- Email: [email protected]
Molecular detection and genotyping of Toxoplasma gondii in free-ranging pigs in Northeastern China
Applied Surface Science 439 (2018) 792–800
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
Effect of water on methane adsorption on the kaolinite (0 0 1) surface based on molecular simulations Bin Zhang, Jianting Kang, Tianhe Kang ⇑ Institute of Mining Technology, Taiyuan University of Technology, Taiyuan 030024, PR China
a r t i c l e
i n f o
Article history: Received 6 June 2017 Revised 13 December 2017 Accepted 28 December 2017 Available online 12 January 2018 Keywords: Methane adsorption Kaolinite Water Molecular simulation
a b s t r a c t CH4 adsorption isotherms of kaolinite with moisture contents ranging from 0 to 5 wt% water, the effects of water on maximum adsorption capacity, kaolinite swelling, and radial distribution function were modelled by the implementing combined Monte Carlo (MC) and molecular dynamics (MD) simulations at 293.15 K (20 °C) and a pressure range of 1–20 MPa. The simulation results showed that the absolute adsorption of CH4 on both dry and moist kaolinite followed a Langmuir isotherm within the simulated pressure range, and both the adsorption capacity and the rate of CH4 adsorption decreased with the water content increases. The adsorption isosteric heats of CH4 on kaolinite decreased linearly with increasing water content, indicating that at higher water contents, the interaction energy between the CH4 and kaolinite was weaker. The interaction between kaolinite and water dominates and was the main contributing factor to kaolinite clay swelling. Water molecules were preferentially adsorbed onto oxygen and hydrogen atoms in kaolinite, while methane showed a tendency to be adsorbed only onto oxygen. The simulation results of our study provide the quantitative analysis of effect of water on CH4 adsorption capacity, adsorption rate, and interaction energy from a microscopic perspective. We hope that our study will contribute to the development of strategies for the further exploration of coal bed methane and shale gas. Ó 2018 Elsevier B.V. All rights reserved.
1. Introduction Clay minerals have large surface areas (approximately 800 m2 g1) and micropore to mesopore structures that can significantly affect the adsorption properties of porous media such as shale and coal [1,2]. Kaolinite is one of the most abundant components in clay minerals [3–5], and understanding the interaction between kaolinite and methane molecules is important for research in the fields of shale gas and coal bed methane. Some coal bed methane and shale gas reservoirs are water saturated [6,7], and the clay minerals that have similar silicoaluminate crystallographic layers with AlAO octahedra and SiAO tetrahedra are hydrophilic [8–10]. Water molecules can be adsorbed onto the surface of the clay mineral without difficulty, which can decrease the total methane sorption of clay minerals [11]. Hence, preloaded water can substantially decrease the total amount of methane adsorbed onto clay-rich rocks [12]. Ross et al. [13] found that the adsorbed capacities of water-saturated montmorillonite and illite were lower than under dry conditions. ⇑ Corresponding author at: Taiyuan University of Technology, No. 18 New Mine Road, West Yingze Street, Taiyuan City 030024, Shanxi Province, PR China. E-mail address: [email protected] (T. Kang). https://doi.org/10.1016/j.apsusc.2017.12.239 0169-4332/Ó 2018 Elsevier B.V. All rights reserved.
Ross and Bustin [12] found that at low pressures (6 MPa), the CH4 adsorption capacities of illite and montmorillonite were lower than that of kaolinite on a moisture-equilibrated basis but were significantly higher than that of kaolinite under dry conditions. Moreover, a few clay minerals, e.g., montmorillonite clay, are able to further enhance the interaction between clay molecules and methane molecules because of the cation-exchange capacity of montmorillonite clay [14]. Some molecular simulations have been carried out on the adsorption of gas in dry and moist clays. Billemont et al. [15] used grand canonical Monte Carlo (GCMC) simulations to consider the influence of water on methane sorption in porous carbons and observed that water molecules have a higher free energy barrier than methane molecules, which cannot displace the water molecules, and the preloaded water molecules notably decreased the adsorption capacity of methane. Jin et al. [16] also utilized GCMC simulations to research the influence of water on methane sorption in montmorillonite clay. However, to our knowledge, no computational and theoretical research has been carried out on the influence of water content on methane sorption in other clay minerals such as kaolinite and illite. Therefore, in this paper, we used the GCMC and molecular dynamics (MD) methods to research the effect of water on methane adsorption in kaolinite over a pres-
793
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
sure range of 1–20 MPa and with pre-adsorbed water contents of 0–5 wt% simulated at 293.15 K. Molecular simulations can reveal the absolute adsorption isotherms, interaction energy, isosteric heat of adsorption, kaolinite swelling, adsorbed phase density, and radial distribution functions between the surface of kaolinite and CH4. The objectives of this work were to better understand CH4/water/kaolinite clay interactions and to shed light on the influence of water on the adsorption capacity and kaolinite swelling of CH4 on kaolinite surfaces at the atomic level. The findings of this study are also expected to shed light on the details of coal bed methane and shale gas adsorption.
Fig. 2. The molecular model of water.
2. Simulation methods 2.1. Models We use methane to represent coal bed gas and shale gas because methane is the main component of coal bed gas and shale gas. Here, we mainly consider the case of methane in hydrated kaolinite. Methane is represented by an all-atom model [17,18]; the CAH bond length and the CAH bond angle were calculated to be 0.109 nm and 109°280 , respectively, as shown in Fig. 1. Water is represented by a simple point charge model [19]; the HAO bond length and the HAOAH bond angle were calculated to be 0.079 nm and 104°60 , respectively, as shown in Fig. 2. The kaolinite mineral consists of 1:1 dioctahedral layers. The dioctahedral layers consist of a sheet of corner-sharing SiO4 tetrahedra and a sheet of edge-sharing AlO6 octahedra linked by common oxygen atoms parallel to the (0 0 1) sheet with a composition of Si4Al4O10(OH)8 [20]. The kaolinite structure used here was that determined experimentally by Bish and Von Dreele [21], and the space group symmetry C1 presents the following kaolinite lattice parameters: a = 0.515 nm, b = 0.894 nm, c = 0.739 nm, a = 91.93°, b = 105.05°, and c = 89.80°. The kaolinite model is shown in Fig. 3, and the atom positions have been used in kaolinite adsorbed water simulations and verification by comparison with experimental results [22,23]. Considering the kaolinite structure model cannot reflect the periodicity and the basal spacing of a natural kaolinite, the 4 2 2 kaolinite super model was established as the research object; the model size is 2.059 nm 1.787 nm 2.519 nm, and the layer spacing is 0.72 nm [24–27], as shown in Fig. 4.
Fig. 3. The molecular model of kaolinite.
2.2. Force field The research object is the system consisting of kaolinite and CH4, in which the total potential energy is composed of the intermolecular energy between kaolinite and CH4 as well as a nonbonded energy between CH4 molecules. Therefore, the Dreiding force field was selected [28], in which the total potential energy E consists of intermolecular energy and non-bonded energy, the formula can be expressed as follows:
Fig. 4. The 4 2 2 super molecular model of kaolinite.
E¼
1X 1X kb ðbi b0 Þ þ kh ðhi h0 Þ i i 2 2 1
2
X 1X þ v i ð1 þ cos uÞ þ i v i ð1 cos 2uÞ þ v i ð1 þ cos 3uÞ i 2 3 " 6 # 12 X X 1X dij dij 2 þ kx ðxi x0 Þ þ 4 i j eij i 2 r ij r ij 4
5
" 10 # 12 X X qi qj Rhb Rhb cos4 ðhDHA Þ þ þ D 5 hb i j r RDA RDA ij Fig. 1. The molecular model of methane.
