the beginning of hydrate formation. The
3.1. Hydrate formation in the CO2/CH4/H2O system mass transfer of gas molecules to the liquid phase for hydrate
growth is hindered with the accumulation of gas hydrates at the
Fig. 1 shows a typical temperature and gas uptake profile for gas/liquid interface, resulting in a significant decrease in the rate
hydrate formation in liquid water using the CO2/CH4 gas mixture of hydrate growth. This explanation agreed with the evolution of
(experiment 3 in Table 1). Fig. 1a shows the enlarged temperature gas uptake curve shown in Fig. 1b as well.
and pressure plot for the first 100 min of the experiment, while
Fig. 1b shows the temperature and gas uptake curves for the entire 3.2. Impact of driving force on CO2 removal in the CO2/CH4/H2O system
experiment. As seen in Fig. 1a, a sudden temperature increase in
the liquid phase was observed at 5.2 min. This was identified as The experiments for CO2 recovery from the CO2/CH4 gas mix-
the onset of hydrate nucleation due to the exothermic nature of ture were performed at two driving forces (DP = 2.5 and 4.0 MPa)
hydrate formation. Gas uptake was observed to increase rapidly in the CO2/CH4/H2O system (experiments 1–8 in Table 1). The ini-
during this short period, indicating the fast growth of gas hydrates tial pressures corresponding to DP = 2.5 and 4.0 MPa were 5.2 and
which was confirmed by the visual observation through the view- 6.7 MPa at 277.15 K. Fig. 3 compared the gas uptake that was
ing windows on the vessel. The temperature increase of liquid obtained at DP = 2.5 (experiments 1–4) and 4.0 MPa (experiments
phase detected at 60 min also indicates the occurrence of hydrate 5–8), respectively. Time zero in the plot corresponds to the hydrate
formation. Then the temperature was observed to decrease and nucleation point. As seen in the figure, for the first 400 min the gas
restored to the set value (277.15 K) at approximately 350 min uptake obtained at DP = 4.0 MPa was larger than that obtained at
(Fig. 1b). It should be noted that the gas uptake at 350 min DP = 2.5 MPa, indicating that the growth of gas hydrates was
(0.0201 mol of gas/mol of water) reached 90% of the total gas enhanced while the driving force was increased from 2.5 MPa to
uptake at the end of the experiment (0.0224 mol of gas/mol of 4.0 MPa. It is interesting to note that 400 min later the gas uptake
wate), indicating the growth of gas hydrates almost completed at DP = 2.5 MPa was observed to surpass the gas uptake at
during this period. The gas uptake curve was observed to plateau DP = 4.0 MPa. This can be explained by the fact that higher driving
after 350 min. This means that the rate of hydrate growth was sig- force resulted in a larger amount of gas hydrates formed in the
nificantly reduced while gas hydrates accumulated in the vessel. bulk liquid during the first 400 min. But the transport of gas mole-
The process of hydrate formation nearly stopped when the gas cules through the bulk hydrates for further growth of gas hydrates
uptake reached a plateau at 900 min. at the higher driving force was more difficult and thus a significant
 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141 137

100 12
0.020 ΔP = 2.5 MPa
Gas uptake (mole of gas/mole of water)

ΔP = 4.0 MPa CO2 recovery

10
80 separation factor
0.016

CO2 Recovery, R (%)

8

separation factor (S )
60
0.012
6

0.008 40
4

0.004 20
2

0.000
0 100 200 300 400 500 600 700 800 900 0 0
2.5 4.0
Time (min)
Driving force (MPa)
Fig. 3. Impact of driving force on the gas consumption for hydrate growth from the
CO2/CH4 gas mixture in liquid water at 277.15 K. Time zero corresponds to the Fig. 5. Impact of driving force on CO2 recovery and separation factor from the CO2/
induction time. CH4 gas mixture in liquid water at 277.15 K.

