- Email: [email protected]
Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations
G Model
ARTICLE IN PRESS
SUPFLU-3377; No. of Pages 8
J. of Supercritical Fluids xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu
Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations Tiantian Wu a,b , Qingzhong Xue a,b,∗ , Xiaofang Li b , Yehan Tao b , Yakang Jin b , Cuicui Ling b , Shuangfang Lu c a
State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580, Shandong, PR China College of Science and Key Laboratory of New Energy Physics & Materials Science in Universities of Shandong, China University of Petroleum, Qingdao 266580, Shandong, PR China c Institute of Unconventional Oil & Gas and New Energy, China University of Petroleum, Qingdao 266580, Shandong, PR China b
a r t i c l e
i n f o
Article history: Received 6 February 2015 Received in revised form 3 July 2015 Accepted 4 July 2015 Available online xxx Keywords: Extraction process Supercritical extraction mechanisms Electronic interaction
a b s t r a c t The extraction process and mechanism of kerogen moieties with supercritical CO2 are elucidated using molecular dynamics simulations. It is demonstrated that supercritical CO2 can effectively dissolve the kerogen moieties adsorbed onto the shale surface, and the kerogen moieties dissolved in supercritical CO2 can be easily extracted from oil shale, because the interaction between the kerogen moieties dissolved in supercritical CO2 and the shale surface is greatly reduced. The dissolving capacity of supercritical CO2 is found to effectively increase with increasing pressure before the pressure reaches a critical value (approximately 50 MPa) and then increases slowly. Moreover, the dissolving capacity of supercritical CO2 increases with increasing temperature at high pressure, which is consistent with experimental results. In addition, the hydroxyl functional groups modified on the shale surface promote the extraction of kerogen moieties with supercritical CO2, and the polar kerogen moieties were more easily dissolved in supercritical CO2 . © 2015 Elsevier B.V. All rights reserved.
1. Introduction The worldwide potential amount of oil shale has been proved to be approximately 80 billion tons [1], which suggests that oil shale is one of the most important sources of liquid fuel after natural petroleum [2]. Using modern petroleum refining technologies, oil shale can be refined into marketable organic products [3]. Shale oil, called “artificial oil”, is produced from organic matter in oil shale, including saturates, aromatics, resins and asphaltenes, by using retorting or pyrolysis [1,2]. The major organic matter in oil shale is kerogen, which is a complex mixture of organic materials tightly adsorbed onto the shale surface and is difficult to be exploited. Therefore, the kerogen extraction from the shale surface is of crucial importance for effective oil extraction from oil shale. The traditional experimental approaches to extract kerogen from oil shale include differential wettability, sink-float separations, chemical methods and pyrolysis [1,4–8]. Differential
∗ Corresponding author at: State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580, Shandong, PR China. Fax: +86 0532 86981169. E-mail address: [email protected] (Q. Xue).