6
7
ð1Þ
794
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
As shown in Eq. (1), Dreiding includes bond energy (term 1), angle energy (term 2), torsion energy (term 3), and inversion energy (term 4) in the bonded potential. The non-bonded potential is composed of a Van der Waals energy (term 5), a Coulomb energy (term 6) and a hydrogen bonds energy (term 7). Atom charges and the Lennard-Jones parameters of CH4 and kaolinite were taken from Mayo, Olafson, and Goddard [28], Smirnov and Bougeard [29], and Martin and Siepmann [30] and shown in Table 1. The Coulomb interactions between charges in the system are estimated using the Ewald summation, and the accuracy is 4.186 103 kJ mol1; the Van deer Waals interactions are determined by the atom based cut-off radius, which is 0.8 nm [22]. We placed an empty space in the simulation cell along the z direction with a length much larger than Lx or Ly. The three-dimensional Ewald summation with the correction term [31,32] was accounted for in the long-range electrostatic interactions and the slab geometry. 2.3. Implementation of simulation The system studied consists of kaolinite, water, and CH4. The initial configuration for hydrated kaolinite is achieved by inserting a certain number of water molecules into the simulation box depending on the water contents. In this study, we consider the number of water molecules 9, 18, 27, 36, and 54, which correspond to 1, 2, 3, 4, and 5 wt%, respectively. All molecules were placed in an empty space of the periodic simulation box which is large enough to accommodate the kaolinite (0 0 1) surface and water. Periodic boundary conditions were applied in three directions. GCMC simulations with lVT ensemble were used to calculate the adsorption isotherms and interaction energies of methane on kaolinite with 0–5 wt% water at 293.15 K up to 20 MPa. For each case, there are three types of operations with equal probability in simulation cell: displacement, creation and deletion. From Eq. (1), total potential energy U (before a certain operation), U0 (after a certain operation) and the increment DU = U0 U are calculated. Then, according to the assumed acceptance probabilities, this operation is accepted or not. For displacement, creation and deletion, the acceptance probabilities are min f1; exp½DU=ðkB TÞg, and n 3 o min 1; kVNi i exp ½ðl þ DUÞ=ðkB TÞ , respectively [27], where qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 k ¼ h =ð2pmkB TÞ, h is Planck constant, m is the mass of CH4 molecule, T is the temperature, l is the chemical potential, V i is the volume of the ith subspace and N i is the number of CH4 in the ith subspace. So, for each case, 1 108 configurations were generated which were adequate for simulations [33–35], and half of which were produced for the system to reach equilibrium, and the other half were produced for calculating the adsorption isotherms and interaction energies. And the final configurations of the GCMC simulations were used as the initial input in the MD simulations [36,37]. MD simulations were used to analyze the kaolinite swelling and the radial distribution function (RDF) in two different ways. To analyze the kaolinite swelling, MD simulations were performed in the NPT ensemble. However, to analyze the radial distribution
function (RDF), MD simulations were performed in the NVE ensemble. Each MD simulation was done in two steps. The first step consisted of a 5 ns simulation to equilibrium, and another step was used to sample (5 ns). All of the GCMC and MD simulations were performed using the Sorption and Forcite modules of the Materials Studio package [38]. 3. Results and discussion 3.1. Modeling To validate the simulated data and to evaluate the rationality of the model, the simulated results were compared with experimental adsorption data from the literature [39]. Our calculated value of the lattice vector a, b, c and the lattice angle a, b, c of the kaolinite structure are in good agreement with the data from the experiment and calculation in Table 2. The lattice vector a, b of the 4 2 2 kaolinite super cell model are 2.059 nm and 1.787 nm, which are very close to the results obtained by Fang et al. [40]. Liming Ji et al. [41] measured an experimental adsorption isotherm of CH4 at 323.15 K in clay-rock derived from a kaolinite quarry in Fugu County, Shanxi Province, China, developed in a Carboniferous–Permian kaolinite-bearing strata. The samples contained 95% kaolinite and 5% quartz. Therefore, an adsorption isotherm of methane was simulated at 323.15 K as well. By comparing the GCMC-simulated isotherms and experimental isotherms (see Fig. 5), we found that the GCMC-simulated isotherm is higher than the experimental isotherm. This difference exists because the experimental samples have 5% quartz, which is not adsorbing CH4, and the simulation model is pure kaolinite. The kaolinite model and the model parameters proposed here are a relatively rational model compared to isotherms in previous studies because of its roots in experimental data from actual kaolinite. Accordingly, this model can be used to further to investigate the adsorption of CH4 in moist kaolinite. 3.2. CH4 adsorption isotherm Absolute adsorption isotherms were simulated for CH4 on both dry and moist kaolinite at 293.15 K up to 20 MPa. Our results, presented in Fig. 6, shows the results demonstrate that the absolute adsorption of CH4 is reduced under moist conditions and decreases with increasing water content. At a pressure up to 20 MPa, the CH4 sorption capacity is reduced by approximately 0.25 (6.6%), 0.304 (8.6%), 0.47 (13.7%), 0.544 (16.5%) and 0.748 molecules/unit cell (23.2%) for 1–5% water content, respectively, compared with dry kaolinite. Jin et al. [16] investigated the effect of moisture on CH4 on montmorillonite clay at 298.15 K up to 6 MPa, and they also found that the CH4 sorption capacity was reduced with increasing water content. To investigate the rate of adsorption, the simulation results of the adsorption were fitted by the Langmuir equation [41]:
V ¼ V L P=ðPL þ PÞ
ð2Þ
Table 1 Lennard-Jones parameters and atomic charge. Molecule
Element
CH4 Kaolinite
Si Al O H
r/Å
e/kcalmol1
q=e
3.73
0.2939
0
4.27 4.39 3.4046 3.195
0.31 0.31 0.0957 0.0152
+1.25 +1.05 0.85 +0.28
795
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800 Table 2 Lattice parameters of kaolinite structure compared with the data from the experiment and calculation.
a b
Kaolinite structure
a (nm)
b (nm)
c (nm)
a (°)
b (°)
c (°)
This work Cala Expb
0.515 0.515 0.515
0.894 0.893 0.894
0.739 0.738 0.740
91.93 91.93 91.70
105.05 105.04 104.86
89.80 89.79 89.82
Fang et al. [40]. Bish [21].
Table 3 Langmuir constant extracted by fitting the Langmuir isotherm model to our simulated absolute adsorption at 293.15 K. Water (wt%)
V L (molecules/uc)
P L (Mpa)
Correlation coefficient R2
0% 1% 2% 3% 4% 5%
4.726 4.552 4.361 4.216 3.967 3.817
4.889 4.819 4.789 4.611 4.579 4.478
0.999 0.999 0.998 0.999 0.998 0.997
Fig. 5. Comparison between simulation and experience results.
Fig. 7. Trend of adsorption capacity of CH4 on kaolinite with water content.
3.3. Kaolinite swelling 0Þ The kaolinite swelling ratio is defined as swelling ratio = ðV iVV 0
Fig. 6. Absolute adsorption isotherm of CH4 on dry and moist kaolinite with 1–5% water at a temperature of 293.15 K.
where V is the absolute amount adsorbed, molecules/unit cell; VL is the maximum absolute amount adsorbed, molecules/unit cell; PL is the Langmuir pressure, MPa; and P is the pressure, MPa. The fitting curves are presented in Fig. 6, and the fitting Langmuir constants are listed in Table 3. From Fig. 6, we can determine that the CH4 absolute adsorption of the dry and moist kaolinite follows the Langmuir isotherm. Many research groups [42,43] reasonably believe that the Langmuir isotherm can fit the microscopic simulation results. From Table 3 and Fig. 7, we can observe that the Langmuir constant VL decreased linearly with water content, following V L ¼ 0:1841 wt% þ 4:7335, and PL also decreased with increasing water content, indicating that both the adsorption capacity and the rate of CH4 adsorption decrease with the moisture content increases.
[44], where V 0 and V i is the initial and current occupied volume by kaolinite molecules, respectively. Fig. 8(a) presents the calculated kaolinite volume as a function of pressure for six moist systems. It was illustrated that the kaolinite volume increases with increasing moisture content. Kaolinite has a small degree expand/swelling by adsorption of water [45]. This is due to the fact that the kaolinite does not swell in the interlayer space by adsorption of water, but involves an increase in volume due to adsorption of water molecules between individual kaolinite clay particles according to the study of Barshad [46]. Abdi and Wild [47] measured the linear expansion vs. soaking time for kaolinite-6 wt% lime-gypsum cylinders initially moist cured at 30 °C and 100% r. h. for one week and soaked in distilled water. The result shown that the linear expansion is close to 2% for kaolinite-6 wt% lime with 0% gypsum. By comparing the experimental data from the study of Abdi and Wild and our molecular simulation date (see Fig. 8a), we found that the molecular simulation date was little higher than the experimental data because the experimental samples had 6% lime, which can reduce substantially the water absorp-
796
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
Fig. 7 showed that moist kaolinite has a significantly lower sorption capacity for CH4 than dry kaolinite. The results of helium pore volume computations indicate that the reduction in sorption capacity by the kaolinite preloaded water is due to the reduction in the space available for CH4 as water progressively occupies the microporosity of the kaolinite matrix. The presence of water results in a lower sorption capacity for CH4; the swelling of the kaolinite demonstrates the important role water plays in kaolinite swelling and CH4 adsorption. To quantify the effect of moisture on kaolinite swelling, in Fig. 8(c), we show the volumetric strain for the kaolinite with different water contents. For the CH4-saturated dry kaolinite, the values range from 0 to 1.19%, while the values range from 2.19% to 3.14% for the kaolinite with 5 wt% water. Heller et al. [48] reported that clay swells to an amount that is approximately linearly proportional to the amount of adsorption. It should be noted that their swelling strain measurement was for methane adsorption-induced swelling at 25 °C and at pressures up to 30 MPa. Pre-adsorbed water adsorption-induced swell on a kaolinite sample were not measured in their work. We believe that this aspect has not been clearly demonstrated in previous research. 3.4. Adsorbed CH4 density We computed the adsorbed phase density and bulk density, and the results are plotted in Fig. 9. Fig. 9(a) was calculated by the absolute adsorption amount divided by pore volume, and Fig. 9 (b) was computed from the Peng–Robinson equation of state
Fig. 8. Kaolinite volume and Free volumes. (a) Kaolinite volume as a function of pressure with 0–5 wt% water at 293.15 K. (b) Free volume as a function of pressure with 0–5 wt% water at 293.15 K. (c) Volumetric strain for a CH4 saturated kaolinite with 0–5 wt% water at 293.15 K.
tion, linear expansion and swelling pressure of the kaolinite, and the simulation model represented pure kaolinite. Fig. 8(b) presents the calculated kaolinite helium pore volume as a function of pressure for six moist systems. In contrast, the helium pore volume decreases with the amount of moisture. The reduction in the helium pore space is due to the increase in the amount of water in the pore structure of the kaolinite matrix.
Fig. 9. Adsorbed and bulk densities of CH4. (a) CH4 adsorbed phase densities on dry and moist kaolinite as a function of pressure at 293.15 K. (b) CH4 bulk densities as a function of pressure at 293.15 K.