decrease in the rate of hydrate formation was observed (Fig. 4). It the kinetics of hydrate formation in the presence of THF for CO2
can be seen in Fig. 4 that the average rate of hydrate formation removal from the CO2/CH4 gas mixture has not been well under-
at DP = 4.0 MPa was 2.5 times larger than that at DP = 2.5 MPa dur- stood. The experiments of hydrate formation with the CO2/CH4
ing the first 15 min, but decreased dramatically and became smal- gas mixture were performed in the presence of 1.0 mol% THF to
ler than that at DP = 2.5 MPa 30 min later. elucidate the impact of driving force (overpressure) on CO2
Fig. 5 shows the impact of driving force on CO2 recovery (R) and removal from the CO2/CH4 gas mixture in the presence of THF
separation factor (S) for the experiments carried out at DP = 2.5 (experiments 9–16 in Table 1). Two driving forces of DP = 2.5
and 4.0 MPa in the CO2/CH4/H2O system. As seen in the figure, (Pexp = 2.8 MPa, experiments 9–12) and DP = 4.0 MPa (Pexp = 4.3 -
the average value of CO2 recovery (R) at DP = 2.5 MPa was 52.3%, MPa, experiments 13–16) at 277.15 K were used.
which was much higher than that obtained at DP = 4.0 MPa Fig. 6 shows the comparison of gas uptake obtained at DP = 2.5
(23%). The CO2 separation factor (S) obtained at DP = 2.5 MPa (experiments 9–12) and 4.0 MPa (experiments 13–16) in the CO2/
(8.8) was also higher than that obtained at DP = 4.0 MPa (1.7). CH4/THF/H2O system. Time zero in the plot corresponds to the
This result indicates that the hydrate selectivity for the separation hydrate nucleation point. Similar to that observed in the CO2/
of CO2 from the CO2/CH4 gas mixture decreased with the increase CH4/H2O system (Fig. 3), hydrate growth in the CO2/CH4/THF/H2O
of driving force. This is because the competition between CH4 system was promoted as the driving force was increased from
molecules and CO2 molecules to occupy the hydrate cavities (512 2.5 MPa to 4.0 MPa. For example, the gas uptake at DP = 4.0 MPa
and 51262 in structure I hydrates) became stronger with the increased faster than that at DP = 2.5 MPa and was larger than that
increase of driving force (overpressure in this work) although obtained at DP = 2.5 MPa during the first 280 min. It is noted that
CO2 molecules were preferred in the enclathration of gas hydrates 280 min later the gas uptake at DP = 2.5 MPa increased higher than
over CH4 molecules. that obtained at DP = 4.0 MPa, and the increasing rate at
DP = 2.5 MPa became larger. The explanation for this phenomenon
was the same as that presented for the result in Fig. 3. The mass
3.3. Impact of driving force on CO2 removal in the CO2/CH4/THF/H2O
transfer of gas molecules through the formed bulk hydrates for
system
the further growth of gas hydrates was hindered under the higher
driving force.
It was reported in our previous work that THF can be employed
as a promoter to reduce the phase equilibrium pressures of gas
hydrates formed with the CO2/CH4 gas mixture [33]. However,
0.014
0.025
ΔP = 2.5 MPa
Gas uptake (mol of gas/mol of water)

0.012 ΔP = 4.0 MPa

Average rate of hydrate formation

0.020 Δ P = 2.5 MPa

0.010
Δ P = 4.0 MPa
(mol of gas/mol of water/h)

0.008
0.015

0.006
0.010
0.004

0.005 0.002

0.000
0.000 0 100 200 300 400 500 600
0 1 2 3 4 5 6 7 8 9 10 11 Time (min)
Time (h)
Fig. 6. Impact of driving force on the gas consumption for hydrate growth from the
Fig. 4. Impact of driving force on the rate of hydrate formation that was obtained in CO2/CH4 gas mixture in the presence of 1.0 mol% THF at 277.15 K. Time zero
liquid water at 277.15 K. corresponds to the induction time.
138 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141

100 12 280 0.010

Tliquid experiment 18

Gas uptake (mol of gas/mol of water)

CO2 recovery
10 gas uptake 1mol% THF + 500 ppm SDS
80 separation factor 279 0.008
CO2 Recovery, R (%)

Temperature (K)
separation factor (S )
8
multiple hydrate nucleation
60 278 0.006

40 277 0.004
4
hydrate nucleation
tind=0.5 min
20 276 0.002
2

0 0 275 0.000
2.5 4.0 0 50 100 150 200 250 300 350
Driving force (MPa) Time (min)

Fig. 7. Impact of driving force on CO2 recovery (R) and separation factor (S) Fig. 8. Temperature and pressure profiles for hydrate formation from the CO2/CH4
obtained in the CO2/CH4/THF/H2O system. The experiments were carried out in the gas mixture in the presence of 1.0 mol% THF and 500 ppm SDS and at 277.15 K.
presence of 1.0 mol% THF and at 277.15 K.