wettability methods are based on the principle of differential wetting of kerogen and minerals. However, the oil shale is required to be of fine particle sizes for a full wetting and successful future separation. Taking advantage of the differential gravity of kerogen and minerals, kerogen and minerals were separated using sink-float separation methods. Although kerogen is not altered chemically by the sink-float method, only low yields of highly enriched organic matters are obtained. Chemical methods, always result in major structural changes in kerogen by chemical agents and are also not reliable for separating kerogen from oil shale for characterization studies. Pyrolysis requires a large amount of energy and results in many harmful toxic compounds. Recently, supercritical fluid extraction (SFE), which uses gases such as carbon dioxide, ethane, propane, toluene and water under supercritical conditions, was reported to be competitive as a new extraction technology. A supercritical fluid has similar density and dissolving ability to a liquid, and has similar diffusivity, viscosity, and surface tension to a gas. Therefore, the great selectivity and rapid mass transfer of supercritical fluid make SFE an efficient, environmentally friendly, and attractive separation technique compared to conventional extraction methods [9,10]. Investigations of kerogen extraction from oil shale with supercritical fluids have been reported [11–14]. However, to our knowledge, the mechanism of SFE of kerogen from the
http://dx.doi.org/10.1016/j.supflu.2015.07.005 0896-8446/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model
ARTICLE IN PRESS
SUPFLU-3377; No. of Pages 8
T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
2
shale surface is still unclear, understanding this mechanism is very important to designing effective methods for oil extraction from oil shale. Supercritical CO2 has attracted considerable attention in many fields such as solvents, material syntheses, supercritical extraction and enhanced oil recovery [15–25]. The introduction of supercritical CO2 was found to be a key factor to encourage the dissolution in the water/granite system [26]. Supercritical CO2 can change the wetting of the surface of silica rock [27], increase oil saturation above the residual saturation, and reduce oil viscosity [21], which can lead to oil flow with better mobility. Moreover, supercritical CO2 has a stronger interaction with aromatic molecules, which has unique properties as an improved oil recovery fluid compared to other supercritical fluids such as N2 and CH4 [28]. Although supercritical CO2 is a nonpolar solvent, it still has good attraction to polar solutes and many large organic molecules due to its large quadrupole moment [3]; in addition, a wide variety of compounds could be extracted by supercritical CO2 [3,9,10,29,30]. Therefore, supercritical CO2 with a low critical point (Tc = 304.3 K and Pc = 7.38 MPa) [16], a high permeability, a great selectivity and a high dissolving ability has the advantage of being able to extract large organic molecules (such as kerogen). Even though many investigations of structures and models of kerogen have been reported [31–37], it is still difficult to investigate a complete kerogen. The kerogen moieties identified in kerogen have been used to evaluate the properties of kerogen [33,38]. In addition, these kerogen moieties are also products of oil shale, the extraction mechanism of which may be important for future oil extraction from oil shale. Therefore, in this study, we investigated the extraction of four representative kerogen moieties with supercritical CO2 using molecular dynamics (MD) simulations. The microscopic behavior of kerogen extraction was revealed using MD simulations, and the supercritical extraction mechanism was proposed. Moreover, the effects of pressure, temperature, spacing between the shale surfaces, functional groups modified on the shale surface and the polarity of kerogen moieties on the extraction of the kerogen moieties with supercritical CO2 were also investigated.
Ecross−term = + + +
b b
F
b
b
+
− 0
Fb b − b0
F − 0
ij
− 0
b − b0
(3)
× [F1 cos + F2 cos 2 + F3 cos 3]
× [V1 cos + V2 cos 2 + V3 cos 3]
K cos − 0
× − 0
Enon-bond =
Fb (b − b0 ) × [V1 cos + V2 cos 2 + V3 cos 3]
+
Fb (b − b0 ) − 0
b
+
Fbb (b − b0 ) b − b0
Aij r9ij
−
Bij r6ij
+ ij
qi qj erij
+ EH-bond
(4)