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
[49]. Billemont et al. [15] reported values up to 0.12 g/cm3 and 0.18 g/cm3 for the densities of confined CH4 at 300 K in the slit carbon nanopore at water contents of 0.26 g/cm3 and 0.052 g/cm3, respectively. Our values were up to 0.107 g/cm3 and 0.118 g/cm3 for the adsorbed phase density of CH4 at 293.15 K and pressures up to 20 MPa in kaolinite with water contents of 3 wt% and 1 wt %, respectively, which are lower than the values reported by Billemont et al. [15]. They also observed a linear relationship between the maximum amount of adsorption and the present water amount. However, close inspection of our data indicates that at pressures higher than 4 MPa, a nonlinear relationship is evident. These quantitative differences might due to the fact that our temperature was lower, and the simplified carbon based slit-pore model might not be sufficient to describe kaolinite chemical properties, complex pore shape and swelling. The adsorbed phase density relies on a cooperative effect involving both adsorption and pore volume change due to kaolinite swelling. Furthermore, our radial distribution results, presented in Section 3.6, shows that oxygen and hydrogen play an important role in water adsorption on kaolinite surfaces. For moist kaolinite, the simplified carbon based slit-pore model is not suitable. 3.5. Interaction energies and isosteric heat of adsorption The present water effect on the adsorption capacity can be further investigated by analyzing the interaction energy between the kaolinite, water, and CH4. The total interaction energy is composed of van der Waals energy and electrostatic energy contributions [27,50]. Simple molecular models that neglect the direct electrostatic and high order interactions can underestimate the total interaction potentials by 5–15% [51,52]. The interaction energy can be defined as [34,53–55],
EInteraction ¼ EAB ðEA þ EB Þ
ð3Þ
where EAB, EA and EB are the energies of AB complex, A and B isolate, respectively. To calculate the interaction energy between hydrated kaolinite and CH4, EAB, EA and EB are energies of the complex of hydrated kaolinite and CH4, hydrated kaolinite and CH4 isolate, respectively. For the interaction energy between kaolinite and water, EAB, EA and EB are energies of the complex of kaolinite adsorbed CH4 and water, kaolinite adsorbed CH4 and water isolate, respectively. And the interaction energies of CH4 and water molecules were obtained by using single-point calculation [56]. The interaction energy between the CH4 and kaolinite, between the CH4 and water, between the CH4 and CH4, and between the kaolinite and water were analysed, as shown in Fig. 10. For the purpose of clarity, only the results for the CH4-saturated kaolinite with 4% water are shown. The preloaded water condition was determined to further investigate the interaction between the kaolinite and the CH4. The dominating effect of the kaolinite–water interaction becomes less apparent with increasing pressure as more CH4 molecules are adsorbed. The interaction energy between the kaolinite and the CH4 decreases logarithmically with increasing pressure, E = 6.2212ln(P) 10.772. In contrast, the interaction energy between the kaolinite and the water increases linearly with increasing pressure, E = 0.1794P 31.403. At a temperature of 293.15 K, the pressure points of 1, 5, 10, 15, and 20 MPa correspond to kaolinite – CH4 interaction energies of 8.829, 24.497, 26.404, 26.813 and 27.138 kJ/mol, respectively. The increasing energies of absolute value reflect greater interactions between the kaolinite and CH4 molecules with increasing pressure. According to the results, the kaolinite–water interaction dominates and the CH4–CH4 and CH4–H2O interactions each contribute less than 0.5% to the total interaction energy. We infer that kaolinite swelling and the decrease of CH4 adsorption capacity in the preloaded water condition is mainly determined by stronger intermolecular
797
Fig. 10. Interaction energy of CH4 – kaolinite, H2O – kaolinite, CH4 – H2O, and CH4 – CH4 for moist kaolinite with 4 wt% water at 293.15 K.
interactions between kaolinite and water. For interactions between kaolinite and CH4, the results for the six moist systems are illustrated in Fig. 11. It is clear that the interaction energy is determined by water content. This energy becomes less negative when the water content is increased from 0 wt% to 5.0 wt%, indicating that the presence of water reduces the interaction between the kaolinite and CH4. In our simulation results, the reduction in sorption capacity by kaolinite in the presence of water could therefore be explained by the fact that water in the kaolinite matrix reduces the interaction between kaolinite and CH4 because the CH4/kaolinite interaction is weaker than the water/kaolinite interaction. The adsorption isosteric heat in unit of kJ/mol can provide information on the release of energy in the adsorption process. These values can be calculated through a set of isotherms according to the Clausius–Clapeyron equation [57],
qst ¼ RT 2
@ðln PÞ @T
ð3Þ N
Fig. 12 shows the adsorption isosteric heats of CH4 on kaolinite for six moist systems at 293.15 K. As seen in Fig. 12, the adsorption isosteric heat of CH4 on kaolinite decreases linearly with increasing water content,
Fig. 11. Effect of water content on the interaction of between kaolinite and CH4 for dry and moist kaolinite at 293.15 K.
798
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
qst ¼ 0:1032 wt% þ 14:088, indicating that at higher water contents, the interaction energy between CH4 and kaolinite is weaker.
pressure, the first peak between CH4 and oxygen is higher, indicating that higher pressure, the more CH4 molecules are packed around oxygen in kaolinite.
3.6. Radial distribution function The atomic density varies as a function of the distance from one particular atom and can be calculated by the radial distribution function g(r), which is defined as follows [58],
g ij ðrÞ ¼
dN 4pqj r 2 dr
ð4Þ
where dN is the number of molecules j from r to r + dr of i and qj is the density of j. From the RDFs, it is possible to acquire information on the density by computing the absolute value of the RDFs, while its shape reflects the structure and interaction force, which can reflect the strength of the intermolecular interaction force. The RDFs between CH4/H2O and kaolinite inner surface species (hydroxyl hydrogen atoms of alumina octahedral surface and oxygen atoms of silicon tetra-hedral surface) with 4 wt% water at 293.15 K are shown in Fig. 13a–d. The first peak between H2O and oxygen of silicon tetra-hedral surface is stronger than the first peak between CH4 and oxygen (Fig. 13a). We infer that H2O and CH4 can be adsorbed onto the sites of the oxygen atoms in of silicon tetra-hedral surface of kaolinite molecules, but H2O is stronger. The close contact peak between CH4 and hydroxyl hydrogen atoms of alumina octahedral surface of kaolinite (Fig. 13b) is barely detected, showing an extremely weak interaction between CH4 and hydrogen. In contrast, the RDFs between H2O and hydrogen show that the first peak is sharp and strong, with a maximum value of 11.73 at a separation of 0.2 nm in kaolinite. We infer that H2O is strongly adsorbed onto the sites of the hydroxyl hydrogen atoms of alumina octahedral surface of kaolinite molecules. Therefore, moist kaolinites have a lower sorption capacity for CH4 than dry kaolinites. To investigate the moisture effect on the interaction between CH4 and oxygen of silicon tetra-hedral surface in kaolinite, we compare the RDFs of CH4–O at different water contents for kaolinite in Fig. 13c. As is shown, at the higher water contents, the first peak between CH4 and oxygen is lower, indicating that with higher water content, the interaction energy between CH4 and oxygen in kaolinite is weaker. To investigate the pressure effect on the packing of CH4, we compare the RDFs of CH4–O at different pressures for kaolinite with 4 wt% water in Fig. 13d. In Fig. 13d, at the higher
Fig. 12. Charge regulation of adsorption heat of CH4 on kaolinite with water content.
Fig. 13. Radial distribution functions (RDFs) between CH4/H2O and atoms in kaolinite in dry and moist kaolinite at 293.15 K. (a) CH4/H2O – O in kaolinite with 4 wt% water; (b) CH4/H2O– H in kaolinite with 4 wt% water; (c) CH4 – O in kaolinite with 0–5 wt% water at 20 MPa; (d) CH4 – O in kaolinite with 4 wt% water at 1, 5, 10, 15 and 20 MPa.
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
4. Conclusion Molecular simulations, which include the MD and MC methods, were used to study the CH4 adsorption on dry ad moist kaolinite at the pressure range of 1–20 MPa and a temperature of 293.15 K. Our results illustrate that the absolute adsorption of CH4 onto kaolinite is reduced in the presence of water and decreases with increasing water content. The absolute adsorption of CH4 on both dry and moist kaolinite followed the Langmuir isotherm within the simulated pressure range. From the Langmuir equation, we observed that the Langmuir constants, and the rate of CH4 adsorption decreased with increasing water content, indicating that both the adsorption capacity moisture content increases. The isosteric heat of adsorption of CH4 on kaolinite decreased linearly with increasing water content, qst ¼ 0:1032 wt% þ 14:088, indicating that at higher water contents, the interaction energy between the CH4 and kaolinite was weaker. The interaction between kaolinite and water dominates and is the main contributing factor to kaolinite clay swelling. Water molecules were preferentially adsorbed onto oxygen and hydrogen atoms in kaolinite, while methane showed a tendency of only being adsorbed onto oxygen. The adsorption isotherm using molecular simulation and the published experimental data were compared, revealing that the simulation is quite satisfactory using the Langmuir theory. Effects of water on CH4 adsorption capacity, adsorption rate, and kaolinite swelling were quantitatively researched from a microscopic perspective in this article. Our findings also offered insights into aspects of the adsorbed phase density for experiments. Molecular simulation is a cost-effective and efficient method to predict gas adsorption behavior in dry and moist kaolinites. We hope that our study will contribute to the development of strategies for the further exploration of coal bed methane and shale gas.
Acknowledgements This research was supported financially by the National Natural Science Foundation of China (51174141 and 50974093), the Postgraduate Innovation Fund of Shanxi Province (80010402100675), and the Taiyuan University of Technology Postgraduate Technology Innovation Fund (8004-02020061). The use of the Materials Studio software package, which is supported by the Key Laboratory of Coal Science and Technology of the Ministry of Education and Shanxi Province, is gratefully acknowledged.
References [1] C.R. Clarkson, N. Solano, R.M. Bustin, A.M.M. Bustin, G.R.L. Chalmers, L. He, Y.B. Melnichenko, A.P. Radlin´ski, T.P. Blach, Pore structure characterization of North American shale gas reservoirs using USANS/SANS, gas adsorption, and mercury intrusion, Fuel 103 (2013) 606–616. [2] S. Zhang, S. Tang, Q. Zheng, Z. Pan, Q. Guo, Evaluation of geological features for deep coalbed methane reservoirs in the Dacheng Salient, Jizhong Depression, China, Int. J. Coal Geol. 133 (2014) 60–71. [3] Benazzouz, B.K. Zaoui, A. Belonoshko, B. Anatoly, Determination of the melting temperature of kaolinite by means of the Z-method, Am. Mineral. 98 (2013) 1881–1885. [4] H.H. Murray, Applied clay mineralogy today and tomorrow, Clay Miner. 34 (1999) 39–49. [5] G.M. Reed, C.J. Mclaughlin, R. Newman, Kaolinite properties, structure and influence of metal retention on pH, Appl. Clay Sci. 23 (2003) 133–139. [6] J. Zhang, Investigations of water inrushes from aquifers under coal seams, Int. J. Rock Mech. Min. Sci. 42 (2005) 350–360. [7] C.D. Jenkins, I.I. Charles Boyer, Coalbed and shale-gas reservoirs, J. Petrol. Technol. 60 (2008) 92–99. [8] Z. Zhai, X. Wang, X. Jin, L. Sun, J. Li, D. Cao, Adsorption and diffusion of shale gas reservoirs in modeled clay minerals at different geological depths, Energy Fuels 28 (2014) 7467–7473. [9] J.L. Provis, P. Duxson, G.C. Lukey, J.S.J.V. Deventer, Statistical thermodynamic model for Si/Al ordering in amorphous aluminosilicates, Chem. Mater. 17 (2005) 2976–2986.