Fig. 7 shows the impact of driving force on CO2 recovery (R) and 280 0.016
separation factor (S) obtained in the CO2/CH4/THF/H2O system. As Tliquid experiment 9 (1.0 mol% THF) 0.014

Gas uptake (mol of gas/mol of water)

seen, CO2 recovery (R) decreased substantially with the increase of gas uptake
279
driving force. For example, CO2 recovery (R) obtained at 0.012
DP = 2.5 MPa was 49.9% but decreased to 24.7% at DP = 4.0 MPa.
Temperature (K)

The separation factor (S) was 3.5 at DP = 2.5 MPa and decreased 278
0.010
by 54% at DP = 4.0 MPa. This indicates that more CH4 molecules
0.008
were trapped in the hydrate phase as the driving force was
increased from 2.5 MPa to 4.0 MPa, causing the decrease of CO2 277
0.006
recovery and separation factor from the CO2/CH4 gas mixture.
The impact of driving force on CO2 selectivity from the CO2/CH4 multiple nucleation 0.004
276
gas mixture by hydrate formation in the presence of THF shows t ind =1.5 min
0.002
the same trend as that observed in the CO2/CH4/H2O system
(Fig. 5). Therefore, the results obtained in the CO2/CH4/H2O system 275 0.000
0 100 200 300 400 500 600
and the CO2/CH4/THF/H2O system elucidated that the hydrate
Time (min)
selectivity for CO2 removal from the CO2/CH4 gas mixture can be
improved at low driving force (low overpressure). Low driving Fig. 9. Temperature and pressure profiles for hydrate formation from the CO2/CH4
force can be realized through using low operation pressure. This gas mixture in the presence of 1.0 mol% THF and at 277.15 K.
is preferred in industry applications because gas compression costs
will be reduced and a higher CO2 separation efficiency will be
obtained. the presence of SDS (350 min in Fig. 8) was significantly shortened
as compared to that in the absence of SDS (600 min in Fig. 9). This
result indicates that the kinetics of hydrate formation with the
3.4. Impact of surfactant SDS on CO2 removal from the CO2/CH4 gas CO2/CH4 gas mixture was enhanced by adding SDS in the THF
mixture solutions.
On the other hand, the average value of CO2 recovery (R)
To improve the kinetics of hydrate formation in the presence of obtained in the presence of SDS was 50.0% (Table 1, experiments
THF for CO2 removal from the CO2/CH4 gas mixture, the surfactant 17–20) which was almost the same as that obtained in the absence
SDS with a low dosage (500 ppm) was employed. The experiments of SDS (49.9%, experiments 9–12 in Table 1). However, as seen in
(experiments 17–20 in Table 1) were carried out at the low driving Fig. 10, CO2 separation factor (S) obtained in the presence of SDS
force (DP = 2.5 MPa) which was found to be an optimum experi- (6.2) was greatly increased as compared with that obtained in
mental condition compared with the high driving force the absence of SDS (3.5), and CO2 content in the hydrate phase
(DP = 4.0 MPa). obtained in the presence of SDS (66.9 mol%) was also higher than
Fig. 8 shows a typical temperature and gas uptake profile for that obtained in the absence of SDS (53.9 mol%). This is probably
hydrate formation with the CO2/CH4 gas mixture in the presence because CO2 solubility in THF solutions was increased by adding
of THF and SDS (experiment 18 in Table 1). The occurrence of a 500 ppm SDS into the THF solutions and thus more CO2 molecules
sudden temperature spike in the liquid phase was observed at were captured in the hydrate phase. Fig. 11 shows the CO2 and CH4
0.5 min. This was identified as the onset of hydrate nucleation solubility in different solutions (liquid water, THF solution) which
and was confirmed by visual observation through the viewing win- were measured at room temperature in the presence and absence
dows on the vessel. Compared to the experiments carried out at of SDS. Obviously, the measured CO2 solubility in pure water
the same experimental conditions without using SDS (Fig. 9), more agreed with the values predicted by Diamond et al.[36] Note that
temperature peaks can be seen after the hydrate nucleation point. CO2 solubility in pure water, THF solution, and THF/SDS solution
This is probably related to the event of multiple hydrate nucle- were very close at low pressures (1.12 and 1.8 MPa), but CO2 solu-
ation. It also can be seen in the figure that the intensity of temper- bility in the THF/SDS solution became higher while the pressure
ature spikes obtained in the presence of SDS was higher than that was increased to 2.8 MPa. This indicates that CO2 solubility in
obtained in the absence of SDS. The process of hydrate growth in THF solutions was promoted in the presence of 500 ppm SDS.
 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141 139

100 12 0.024
liquid water
1mol% THF

Gas uptake (mol of gas/mol of water)

CO2 content in hydrate
10 0.020
CO2 content in hydrate (mol%)

80
1mol% THF + 500ppm SDS
separation factor

separation factor, S
8 0.016
60

6 0.012

40
4 0.008

20
2 0.004 ΔP = 2.5 MPa, T = 277.15 K

0 0 0.000
THF solution THF/SDS solution 0 100 200 300 400 500 600 700 800 900
Time (min)
Fig. 10. CO2 content in the hydrates and CO2 separation factor obtained in THF and
THF/SDS solutions at 277.15 K. The initial experimental pressure was set at 2.8 MPa. Fig. 12. Comparison of the gas uptake obtained in CO2/CH4/H2O, CO2/CH4/THF/H2O,
1.0 mol% THF and 500 ppm SDS were employed. and CO2/CH4/THF/SDS/H2O systems under the driving force of 2.5 MPa. The
experimental temperature was fixed at 277.15 K.

2.0 1.0
CO2, liquid water (this work)
1.8 CO2, liquid water (Diamond et al.) formation from the CO2/CH4 gas mixture was enhanced while THF
1.6 CO2, 1.0 mol% THF 0.8
or THF/SDS was added into the liquid water. Compared to hydrate
CO2, 1.0 mol% THF + 500 ppm SDS formation in THF solutions, the enhancement of hydrate formation
CO2 solubility (mol%)

1.4
CH4 solubility (mol%)

CH4, 1.0 mol% THF in THF/SDS solutions is probably due to the significant reduction of
1.2 CH4, 1.0 mol% THF + 500 ppm SDS 0.6 liquid surface tension in the presence of SDS.
1.0 Fig. 13 shows the comparison of CO2 recovery (R) and separa-
tion factor (S) obtained in liquid water, THF solution, and THF/
0.8 0.4
SDS solution at a fixed driving force and experimental temperature
0.6 (DP = 2.5 MPa and T = 277.15 K). It can be seen that CO2 recovery
0.4 0.2 (R) obtained in the CO2/CH4/H2O system was 52.3%, which was
higher than that obtained in the CO2/CH4/THF/H2O (49.9%) and
0.2
CO2/CH4/THF/SDS/H2O systems (50.0%). The CO2 separation factor
0.0 0.0
0.8 1.2 1.6 2.0 2.4 2.8 3.2 (S) of the CO2/CH4/H2O system (8.8) was also larger than that
Pressure (MPa) obtained in the CO2/CH4/THF/H2O (3.5) and CO2/CH4/THF/SDS/
H2O systems (6.2). This result indicates that although the rate of
Fig. 11. CO2 solubility in liquid water, THF and THF/SDS solutions. CH4 solubility in hydrate growth was promoted in the presence of THF or THF/SDS
THF and THF/SDS solutions. The measurements were carried out at room temper- (Fig. 12), the selectivity of CO2 from the 40 mol% CO2/CH4 gas mix-
ature and different pressures.
ture was reduced in the THF and THF/SDS solutions as compared to
that obtained in liquid water. This means that CO2 recovery from
Interestingly, the CH4 solubility obtained in THF/SDS solutions was the 40 mol% CO2/CH4 gas mixture using hydrate formation was
slightly lower than that obtained in the absence of SDS at 1.8 MPa, compromised due to the presence of THF. One reason is that the
and decreased largely as compared to that obtained in the absence large cavities of the structure II hydrates formed from the CO2/
of SDS while the pressure was increased to 2.8 MPa. This indicates CH4 gas mixture were assumed to encage CO2 molecules but were
that the presence of SDS reduced the CH4 solubility in THF solu- partially occupied by THF molecules. Another reason is that the
tions. So the reason why CO2 selectivity obtained in the presence
of SDS was higher is probably because the presence of SDS
increased the CO2 solubility in THF solutions but reduced the CH4
100 12
solubility in THF solutions at the same time.
ΔP = 2.5 MPa,T = 277.15 K
CO2 recovery 10
80
3.5. Comparison of CO2 removal using hydrate formation in different separation factor
CO2 Recovery, R (%)