799
[10] P. Duxson, J.L. Provis, G.C. Lukey, F. Separovic, J.S. van Deventer, 29Si NMR study of structural ordering in aluminosilicate geopolymer gels, Langmuir Acs J. Surf. Colloids 21 (2005) 3028. [11] V. Marry, P. Turq, T. Cartailler, D. Levesque, Microscopic simulation of structure and dynamics of water and counterions in a monohydrated montmorillonite, J. Chem. Phys. 117 (2002) 3454–3463. [12] D.J.K. Ross, R.M. Bustin, The importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs, Mar. Pet. Geol. 26 (2009) 916–927. [13] D.J.K. Ross, R.M. Bustin, Shale gas potential of the Lower Jurassic Gordondale Member, northeastern British Columbia, Canada, Bull. Can. Pet. Geol. 55 (2007) 51–75. [14] M. Chávezpáez, K.V. Workum, L.D. Pablo, J.J.D. Pablo, Monte Carlo simulations of Wyoming sodium montmorillonite hydrates, J. Chem. Phys. 114 (2001) 1405–1413. [15] P. Billemont, B. Coasne, W.G. De, An experimental and molecular simulation study of the adsorption of carbon dioxide and methane in nanoporous carbons in the presence of water, Langmuir Acs J. Surf. Colloids 27 (2011) 1015–1024. [16] Z. Jin, A. Firoozabadi, Effect of water on methane and carbon dioxide sorption in clay minerals by Monte Carlo simulations, Fluid Phase Equilib. 382 (2014) 10–20. [17] D.P. Cao, X.R. Zhang, J.F. Chen, W.C. Wang, J. Yun, Optimization of single-walled carbon nanotube arrays for methane storage at room temperature, J. Phys. Chem. B 107 (2003) 13286–13292. [18] J. Foerster, J. Lansford, J. Laskar, T. Rappaport, S. Kato, High uptakes of methane in Li-doped 3D covalent organic frameworks, Langmuir Acs J. Surf. Colloids 26 (2010) 220–226. [19] J. Hermans, H.J.C. Berendsen, W.F.V. Gunsteren, J.P.M. Postma, A consistent empirical potential for water–protein interactions, Biopolymers 23 (1984) 1513–1518. [20] M.R. Warne, N.L. Allan, T. Cosgrove, Computer simulation of water molecules at kaolinite and silica surfaces, Phys.chem.chem.phys 2 (2000) 3663–3668. [21] D.L. Bish, Rietveld refinement of non-hydrogen atomic positions in kaolinite, Clays Clay Miner. 37 (1989) 289–296. [22] G. Rutkai, T. Kristóf, Molecular simulation study of intercalation of small molecules in kaolinite, Chem. Phys. Lett. 462 (2008) 269–274. [23] M.C. He, J. Zhao, Effects of Mg, Ca, and Fe(II) doping on the kaolinite (0 0 1) surface with H2O adsorption, Clays Clay Miner. 60 (2012) 330–337. [24] X. Hua, F.Z. Wang, X.Q. Li, Preparation and characterization of kaolintepotassium acetate intercalation comple, J. Harbin Inst. Technol. 38 (2006) 126–129. [25] L. Zhang, C.J. Wang, Z.F. Yan, X.L. Wu, Y.Q. Wang, D.W. Meng, H.X. Xie, Kaolinite nanomaterial: Intercalation of 1-butyl-3-methylimidazolium bromine in a methanol–kaolinite pre-intercalate, Appl. Clay Sci. 86 (2013) 106–110. [26] S. Zhang, Q. Liu, H. Cheng, X. Li, F. Zeng, R.L. Frost, Intercalation of dodecylamine into kaolinite and its layering structure investigated by molecular dynamics simulation, J Colloid Interface Sci 430 (2014) 345–350. [27] B. Zhang, J. Kang, T. Kang, Monte Carlo simulations of methane adsorption on kaolinite as a function of pore size, J. Nat. Gas Sci. Eng. 49C (2018) 410–416. [28] S.L. Mayo, B.D. Olafson, W.A. Goddard, DREIDING: a generic force field for molecular simulations, J. Phys. Chem. 94 (1990) 8897–8909. [29] K.S. Smirnov, D. Bougeard, A molecular dynamics study of structure and shorttime dynamics of water in kaolinite, J. Phys. Chem. B 103 (1999) 5266–5273. [30] M.G. Martin, J.I. Siepmann, A molecular dynamics study of structure and shorttime dynamics of water in kaolinite, J. Phys. Chem. B 103 (1999) 5266–5273. [31] I.C. Yeh, M.L. Berkowitz, Ewald summation for systems with slab geometry, J. Chem. Phys. 111 (1999) 3155–3162. [32] P.S. Crozier, R.L. Rowley, E. Spohr, D. Henderson, Comparison of charged sheets and corrected 3D Ewald calculations of long-range forces in slab geometry electrolyte systems with solvent molecules, J. Chem. Phys. 112 (2000) 9253– 9257. [33] A. Hirotani, K. Mizukami, R. Miura, H. Takaba, T. Miya, A. Fahmi, A. Stirling, M. Kubo, A. Miyamoto, Grand canonical Monte Carlo simulation of the adsorption of CO2 on silicalite and NaZSM-5, Appl. Surf. Sci. 120 (1997) 81–84. [34] X.Q. Liu, X. He, N.X. Qiu, X. Yang, Z.Y. Tian, M.J. Li, Y. Xue, Molecular simulation of CH4, CO2, H2O and N2 molecules adsorption on heterogeneous surface models of coal, Appl. Surf. Sci. 389 (2016) 894–905. [35] A. Gotzias, H. Heiberg-Andersen, M. Kainourgiakis, T. Steriotis, Grand canonical Monte Carlo simulations of hydrogen adsorption in carbon cones, Appl. Surf. Sci. 256 (2010) 5226–5231. [36] Z. Xiang, D. Cao, Porous covalent-organic materials: synthesis, clean energy application and design, J. Mater. Chem. A 1 (2013) 2691–2718. [37] Z. Yang, D. Cao, Effect of Li doping on diffusion and separation of hydrogen and methane in covalent organic frameworks, J. Phys. Chem. C 116 (2012) 12591– 12598. [38] X. Accelrys, Accelrys. Materials Studio Release Notes, Release 6.0, Accelrys Software Inc, San Diego, 2010. [39] R.S. Shulyak, V.V. Primachenko, L.N. Zolotukhina, A.N. Zakovorotnyi, Investigating Chalganovsk kaolin deposits, Refractories 33 (1992) 323–325. [40] Q. Fang, S. Huang, W. Wang, Intercalation of dimethyl sulphoxide in kaolinite: molecular dynamics simulation study, Chem. Phys. Lett. 411 (2005) 233–237. [41] L. Ji, T. Zhang, K.L. Milliken, J. Qu, X. Zhang, Experimental investigation of main controls to methane adsorption in clay-rich rocks, Appl. Geochem. 27 (2012) 2533–2545. [42] B. Sun, S. Kayal, A. Chakraborty, Study of HKUST (Copper benzene-1,3,5tricarboxylate, Cu-BTC MOF)-1 metal organic frameworks for CH4 adsorption:
800
[43]
[44]
[45] [46] [47] [48] [49] [50]
[51]
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800 an experimental investigation with GCMC (grand canonical Monte-carlo) simulation, Energy 76 (2014) 419–427. J. Zhang, M.B. Clennell, D.N. Dewhurst, K. Liu, Combined Monte Carlo and molecular dynamics simulation of methane adsorption on dry and moist coal, Fuel 122 (2014) 186–197. J. Liu, S. Keskin, D.S. Sholl, J.K. Johnson, Molecular simulations and theoretical predictions for adsorption and diffusion of CH4/H2 and CO2/CH4 mixtures in ZIFs, J. Phys. Chem. C 115 (2011) 12560–12566. C.X. Chen, C.H. Liu, A study on expanding characteristic of weathered granite, Key Eng. Mater. 261–263 (2004) 1391–1396. I. Barshad, Adsorptive and swelling properties of clay-water system, Clays Clay Miner. 1 (1952) 70–77. M.R. Abdi, S. Wild, Sulphate expansion of lime-stabilized kaolinite: I. Physical characteristics, Clay Miner. 28 (1993) 555–567. R. Heller, M. Zoback, Adsorption of methane and carbon dioxide on gas shale and pure mineral samples, J. Unconvention. Oil Gas Resour. 8 (2014) 14–24. D.Y. Peng, D.B. Robinson, A new two-constant equation of state, Ind. Eng. Chem. Fundam. 15 (1976) 92–94. F. Wu, C. Anutosh, Investigation of the interaction of polar molecules on graphite surface: prediction of isosteric heat of adsorption at zero surface coverage, J. Phys. Chem. C 120 (2016) 23490–23499. S. Namsani, Molecular simulation of shale gas adsorption and diffusion in inorganic nanopores, Mol. Simul. 41 (2015) 414–422.
[52] T. Zhang, G.S. Ellis, S.C. Ruppel, K. Milliken, R. Yang, Effect of organic-matter type and thermal maturity on methane adsorption in shale-gas systems, Org. Geochem. 47 (2012) 120–131. [53] J.S. Yang, C.L. Yang, M.S. Wang, B.D. Chen, X.G. Ma, Effect of functionalization on the interfacial binding energy of carbon nanotube/nylon 6 nanocomposites: a molecular dynamics study, Rsc Adv. 2 (2012) 2836–2841. [54] Q. Zheng, Q. Xue, K. Yan, X. Gao, Q. Li, L. Hao, Effect of chemisorption on the interfacial bonding characteristics of carbon nanotube–polymer composites, Polymer 49 (2008) 800–808. [55] C. Lv, Q. Xue, D. Xia, M. Ma, J. Xie, H. Chen, Effect of chemisorption on the interfacial bonding characteristics of graphenepolymer composites, J. Phys. Chem. C 114 (2010) 6588–6594. [56] L.A. García-Serrano, C.A. Flores-Sandoval, I.P. Zaragoza, Theoretical study of the adsorption of isobutane over H-mordenite zeolite by ab initio and DFT methods, J. Mol. Catal. A: Chem. 200 (2003) 205–212. [57] H. Pan, J.A. Ritter, P.B. Balbuena, Examination of the approximations used in determining the isosteric heat of adsorption from the ClausiusClapeyron equation, Langmuir 14 (2015) 535–542. [58] X.P. Kong, J. Wang, Copper (II) adsorption on the kaolinite (0 0 1) surface: insights from first-principles calculations and molecular dynamics simulations, Appl. Surf. Sci. 389 (2016) 316–323.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
Effect of water on methane adsorption on the kaolinite (0 0 1) surface based on molecular simulations Bin Zhang, Jianting Kang, Tianhe Kang ⇑ Institute of Mining Technology, Taiyuan University of Technology, Taiyuan 030024, PR China
a r t i c l e
i n f o
Article history: Received 6 June 2017 Revised 13 December 2017 Accepted 28 December 2017 Available online 12 January 2018 Keywords: Methane adsorption Kaolinite Water Molecular simulation
a b s t r a c t CH4 adsorption isotherms of kaolinite with moisture contents ranging from 0 to 5 wt% water, the effects of water on maximum adsorption capacity, kaolinite swelling, and radial distribution function were modelled by the implementing combined Monte Carlo (MC) and molecular dynamics (MD) simulations at 293.15 K (20 °C) and a pressure range of 1–20 MPa. The simulation results showed that the absolute adsorption of CH4 on both dry and moist kaolinite followed a Langmuir isotherm within the simulated pressure range, and both the adsorption capacity and the rate of CH4 adsorption decreased with the water content increases. The adsorption isosteric heats of CH4 on kaolinite decreased linearly with increasing water content, indicating that at higher water contents, the interaction energy between the CH4 and kaolinite was weaker. The interaction between kaolinite and water dominates and was the main contributing factor to kaolinite clay swelling. Water molecules were preferentially adsorbed onto oxygen and hydrogen atoms in kaolinite, while methane showed a tendency to be adsorbed only onto oxygen. The simulation results of our study provide the quantitative analysis of effect of water on CH4 adsorption capacity, adsorption rate, and interaction energy from a microscopic perspective. We hope that our study will contribute to the development of strategies for the further exploration of coal bed methane and shale gas. Ó 2018 Elsevier B.V. All rights reserved.
1. Introduction Clay minerals have large surface areas (approximately 800 m2 g1) and micropore to mesopore structures that can significantly affect the adsorption properties of porous media such as shale and coal [1,2]. Kaolinite is one of the most abundant components in clay minerals [3–5], and understanding the interaction between kaolinite and methane molecules is important for research in the fields of shale gas and coal bed methane. Some coal bed methane and shale gas reservoirs are water saturated [6,7], and the clay minerals that have similar silicoaluminate crystallographic layers with AlAO octahedra and SiAO tetrahedra are hydrophilic [8–10]. Water molecules can be adsorbed onto the surface of the clay mineral without difficulty, which can decrease the total methane sorption of clay minerals [11]. Hence, preloaded water can substantially decrease the total amount of methane adsorbed onto clay-rich rocks [12]. Ross et al. [13] found that the adsorbed capacities of water-saturated montmorillonite and illite were lower than under dry conditions. ⇑ Corresponding author at: Taiyuan University of Technology, No. 18 New Mine Road, West Yingze Street, Taiyuan City 030024, Shanxi Province, PR China. E-mail address: [email protected] (T. Kang). https://doi.org/10.1016/j.apsusc.2017.12.239 0169-4332/Ó 2018 Elsevier B.V. All rights reserved.
Ross and Bustin [12] found that at low pressures (6 MPa), the CH4 adsorption capacities of illite and montmorillonite were lower than that of kaolinite on a moisture-equilibrated basis but were significantly higher than that of kaolinite under dry conditions. Moreover, a few clay minerals, e.g., montmorillonite clay, are able to further enhance the interaction between clay molecules and methane molecules because of the cation-exchange capacity of montmorillonite clay [14]. Some molecular simulations have been carried out on the adsorption of gas in dry and moist clays. Billemont et al. [15] used grand canonical Monte Carlo (GCMC) simulations to consider the influence of water on methane sorption in porous carbons and observed that water molecules have a higher free energy barrier than methane molecules, which cannot displace the water molecules, and the preloaded water molecules notably decreased the adsorption capacity of methane. Jin et al. [16] also utilized GCMC simulations to research the influence of water on methane sorption in montmorillonite clay. However, to our knowledge, no computational and theoretical research has been carried out on the influence of water content on methane sorption in other clay minerals such as kaolinite and illite. Therefore, in this paper, we used the GCMC and molecular dynamics (MD) methods to research the effect of water on methane adsorption in kaolinite over a pres-
793
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
sure range of 1–20 MPa and with pre-adsorbed water contents of 0–5 wt% simulated at 293.15 K. Molecular simulations can reveal the absolute adsorption isotherms, interaction energy, isosteric heat of adsorption, kaolinite swelling, adsorbed phase density, and radial distribution functions between the surface of kaolinite and CH4. The objectives of this work were to better understand CH4/water/kaolinite clay interactions and to shed light on the influence of water on the adsorption capacity and kaolinite swelling of CH4 on kaolinite surfaces at the atomic level. The findings of this study are also expected to shed light on the details of coal bed methane and shale gas adsorption.
Fig. 2. The molecular model of water.
2. Simulation methods 2.1. Models We use methane to represent coal bed gas and shale gas because methane is the main component of coal bed gas and shale gas. Here, we mainly consider the case of methane in hydrated kaolinite. Methane is represented by an all-atom model [17,18]; the CAH bond length and the CAH bond angle were calculated to be 0.109 nm and 109°280 , respectively, as shown in Fig. 1. Water is represented by a simple point charge model [19]; the HAO bond length and the HAOAH bond angle were calculated to be 0.079 nm and 104°60 , respectively, as shown in Fig. 2. The kaolinite mineral consists of 1:1 dioctahedral layers. The dioctahedral layers consist of a sheet of corner-sharing SiO4 tetrahedra and a sheet of edge-sharing AlO6 octahedra linked by common oxygen atoms parallel to the (0 0 1) sheet with a composition of Si4Al4O10(OH)8 [20]. The kaolinite structure used here was that determined experimentally by Bish and Von Dreele [21], and the space group symmetry C1 presents the following kaolinite lattice parameters: a = 0.515 nm, b = 0.894 nm, c = 0.739 nm, a = 91.93°, b = 105.05°, and c = 89.80°. The kaolinite model is shown in Fig. 3, and the atom positions have been used in kaolinite adsorbed water simulations and verification by comparison with experimental results [22,23]. Considering the kaolinite structure model cannot reflect the periodicity and the basal spacing of a natural kaolinite, the 4 2 2 kaolinite super model was established as the research object; the model size is 2.059 nm 1.787 nm 2.519 nm, and the layer spacing is 0.72 nm [24–27], as shown in Fig. 4.
Fig. 3. The molecular model of kaolinite.
2.2. Force field The research object is the system consisting of kaolinite and CH4, in which the total potential energy is composed of the intermolecular energy between kaolinite and CH4 as well as a nonbonded energy between CH4 molecules. Therefore, the Dreiding force field was selected [28], in which the total potential energy E consists of intermolecular energy and non-bonded energy, the formula can be expressed as follows:
Fig. 4. The 4 2 2 super molecular model of kaolinite.
E¼
1X 1X kb ðbi b0 Þ þ kh ðhi h0 Þ i i 2 2 1
2
X 1X þ v i ð1 þ cos uÞ þ i v i ð1 cos 2uÞ þ v i ð1 þ cos 3uÞ i 2 3 " 6 # 12 X X 1X dij dij 2 þ kx ðxi x0 Þ þ 4 i j eij i 2 r ij r ij 4
5
" 10 # 12 X X qi qj Rhb Rhb cos4 ðhDHA Þ þ þ D 5 hb i j r RDA RDA ij Fig. 1. The molecular model of methane.