systems 8
separation factor, S

60
Fig. 12 shows the comparison of gas uptake obtained in liquid 6
water, THF solution, and THF/SDS solution at a fixed driving force
40
(DP = 2.5 MPa). Note that the gas uptake shown in the plot was
4
an average value of the experimental data which were obtained
at the same experimental condition in each system. Time zero in 20
2
the figure corresponds to hydrate nucleation point. It can be seen
that the final gas uptake obtained in liquid water was 0.0190 mol
of gas/mol of water, which was 2.3 times larger than that obtained 0 0
H2O THF/H2O THF/SDS/H2O
in THF solutions and 1.6 times larger than that obtained in THF/SDS
solutions. On the other hand, the process of hydrate formation in Fig. 13. Comparison of CO2 recovery and separation factor obtained in CO2/CH4/
THF solutions and THF/SDS solutions was shortened as compared H2O, CO2/CH4/THF/H2O, and CO2/CH4/THF/SDS/H2O systems under the driving force
with that in liquid water. This indicates that the kinetics of hydrate of 2.5 MPa. The experimental temperature was fixed at 277.15 K.
140 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141

Flue Gas CO2 Capture CO2 Compression SC-CO2 Shale gas

System process formations

CO2
Recovered
Hydrate shale gas
decomposition CO2/CH4

CO2 hydrate
Water
Natural gas HBGS process
CH4-rich stream
pipelines

Fig. 14. Flow diagram of using the hydrate based gas separation (HBGS) method for CO2 removal in the engineering process.