6
7
ð1Þ
794
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
As shown in Eq. (1), Dreiding includes bond energy (term 1), angle energy (term 2), torsion energy (term 3), and inversion energy (term 4) in the bonded potential. The non-bonded potential is composed of a Van der Waals energy (term 5), a Coulomb energy (term 6) and a hydrogen bonds energy (term 7). Atom charges and the Lennard-Jones parameters of CH4 and kaolinite were taken from Mayo, Olafson, and Goddard [28], Smirnov and Bougeard [29], and Martin and Siepmann [30] and shown in Table 1. The Coulomb interactions between charges in the system are estimated using the Ewald summation, and the accuracy is 4.186 103 kJ mol1; the Van deer Waals interactions are determined by the atom based cut-off radius, which is 0.8 nm [22]. We placed an empty space in the simulation cell along the z direction with a length much larger than Lx or Ly. The three-dimensional Ewald summation with the correction term [31,32] was accounted for in the long-range electrostatic interactions and the slab geometry. 2.3. Implementation of simulation The system studied consists of kaolinite, water, and CH4. The initial configuration for hydrated kaolinite is achieved by inserting a certain number of water molecules into the simulation box depending on the water contents. In this study, we consider the number of water molecules 9, 18, 27, 36, and 54, which correspond to 1, 2, 3, 4, and 5 wt%, respectively. All molecules were placed in an empty space of the periodic simulation box which is large enough to accommodate the kaolinite (0 0 1) surface and water. Periodic boundary conditions were applied in three directions. GCMC simulations with lVT ensemble were used to calculate the adsorption isotherms and interaction energies of methane on kaolinite with 0–5 wt% water at 293.15 K up to 20 MPa. For each case, there are three types of operations with equal probability in simulation cell: displacement, creation and deletion. From Eq. (1), total potential energy U (before a certain operation), U0 (after a certain operation) and the increment DU = U0 U are calculated. Then, according to the assumed acceptance probabilities, this operation is accepted or not. For displacement, creation and deletion, the acceptance probabilities are min f1; exp½DU=ðkB TÞg, and n 3 o min 1; kVNi i exp ½ðl þ DUÞ=ðkB TÞ , respectively [27], where qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 k ¼ h =ð2pmkB TÞ, h is Planck constant, m is the mass of CH4 molecule, T is the temperature, l is the chemical potential, V i is the volume of the ith subspace and N i is the number of CH4 in the ith subspace. So, for each case, 1 108 configurations were generated which were adequate for simulations [33–35], and half of which were produced for the system to reach equilibrium, and the other half were produced for calculating the adsorption isotherms and interaction energies. And the final configurations of the GCMC simulations were used as the initial input in the MD simulations [36,37]. MD simulations were used to analyze the kaolinite swelling and the radial distribution function (RDF) in two different ways. To analyze the kaolinite swelling, MD simulations were performed in the NPT ensemble. However, to analyze the radial distribution
function (RDF), MD simulations were performed in the NVE ensemble. Each MD simulation was done in two steps. The first step consisted of a 5 ns simulation to equilibrium, and another step was used to sample (5 ns). All of the GCMC and MD simulations were performed using the Sorption and Forcite modules of the Materials Studio package [38]. 3. Results and discussion 3.1. Modeling To validate the simulated data and to evaluate the rationality of the model, the simulated results were compared with experimental adsorption data from the literature [39]. Our calculated value of the lattice vector a, b, c and the lattice angle a, b, c of the kaolinite structure are in good agreement with the data from the experiment and calculation in Table 2. The lattice vector a, b of the 4 2 2 kaolinite super cell model are 2.059 nm and 1.787 nm, which are very close to the results obtained by Fang et al. [40]. Liming Ji et al. [41] measured an experimental adsorption isotherm of CH4 at 323.15 K in clay-rock derived from a kaolinite quarry in Fugu County, Shanxi Province, China, developed in a Carboniferous–Permian kaolinite-bearing strata. The samples contained 95% kaolinite and 5% quartz. Therefore, an adsorption isotherm of methane was simulated at 323.15 K as well. By comparing the GCMC-simulated isotherms and experimental isotherms (see Fig. 5), we found that the GCMC-simulated isotherm is higher than the experimental isotherm. This difference exists because the experimental samples have 5% quartz, which is not adsorbing CH4, and the simulation model is pure kaolinite. The kaolinite model and the model parameters proposed here are a relatively rational model compared to isotherms in previous studies because of its roots in experimental data from actual kaolinite. Accordingly, this model can be used to further to investigate the adsorption of CH4 in moist kaolinite. 3.2. CH4 adsorption isotherm Absolute adsorption isotherms were simulated for CH4 on both dry and moist kaolinite at 293.15 K up to 20 MPa. Our results, presented in Fig. 6, shows the results demonstrate that the absolute adsorption of CH4 is reduced under moist conditions and decreases with increasing water content. At a pressure up to 20 MPa, the CH4 sorption capacity is reduced by approximately 0.25 (6.6%), 0.304 (8.6%), 0.47 (13.7%), 0.544 (16.5%) and 0.748 molecules/unit cell (23.2%) for 1–5% water content, respectively, compared with dry kaolinite. Jin et al. [16] investigated the effect of moisture on CH4 on montmorillonite clay at 298.15 K up to 6 MPa, and they also found that the CH4 sorption capacity was reduced with increasing water content. To investigate the rate of adsorption, the simulation results of the adsorption were fitted by the Langmuir equation [41]:
V ¼ V L P=ðPL þ PÞ
ð2Þ
Table 1 Lennard-Jones parameters and atomic charge. Molecule
Element
CH4 Kaolinite
Si Al O H
r/Å
e/kcalmol1
q=e
3.73
0.2939
0
4.27 4.39 3.4046 3.195
0.31 0.31 0.0957 0.0152
+1.25 +1.05 0.85 +0.28
795
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800 Table 2 Lattice parameters of kaolinite structure compared with the data from the experiment and calculation.
a b
Kaolinite structure
a (nm)
b (nm)
c (nm)
a (°)
b (°)
c (°)
This work Cala Expb
0.515 0.515 0.515
0.894 0.893 0.894
0.739 0.738 0.740
91.93 91.93 91.70
105.05 105.04 104.86
89.80 89.79 89.82
Fang et al. [40]. Bish [21].
Table 3 Langmuir constant extracted by fitting the Langmuir isotherm model to our simulated absolute adsorption at 293.15 K. Water (wt%)
V L (molecules/uc)
P L (Mpa)
Correlation coefficient R2
0% 1% 2% 3% 4% 5%
4.726 4.552 4.361 4.216 3.967 3.817
4.889 4.819 4.789 4.611 4.579 4.478
0.999 0.999 0.998 0.999 0.998 0.997
Fig. 5. Comparison between simulation and experience results.
Fig. 7. Trend of adsorption capacity of CH4 on kaolinite with water content.
3.3. Kaolinite swelling 0Þ The kaolinite swelling ratio is defined as swelling ratio = ðV iVV 0
Fig. 6. Absolute adsorption isotherm of CH4 on dry and moist kaolinite with 1–5% water at a temperature of 293.15 K.
where V is the absolute amount adsorbed, molecules/unit cell; VL is the maximum absolute amount adsorbed, molecules/unit cell; PL is the Langmuir pressure, MPa; and P is the pressure, MPa. The fitting curves are presented in Fig. 6, and the fitting Langmuir constants are listed in Table 3. From Fig. 6, we can determine that the CH4 absolute adsorption of the dry and moist kaolinite follows the Langmuir isotherm. Many research groups [42,43] reasonably believe that the Langmuir isotherm can fit the microscopic simulation results. From Table 3 and Fig. 7, we can observe that the Langmuir constant VL decreased linearly with water content, following V L ¼ 0:1841 wt% þ 4:7335, and PL also decreased with increasing water content, indicating that both the adsorption capacity and the rate of CH4 adsorption decrease with the moisture content increases.
[44], where V 0 and V i is the initial and current occupied volume by kaolinite molecules, respectively. Fig. 8(a) presents the calculated kaolinite volume as a function of pressure for six moist systems. It was illustrated that the kaolinite volume increases with increasing moisture content. Kaolinite has a small degree expand/swelling by adsorption of water [45]. This is due to the fact that the kaolinite does not swell in the interlayer space by adsorption of water, but involves an increase in volume due to adsorption of water molecules between individual kaolinite clay particles according to the study of Barshad [46]. Abdi and Wild [47] measured the linear expansion vs. soaking time for kaolinite-6 wt% lime-gypsum cylinders initially moist cured at 30 °C and 100% r. h. for one week and soaked in distilled water. The result shown that the linear expansion is close to 2% for kaolinite-6 wt% lime with 0% gypsum. By comparing the experimental data from the study of Abdi and Wild and our molecular simulation date (see Fig. 8a), we found that the molecular simulation date was little higher than the experimental data because the experimental samples had 6% lime, which can reduce substantially the water absorp-
796
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
Fig. 7 showed that moist kaolinite has a significantly lower sorption capacity for CH4 than dry kaolinite. The results of helium pore volume computations indicate that the reduction in sorption capacity by the kaolinite preloaded water is due to the reduction in the space available for CH4 as water progressively occupies the microporosity of the kaolinite matrix. The presence of water results in a lower sorption capacity for CH4; the swelling of the kaolinite demonstrates the important role water plays in kaolinite swelling and CH4 adsorption. To quantify the effect of moisture on kaolinite swelling, in Fig. 8(c), we show the volumetric strain for the kaolinite with different water contents. For the CH4-saturated dry kaolinite, the values range from 0 to 1.19%, while the values range from 2.19% to 3.14% for the kaolinite with 5 wt% water. Heller et al. [48] reported that clay swells to an amount that is approximately linearly proportional to the amount of adsorption. It should be noted that their swelling strain measurement was for methane adsorption-induced swelling at 25 °C and at pressures up to 30 MPa. Pre-adsorbed water adsorption-induced swell on a kaolinite sample were not measured in their work. We believe that this aspect has not been clearly demonstrated in previous research. 3.4. Adsorbed CH4 density We computed the adsorbed phase density and bulk density, and the results are plotted in Fig. 9. Fig. 9(a) was calculated by the absolute adsorption amount divided by pore volume, and Fig. 9 (b) was computed from the Peng–Robinson equation of state
Fig. 8. Kaolinite volume and Free volumes. (a) Kaolinite volume as a function of pressure with 0–5 wt% water at 293.15 K. (b) Free volume as a function of pressure with 0–5 wt% water at 293.15 K. (c) Volumetric strain for a CH4 saturated kaolinite with 0–5 wt% water at 293.15 K.
tion, linear expansion and swelling pressure of the kaolinite, and the simulation model represented pure kaolinite. Fig. 8(b) presents the calculated kaolinite helium pore volume as a function of pressure for six moist systems. In contrast, the helium pore volume decreases with the amount of moisture. The reduction in the helium pore space is due to the increase in the amount of water in the pore structure of the kaolinite matrix.
Fig. 9. Adsorbed and bulk densities of CH4. (a) CH4 adsorbed phase densities on dry and moist kaolinite as a function of pressure at 293.15 K. (b) CH4 bulk densities as a function of pressure at 293.15 K.
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
[49]. Billemont et al. [15] reported values up to 0.12 g/cm3 and 0.18 g/cm3 for the densities of confined CH4 at 300 K in the slit carbon nanopore at water contents of 0.26 g/cm3 and 0.052 g/cm3, respectively. Our values were up to 0.107 g/cm3 and 0.118 g/cm3 for the adsorbed phase density of CH4 at 293.15 K and pressures up to 20 MPa in kaolinite with water contents of 3 wt% and 1 wt %, respectively, which are lower than the values reported by Billemont et al. [15]. They also observed a linear relationship between the maximum amount of adsorption and the present water amount. However, close inspection of our data indicates that at pressures higher than 4 MPa, a nonlinear relationship is evident. These quantitative differences might due to the fact that our temperature was lower, and the simplified carbon based slit-pore model might not be sufficient to describe kaolinite chemical properties, complex pore shape and swelling. The adsorbed phase density relies on a cooperative effect involving both adsorption and pore volume change due to kaolinite swelling. Furthermore, our radial distribution results, presented in Section 3.6, shows that oxygen and hydrogen play an important role in water adsorption on kaolinite surfaces. For moist kaolinite, the simplified carbon based slit-pore model is not suitable. 3.5. Interaction energies and isosteric heat of adsorption The present water effect on the adsorption capacity can be further investigated by analyzing the interaction energy between the kaolinite, water, and CH4. The total interaction energy is composed of van der Waals energy and electrostatic energy contributions [27,50]. Simple molecular models that neglect the direct electrostatic and high order interactions can underestimate the total interaction potentials by 5–15% [51,52]. The interaction energy can be defined as [34,53–55],
EInteraction ¼ EAB ðEA þ EB Þ
ð3Þ
where EAB, EA and EB are the energies of AB complex, A and B isolate, respectively. To calculate the interaction energy between hydrated kaolinite and CH4, EAB, EA and EB are energies of the complex of hydrated kaolinite and CH4, hydrated kaolinite and CH4 isolate, respectively. For the interaction energy between kaolinite and water, EAB, EA and EB are energies of the complex of kaolinite adsorbed CH4 and water, kaolinite adsorbed CH4 and water isolate, respectively. And the interaction energies of CH4 and water molecules were obtained by using single-point calculation [56]. The interaction energy between the CH4 and kaolinite, between the CH4 and water, between the CH4 and CH4, and between the kaolinite and water were analysed, as shown in Fig. 10. For the purpose of clarity, only the results for the CH4-saturated kaolinite with 4% water are shown. The preloaded water condition was determined to further investigate the interaction between the kaolinite and the CH4. The dominating effect of the kaolinite–water interaction becomes less apparent with increasing pressure as more CH4 molecules are adsorbed. The interaction energy between the kaolinite and the CH4 decreases logarithmically with increasing pressure, E = 6.2212ln(P) 10.772. In contrast, the interaction energy between the kaolinite and the water increases linearly with increasing pressure, E = 0.1794P 31.403. At a temperature of 293.15 K, the pressure points of 1, 5, 10, 15, and 20 MPa correspond to kaolinite – CH4 interaction energies of 8.829, 24.497, 26.404, 26.813 and 27.138 kJ/mol, respectively. The increasing energies of absolute value reflect greater interactions between the kaolinite and CH4 molecules with increasing pressure. According to the results, the kaolinite–water interaction dominates and the CH4–CH4 and CH4–H2O interactions each contribute less than 0.5% to the total interaction energy. We infer that kaolinite swelling and the decrease of CH4 adsorption capacity in the preloaded water condition is mainly determined by stronger intermolecular
797
Fig. 10. Interaction energy of CH4 – kaolinite, H2O – kaolinite, CH4 – H2O, and CH4 – CH4 for moist kaolinite with 4 wt% water at 293.15 K.
interactions between kaolinite and water. For interactions between kaolinite and CH4, the results for the six moist systems are illustrated in Fig. 11. It is clear that the interaction energy is determined by water content. This energy becomes less negative when the water content is increased from 0 wt% to 5.0 wt%, indicating that the presence of water reduces the interaction between the kaolinite and CH4. In our simulation results, the reduction in sorption capacity by kaolinite in the presence of water could therefore be explained by the fact that water in the kaolinite matrix reduces the interaction between kaolinite and CH4 because the CH4/kaolinite interaction is weaker than the water/kaolinite interaction. The adsorption isosteric heat in unit of kJ/mol can provide information on the release of energy in the adsorption process. These values can be calculated through a set of isotherms according to the Clausius–Clapeyron equation [57],
qst ¼ RT 2
@ðln PÞ @T
ð3Þ N
Fig. 12 shows the adsorption isosteric heats of CH4 on kaolinite for six moist systems at 293.15 K. As seen in Fig. 12, the adsorption isosteric heat of CH4 on kaolinite decreases linearly with increasing water content,
Fig. 11. Effect of water content on the interaction of between kaolinite and CH4 for dry and moist kaolinite at 293.15 K.
798
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
qst ¼ 0:1032 wt% þ 14:088, indicating that at higher water contents, the interaction energy between CH4 and kaolinite is weaker.
pressure, the first peak between CH4 and oxygen is higher, indicating that higher pressure, the more CH4 molecules are packed around oxygen in kaolinite.
3.6. Radial distribution function The atomic density varies as a function of the distance from one particular atom and can be calculated by the radial distribution function g(r), which is defined as follows [58],
g ij ðrÞ ¼
dN 4pqj r 2 dr
ð4Þ
where dN is the number of molecules j from r to r + dr of i and qj is the density of j. From the RDFs, it is possible to acquire information on the density by computing the absolute value of the RDFs, while its shape reflects the structure and interaction force, which can reflect the strength of the intermolecular interaction force. The RDFs between CH4/H2O and kaolinite inner surface species (hydroxyl hydrogen atoms of alumina octahedral surface and oxygen atoms of silicon tetra-hedral surface) with 4 wt% water at 293.15 K are shown in Fig. 13a–d. The first peak between H2O and oxygen of silicon tetra-hedral surface is stronger than the first peak between CH4 and oxygen (Fig. 13a). We infer that H2O and CH4 can be adsorbed onto the sites of the oxygen atoms in of silicon tetra-hedral surface of kaolinite molecules, but H2O is stronger. The close contact peak between CH4 and hydroxyl hydrogen atoms of alumina octahedral surface of kaolinite (Fig. 13b) is barely detected, showing an extremely weak interaction between CH4 and hydrogen. In contrast, the RDFs between H2O and hydrogen show that the first peak is sharp and strong, with a maximum value of 11.73 at a separation of 0.2 nm in kaolinite. We infer that H2O is strongly adsorbed onto the sites of the hydroxyl hydrogen atoms of alumina octahedral surface of kaolinite molecules. Therefore, moist kaolinites have a lower sorption capacity for CH4 than dry kaolinites. To investigate the moisture effect on the interaction between CH4 and oxygen of silicon tetra-hedral surface in kaolinite, we compare the RDFs of CH4–O at different water contents for kaolinite in Fig. 13c. As is shown, at the higher water contents, the first peak between CH4 and oxygen is lower, indicating that with higher water content, the interaction energy between CH4 and oxygen in kaolinite is weaker. To investigate the pressure effect on the packing of CH4, we compare the RDFs of CH4–O at different pressures for kaolinite with 4 wt% water in Fig. 13d. In Fig. 13d, at the higher
Fig. 12. Charge regulation of adsorption heat of CH4 on kaolinite with water content.
Fig. 13. Radial distribution functions (RDFs) between CH4/H2O and atoms in kaolinite in dry and moist kaolinite at 293.15 K. (a) CH4/H2O – O in kaolinite with 4 wt% water; (b) CH4/H2O– H in kaolinite with 4 wt% water; (c) CH4 – O in kaolinite with 0–5 wt% water at 20 MPa; (d) CH4 – O in kaolinite with 4 wt% water at 1, 5, 10, 15 and 20 MPa.
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800
4. Conclusion Molecular simulations, which include the MD and MC methods, were used to study the CH4 adsorption on dry ad moist kaolinite at the pressure range of 1–20 MPa and a temperature of 293.15 K. Our results illustrate that the absolute adsorption of CH4 onto kaolinite is reduced in the presence of water and decreases with increasing water content. The absolute adsorption of CH4 on both dry and moist kaolinite followed the Langmuir isotherm within the simulated pressure range. From the Langmuir equation, we observed that the Langmuir constants, and the rate of CH4 adsorption decreased with increasing water content, indicating that both the adsorption capacity moisture content increases. The isosteric heat of adsorption of CH4 on kaolinite decreased linearly with increasing water content, qst ¼ 0:1032 wt% þ 14:088, indicating that at higher water contents, the interaction energy between the CH4 and kaolinite was weaker. The interaction between kaolinite and water dominates and is the main contributing factor to kaolinite clay swelling. Water molecules were preferentially adsorbed onto oxygen and hydrogen atoms in kaolinite, while methane showed a tendency of only being adsorbed onto oxygen. The adsorption isotherm using molecular simulation and the published experimental data were compared, revealing that the simulation is quite satisfactory using the Langmuir theory. Effects of water on CH4 adsorption capacity, adsorption rate, and kaolinite swelling were quantitatively researched from a microscopic perspective in this article. Our findings also offered insights into aspects of the adsorbed phase density for experiments. Molecular simulation is a cost-effective and efficient method to predict gas adsorption behavior in dry and moist kaolinites. We hope that our study will contribute to the development of strategies for the further exploration of coal bed methane and shale gas.
Acknowledgements This research was supported financially by the National Natural Science Foundation of China (51174141 and 50974093), the Postgraduate Innovation Fund of Shanxi Province (80010402100675), and the Taiyuan University of Technology Postgraduate Technology Innovation Fund (8004-02020061). The use of the Materials Studio software package, which is supported by the Key Laboratory of Coal Science and Technology of the Ministry of Education and Shanxi Province, is gratefully acknowledged.
References [1] C.R. Clarkson, N. Solano, R.M. Bustin, A.M.M. Bustin, G.R.L. Chalmers, L. He, Y.B. Melnichenko, A.P. Radlin´ski, T.P. Blach, Pore structure characterization of North American shale gas reservoirs using USANS/SANS, gas adsorption, and mercury intrusion, Fuel 103 (2013) 606–616. [2] S. Zhang, S. Tang, Q. Zheng, Z. Pan, Q. Guo, Evaluation of geological features for deep coalbed methane reservoirs in the Dacheng Salient, Jizhong Depression, China, Int. J. Coal Geol. 133 (2014) 60–71. [3] Benazzouz, B.K. Zaoui, A. Belonoshko, B. Anatoly, Determination of the melting temperature of kaolinite by means of the Z-method, Am. Mineral. 98 (2013) 1881–1885. [4] H.H. Murray, Applied clay mineralogy today and tomorrow, Clay Miner. 34 (1999) 39–49. [5] G.M. Reed, C.J. Mclaughlin, R. Newman, Kaolinite properties, structure and influence of metal retention on pH, Appl. Clay Sci. 23 (2003) 133–139. [6] J. Zhang, Investigations of water inrushes from aquifers under coal seams, Int. J. Rock Mech. Min. Sci. 42 (2005) 350–360. [7] C.D. Jenkins, I.I. Charles Boyer, Coalbed and shale-gas reservoirs, J. Petrol. Technol. 60 (2008) 92–99. [8] Z. Zhai, X. Wang, X. Jin, L. Sun, J. Li, D. Cao, Adsorption and diffusion of shale gas reservoirs in modeled clay minerals at different geological depths, Energy Fuels 28 (2014) 7467–7473. [9] J.L. Provis, P. Duxson, G.C. Lukey, J.S.J.V. Deventer, Statistical thermodynamic model for Si/Al ordering in amorphous aluminosilicates, Chem. Mater. 17 (2005) 2976–2986.