phase equilibrium conditions for the CO2/CH4 mixed hydrates in Funds for the Central Universities (No. CDJZR12248801, No.
the presence of THF were significantly reduced as compared to that 2015CDJXY145502), and the Chongqing Graduate Student
in pure water. As a result, CH4 molecules occupied the hydrated Research Innovation Project (2015) is greatly appreciated.
cavities more easily in the presence of THF and hence the hydrate
selectivity for CO2 removal from the CO2/CH4 gas mixture was
References
reduced. Therefore, enhancing the kinetics of hydrate growth with-
out deceasing CO2 selectivity (using porous media such as silica [1] Weijermars R. Economic appraisal of shale gas plays in Continental Europe.
sand, silica gel, and activated carbon) is the objective of optimizing Appl Energy 2013;106:100–15.
the hydrate-based separation process for CO2 removal from the [2] Wakamatsu H, Aruga K. The impact of the shale gas revolution on the U.S. and
Japanese natural gas markets. Energy Policy 2013;62:1002–9.
CO2/CH4 gas mixture in future. [3] Armor JN. Emerging importance of shale gas to both the energy & chemicals
Fig. 14 shows the flow diagram of configuring the hydrate based landscape. J Energy Chem 2013;22:21–6.
gas separation (HBGS) method for CO2 removal in the engineering [4] George DL, Bowles EB. Shale gas measurement and associated issues. Pipe Gas J
2011. <http://www.pipelineandgasjournal.com/shale-gas-measurement-and-
process. As seen in the figure, the greenhouse gas CO2 is obtained
associated-issues?page=show>.
from flue gas (CO2/N2) which is emitted from power plants, and [5] Li D, Zhang L, Lu D. Effect of distinguishing apparent permeability on flowing
then is compressed to supercritical state (SC-CO2) before being gas composition, composition change and composition derivative in tight- and
shale-gas reservoir. J Petrol Sci Eng 2015;128:107–14.
injected into the shale gas formations. When shale gas is produced
[6] Martin M, Grossmann IE. Optimal use of hybrid feedstock, switchgrass and
by the injected SC-CO2, CO2 in the recovered shale gas (CO2/CH4) shale gas for the simultaneous production of hydrogen and liquid fuels. Energy
will be removed through the HBGS process. Thus, the recovered 2013;55:378–91.
CH4-rich stream can be transferred to natural gas pipelines for fur- [7] Stamford L, Azapagic A. Life cycle environmental impacts of UK shale gas. Appl
Energy 2014;134:506–18.
ther use while CO2 collected from the HBGS process can be used [8] Engelder T, Cathles LM, Bryndzia LT. The fate of residual treatment water in gas
again for the production of shale gas. In this way, the production shale. J Unconventional Oil Gas Resour 2014;7:33–48.
of shale gas is enhanced and the geological storage of CO2 is [9] Godec M, Koperna G, Petrusak R, Oudinot A. Potential for enhanced gas
recovery and CO2 storage in the Marcellus Shale in the Eastern United States.
achieved. Int J Coal Geol 2013;118:95–104.
[10] Middleton RS, Carey JW, Currier RP, Hyman JD, Kang Q, Karra S, et al. Shale gas
and non-aqueous fracturing fluids: opportunities and challenges for
4. Conclusions
supercritical CO2. Appl Energy 2015;147:500–9.
[11] Davidson DW. Clathrate hydrates. In: Frank F, editor. Water: a comprehensive
In this work, the kinetics of hydrate formation for CO2 removal treatise. New York: Plenum Press; 1973.
[12] Sloan ED, Koh CA. Clathrate hydrates of natural gases. 3rd ed. Boca Raton,
from a CO2/CH4 gas mixture (40 mol% CO2 and 60 mol% CH4) was
FL: CRC Press; 2008.
evaluated in three systems such as liquid water, THF solution, [13] Englezos P. Clathrate hydrates. Ind Eng Chem Res 1993;32(7):1251–74.
and THF/SDS solution. The impact of driving force (overpressure) [14] Kim NJ, Hwan Lee J, Cho YS, Chun W. Formation enhancement of methane
on hydrate growth and CO2 selectivity were studied in liquid water hydrate for natural gas transport and storage. Energy 2010;35(6):2717–22.
[15] Kondo W, Ohtsuka K, Ohmura R, Takeya S, Mori YH. Clathrate-hydrate
and in THF solutions, respectively. It was found that the final gas formation from a hydrocarbon gas mixture: compositional evolution of formed
uptake, CO2 recovery, and the separation factor decreased with hydrate during an isobaric semi-batch hydrate-forming operation. Appl
the increase of driving force, indicating that the competition Energy 2014;113:864–71.
[16] Lee JD, Park KN, Hong SY, Lee JW, Kang KC, Lee YC, et al. A new apparatus for
between CH4 and CO2 molecules for the occupancy of hydrate cav- seawater desalination by gas hydrate process and removal characteristics of
ities became stronger at high driving force. The kinetics of hydrate dissolved minerals (Na+, Mg2+, Ca2+, K+, B3+). Desalination 2011;274:91–6.
growth with the CO2/CH4 gas mixture was improved by adding THF [17] Kang KC, Linga P, Park KN, Choi SJ, Lee JD. Seawater desalination by gas hydrate
process and removal characteristics of dissolved ions (Na+, K+, Mg2+, Ca2+, B3+,
and SDS into the liquid water, but the hydrate selectivity for CO2 Cl1, SO42). Desalination 2014;353:84–90.
removal from the CO2/CH4 gas mixture was compromised as com- [18] Xie YM, Li G, Liu DP, Liu N, Qi YX, Liang DQ, et al. Experimental study on a small
pared to that obtained in liquid water. The CO2 recovery (52%) and scale of gas hydrate cold storage apparatus. Appl Energy 2010;87(11):3340–6.
[19] Shi XJ, Zhang P. A comparative study of different methods for the generation of
separation factor (8.8) obtained in liquid water at a low driving
tetra-n-butyl ammonium bromide clathrate hydrate slurry in a cold storage
force were much higher than those obtained in THF and THF/SDS air-conditioning system. Appl Energy 2014;112:1393–402.
solutions. [20] Zhong DL, Englezos P. Methane separation from coal mine methane gas by
tetra-n-butyl ammonium bromide semiclathrate hydrate formation. Energy
Fuels 2012;26(4):2098–106.
Acknowledgements [21] Liu H, Mu L, Wang B, Liu B, Wang J, Zhang X, et al. Separation of ethylene from
refinery dry gas via forming hydrate in w/o dispersion system. Sep Purif
The financial support from the Ministry of Education Innovation Technol 2013;116:342–50.
[22] Zhong DL, Ding K, Lu YY, Yan J, Zhao WL. Methane recovery from coal mine gas
Research Team (IRT13043), the National Key Basic Research Pro- using hydrate formation in water-in-oil emulsions. Appl Energy 2014. http://
gram of China (No. 2014CB239206), the Fundamental Research dx.doi.org/10.1016/j.apenergy.2014.11.010.
 D.-L. Zhong et al. / Applied Energy 158 (2015) 133–141 141