799
[10] P. Duxson, J.L. Provis, G.C. Lukey, F. Separovic, J.S. van Deventer, 29Si NMR study of structural ordering in aluminosilicate geopolymer gels, Langmuir Acs J. Surf. Colloids 21 (2005) 3028. [11] V. Marry, P. Turq, T. Cartailler, D. Levesque, Microscopic simulation of structure and dynamics of water and counterions in a monohydrated montmorillonite, J. Chem. Phys. 117 (2002) 3454–3463. [12] D.J.K. Ross, R.M. Bustin, The importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs, Mar. Pet. Geol. 26 (2009) 916–927. [13] D.J.K. Ross, R.M. Bustin, Shale gas potential of the Lower Jurassic Gordondale Member, northeastern British Columbia, Canada, Bull. Can. Pet. Geol. 55 (2007) 51–75. [14] M. Chávezpáez, K.V. Workum, L.D. Pablo, J.J.D. Pablo, Monte Carlo simulations of Wyoming sodium montmorillonite hydrates, J. Chem. Phys. 114 (2001) 1405–1413. [15] P. Billemont, B. Coasne, W.G. De, An experimental and molecular simulation study of the adsorption of carbon dioxide and methane in nanoporous carbons in the presence of water, Langmuir Acs J. Surf. Colloids 27 (2011) 1015–1024. [16] Z. Jin, A. Firoozabadi, Effect of water on methane and carbon dioxide sorption in clay minerals by Monte Carlo simulations, Fluid Phase Equilib. 382 (2014) 10–20. [17] D.P. Cao, X.R. Zhang, J.F. Chen, W.C. Wang, J. Yun, Optimization of single-walled carbon nanotube arrays for methane storage at room temperature, J. Phys. Chem. B 107 (2003) 13286–13292. [18] J. Foerster, J. Lansford, J. Laskar, T. Rappaport, S. Kato, High uptakes of methane in Li-doped 3D covalent organic frameworks, Langmuir Acs J. Surf. Colloids 26 (2010) 220–226. [19] J. Hermans, H.J.C. Berendsen, W.F.V. Gunsteren, J.P.M. Postma, A consistent empirical potential for water–protein interactions, Biopolymers 23 (1984) 1513–1518. [20] M.R. Warne, N.L. Allan, T. Cosgrove, Computer simulation of water molecules at kaolinite and silica surfaces, Phys.chem.chem.phys 2 (2000) 3663–3668. [21] D.L. Bish, Rietveld refinement of non-hydrogen atomic positions in kaolinite, Clays Clay Miner. 37 (1989) 289–296. [22] G. Rutkai, T. Kristóf, Molecular simulation study of intercalation of small molecules in kaolinite, Chem. Phys. Lett. 462 (2008) 269–274. [23] M.C. He, J. Zhao, Effects of Mg, Ca, and Fe(II) doping on the kaolinite (0 0 1) surface with H2O adsorption, Clays Clay Miner. 60 (2012) 330–337. [24] X. Hua, F.Z. Wang, X.Q. Li, Preparation and characterization of kaolintepotassium acetate intercalation comple, J. Harbin Inst. Technol. 38 (2006) 126–129. [25] L. Zhang, C.J. Wang, Z.F. Yan, X.L. Wu, Y.Q. Wang, D.W. Meng, H.X. Xie, Kaolinite nanomaterial: Intercalation of 1-butyl-3-methylimidazolium bromine in a methanol–kaolinite pre-intercalate, Appl. Clay Sci. 86 (2013) 106–110. [26] S. Zhang, Q. Liu, H. Cheng, X. Li, F. Zeng, R.L. Frost, Intercalation of dodecylamine into kaolinite and its layering structure investigated by molecular dynamics simulation, J Colloid Interface Sci 430 (2014) 345–350. [27] B. Zhang, J. Kang, T. Kang, Monte Carlo simulations of methane adsorption on kaolinite as a function of pore size, J. Nat. Gas Sci. Eng. 49C (2018) 410–416. [28] S.L. Mayo, B.D. Olafson, W.A. Goddard, DREIDING: a generic force field for molecular simulations, J. Phys. Chem. 94 (1990) 8897–8909. [29] K.S. Smirnov, D. Bougeard, A molecular dynamics study of structure and shorttime dynamics of water in kaolinite, J. Phys. Chem. B 103 (1999) 5266–5273. [30] M.G. Martin, J.I. Siepmann, A molecular dynamics study of structure and shorttime dynamics of water in kaolinite, J. Phys. Chem. B 103 (1999) 5266–5273. [31] I.C. Yeh, M.L. Berkowitz, Ewald summation for systems with slab geometry, J. Chem. Phys. 111 (1999) 3155–3162. [32] P.S. Crozier, R.L. Rowley, E. Spohr, D. Henderson, Comparison of charged sheets and corrected 3D Ewald calculations of long-range forces in slab geometry electrolyte systems with solvent molecules, J. Chem. Phys. 112 (2000) 9253– 9257. [33] A. Hirotani, K. Mizukami, R. Miura, H. Takaba, T. Miya, A. Fahmi, A. Stirling, M. Kubo, A. Miyamoto, Grand canonical Monte Carlo simulation of the adsorption of CO2 on silicalite and NaZSM-5, Appl. Surf. Sci. 120 (1997) 81–84. [34] X.Q. Liu, X. He, N.X. Qiu, X. Yang, Z.Y. Tian, M.J. Li, Y. Xue, Molecular simulation of CH4, CO2, H2O and N2 molecules adsorption on heterogeneous surface models of coal, Appl. Surf. Sci. 389 (2016) 894–905. [35] A. Gotzias, H. Heiberg-Andersen, M. Kainourgiakis, T. Steriotis, Grand canonical Monte Carlo simulations of hydrogen adsorption in carbon cones, Appl. Surf. Sci. 256 (2010) 5226–5231. [36] Z. Xiang, D. Cao, Porous covalent-organic materials: synthesis, clean energy application and design, J. Mater. Chem. A 1 (2013) 2691–2718. [37] Z. Yang, D. Cao, Effect of Li doping on diffusion and separation of hydrogen and methane in covalent organic frameworks, J. Phys. Chem. C 116 (2012) 12591– 12598. [38] X. Accelrys, Accelrys. Materials Studio Release Notes, Release 6.0, Accelrys Software Inc, San Diego, 2010. [39] R.S. Shulyak, V.V. Primachenko, L.N. Zolotukhina, A.N. Zakovorotnyi, Investigating Chalganovsk kaolin deposits, Refractories 33 (1992) 323–325. [40] Q. Fang, S. Huang, W. Wang, Intercalation of dimethyl sulphoxide in kaolinite: molecular dynamics simulation study, Chem. Phys. Lett. 411 (2005) 233–237. [41] L. Ji, T. Zhang, K.L. Milliken, J. Qu, X. Zhang, Experimental investigation of main controls to methane adsorption in clay-rich rocks, Appl. Geochem. 27 (2012) 2533–2545. [42] B. Sun, S. Kayal, A. Chakraborty, Study of HKUST (Copper benzene-1,3,5tricarboxylate, Cu-BTC MOF)-1 metal organic frameworks for CH4 adsorption:
800
[43]
[44]
[45] [46] [47] [48] [49] [50]
[51]
B. Zhang et al. / Applied Surface Science 439 (2018) 792–800 an experimental investigation with GCMC (grand canonical Monte-carlo) simulation, Energy 76 (2014) 419–427. J. Zhang, M.B. Clennell, D.N. Dewhurst, K. Liu, Combined Monte Carlo and molecular dynamics simulation of methane adsorption on dry and moist coal, Fuel 122 (2014) 186–197. J. Liu, S. Keskin, D.S. Sholl, J.K. Johnson, Molecular simulations and theoretical predictions for adsorption and diffusion of CH4/H2 and CO2/CH4 mixtures in ZIFs, J. Phys. Chem. C 115 (2011) 12560–12566. C.X. Chen, C.H. Liu, A study on expanding characteristic of weathered granite, Key Eng. Mater. 261–263 (2004) 1391–1396. I. Barshad, Adsorptive and swelling properties of clay-water system, Clays Clay Miner. 1 (1952) 70–77. M.R. Abdi, S. Wild, Sulphate expansion of lime-stabilized kaolinite: I. Physical characteristics, Clay Miner. 28 (1993) 555–567. R. Heller, M. Zoback, Adsorption of methane and carbon dioxide on gas shale and pure mineral samples, J. Unconvention. Oil Gas Resour. 8 (2014) 14–24. D.Y. Peng, D.B. Robinson, A new two-constant equation of state, Ind. Eng. Chem. Fundam. 15 (1976) 92–94. F. Wu, C. Anutosh, Investigation of the interaction of polar molecules on graphite surface: prediction of isosteric heat of adsorption at zero surface coverage, J. Phys. Chem. C 120 (2016) 23490–23499. S. Namsani, Molecular simulation of shale gas adsorption and diffusion in inorganic nanopores, Mol. Simul. 41 (2015) 414–422.
[52] T. Zhang, G.S. Ellis, S.C. Ruppel, K. Milliken, R. Yang, Effect of organic-matter type and thermal maturity on methane adsorption in shale-gas systems, Org. Geochem. 47 (2012) 120–131. [53] J.S. Yang, C.L. Yang, M.S. Wang, B.D. Chen, X.G. Ma, Effect of functionalization on the interfacial binding energy of carbon nanotube/nylon 6 nanocomposites: a molecular dynamics study, Rsc Adv. 2 (2012) 2836–2841. [54] Q. Zheng, Q. Xue, K. Yan, X. Gao, Q. Li, L. Hao, Effect of chemisorption on the interfacial bonding characteristics of carbon nanotube–polymer composites, Polymer 49 (2008) 800–808. [55] C. Lv, Q. Xue, D. Xia, M. Ma, J. Xie, H. Chen, Effect of chemisorption on the interfacial bonding characteristics of graphenepolymer composites, J. Phys. Chem. C 114 (2010) 6588–6594. [56] L.A. García-Serrano, C.A. Flores-Sandoval, I.P. Zaragoza, Theoretical study of the adsorption of isobutane over H-mordenite zeolite by ab initio and DFT methods, J. Mol. Catal. A: Chem. 200 (2003) 205–212. [57] H. Pan, J.A. Ritter, P.B. Balbuena, Examination of the approximations used in determining the isosteric heat of adsorption from the ClausiusClapeyron equation, Langmuir 14 (2015) 535–542. [58] X.P. Kong, J. Wang, Copper (II) adsorption on the kaolinite (0 0 1) surface: insights from first-principles calculations and molecular dynamics simulations, Appl. Surf. Sci. 389 (2016) 316–323.