[23] Li XS, Xia ZM, Chen ZY, Yan KF, Li G, Wu HJ. Gas hydrate formation process for [30] van Denderen M, Ineke E, Golombok M. CO2 removal from contaminated
capture of carbon dioxide from fuel gas mixture. Ind Eng Chem Res 2010;49 natural gas mixtures by hydrate formation. Ind Eng Chem Res 2009;48
(22):11614–9. (12):5802–7.
[24] Li XS, Zhan H, Xu CG, Zeng ZY, Lv QN, Yan KF. Effects of tetrabutyl- [31] Fan S, Li Q, Nie J, Lang X, Wen Y, Wang Y. Semiclathrate hydrate phase
(ammonium/phosphonium) salts on clathrate hydrate capture of CO2 from equilibrium for CO2/CH4 gas mixtures in the presence of tetrabutylammonium
simulated flue gas. Energy Fuels 2012;26(4):2518–27. halide (Bromide, Chloride, or Fluoride). J Chem Eng Data 2013;58
[25] Xu CG, Zhang SH, Cai J, Chen ZY, Li XS. CO2 (carbon dioxide) separation from (11):3137–41.
CO2–H2 (hydrogen) gas mixtures by gas hydrates in TBAB (tetra-n-butyl [32] Ricaurte M, Dicharry C, Broseta D, Renaud X, Torre JP. CO2 removal from a CO2–
ammonium bromide) solution and Raman spectroscopic analysis. Energy CH4 gas mixture by clathrate hydrate formation using THF and SDS as water-
2013;59:719–25. soluble hydrate promoters. Ind Eng Chem Res 2013;52(2):899–910.
[26] Xu CG, Chen ZY, Cai J, Li XS. Study on pilot-scale CO2 separation from flue gas [33] Zhong DL, Li Z, Lu YY, Sun DJ. Phase equilibrium data of gas hydrates formed
by the hydrate method. Energy Fuels 2014;28(2):1242–8. from a CO2 + CH4 gas mixture in the presence of tetrahydrofuran. J Chem Eng
[27] Yang M, Song Y, Jiang L, Zhao Y, Ruan X, Zhang Y, et al. Hydrate-based Data 2014;59(12):4110–7.
technology for CO2 capture from fossil fuel power plants. Appl Energy [34] Smith JM, Van ness HC, Abbott MW. Introduction to chemical engineering
2014;116:26–40. thermodynamics. New York: Mcgraw-Hill, Inc.; 2001.
[28] Babu P, Ho CY, Kumar R, Linga P. Enhanced kinetics for the clathrate process in [35] Linga P, Kumar R, Englezos P. The clathrate hydrate process for post and pre-
a fixed bed reactor in the presence of liquid promoters for pre-combustion combustion capture of carbon dioxide. J Hazard Mater 2007;149(3):625–9.
carbon dioxide capture. Energy 2014;70:664–73. [36] Diamond LW, Akinfiev NN. Solubility of CO2 in water from 5 to 100 °C and
[29] Uchida T, Ikeda IY, Takeya S, Kamata Y, Ohmura R, Nagao J, et al. Kinetics and from 0.1 to 100 MPa: evaluation of literature data and thermodynamic
stability of CH4–CO2 mixed gas hydrates during formation and long-term modelling. Fluid Phase Equilib 2003;208:265–90.
storage. ChemPhysChem 2005;6(4):646–54.