The valence energy, Evalence , is generally accounted for by terms of bond stretching, valence angle bending, dihedral angle torsion, and inversion. The cross-term interaction energy, Ecross−term , accounts for factors such as bond or angle distortions caused by nearby atoms to accurately reproduce the dynamic properties of molecules. The non-bonding interaction term, Enon−bond , accounts for the interactions between non-bonded atoms and results mainly from van der Waals (vdW) interactions. In Eqs. (1)–(4), q is the atomic charge, is the dielectric constant, and rij is the i-j atomic separation distance. b and b are the lengths of two adjacent bonds, is the two-bond angle, is the dihedral torsion angle, and is the out of plane angle. b0 , ki (i = 2 − 4), 0 , Hi (i = 2 − 4), i0 (i = 1 − 3), Vi (i = 1 − 3), Fbb , b0 , F , 0 , Fb , Fb , Fb , Fi (i = 1 − 3), F , K , Aij , and Bij are fitted from quantum mechanical calculations and are implemented into the Discover module of the Materials Studio. 2.1. Oil shale surface
2. Models and methods MD simulations are implemented by the DISCOVER code embedded in the Material Studio software developed by Accelrys Inc. The condensed-phase optimized molecular potential for atomistic simulation studies (COMPASS) module in the Material Studio software was used to conduct force-field computations to account for the interactions between atoms and molecules [39]. This is the first ab initio force field that is parametrized and validated using condensed-phase properties in addition to various ab initio and empirical data. The force field is expressed as a sum of valence (or bonding), cross-terms, and non-bonding interactions: Etotal = Evalence + Ecross−term + Enon−bond
Evalence = + + +
2
3
K2 (b − b0 ) + K3 (b − b0 ) + K4 (b − b0 )
b
H2 − 0
2
+ H3 − 0
3
(1)
4
+ H4 − 0
4
V1 1 − cos( − 10 )] + V2 [1 − cos(2 − 20 )] + V3 [1 − cos(3−30 )
Kx 2 + EUB
x
(2)
The most important inorganic constituents of oil shale are clay, calcite, and silica. Silica is the major mineral constituent in most shale formations. Therefore, the silica surface was investigated as oil shale surface in this work [40–42]. The initial silica lattice was derived from the database of the Material Studio software. The (100) silica surface was cleaved with a thickness of 1.2 nm. To investigate the effect of functional groups modified on the shale surface on the extraction of kerogen moieties with supercritical CO2 , the silica surfaces were modified with OH (strong polarity) and H (weak polarity). The structures of the modified silica surface are shown in Fig. S1. 2.2. Supercritical CO2 Carbon dioxide in the gas phase was built in a 281 × 281 × 281 Å3 box with 600 CO2 molecules. To obtain supercritical CO2 , we performed a constant-pressure/constant-temperature dynamics (NPT) simulation controlled by the Andersen thermostat method, with a fixed time step of 1 fs. Data were collected every 5 ps, and the fullprecision trajectory was then recorded. By changing the simulation pressure and temperature, the supercritical CO2 in the equilibrium state was finally obtained. Table S1 shows the density of supercritical CO2 at different temperatures and pressures. The densities of supercritical CO2 at 333.15 K and 373.15 K calculated in our simulation are quite similar to those in the database reported by Aimoli group [43].
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
3
Fig. 1. Molecular structures of (a) dodecane, (b) pyridine, (c) quinaldine, and (d) heptylamine.
2.3. Kerogen moieties The structures of kerogen moieties used in our work are shown in Fig. 1. To ensure a similar thickness of the layer of kerogen moieties and simultaneously fix the atom numbers, the numbers of dodecane, pyridine, quinaldine and heptylamine molecules used in our simulations are 50, 100, 50, and 100, respectively. 2.4. Extraction model According to the previously reported models [44], the extraction system is composed of the shale surfaces, the supercritical CO2 layer and the layer of kerogen moieties, with a total size of 35 × 36 × 177 Å3 , as shown in Fig. 2. The whole system was simulated using a constant-volume/constant-temperature dynamics (NVT) ensemble for 5 ns. The NVT dynamics simulation parameters were the same as those of the NPT simulation. Periodic boundary conditions were applied in the whole process. Subsequently, the results were analysed. 3. Results and discussion 3.1. The effects of pressure, temperature and functional group modification on the silica surface on extraction We first investigated dodecane extraction from the silica surface with supercritical CO2 at 373.15 K and 100 MPa. Fig. 3 shows the snapshots of dodecane extraction in the simulation from 0.25 ns to 5 ns. Initially, the dodecane molecules were tightly adsorbed onto the silica surface. First, several supercritical CO2 molecules permeated into the dodecane layer, and the dodecane molecules far away from the silica surface were first extracted to achieve a stronger
non-bonding energy between dodecane and supercritical CO2 compared to that between dodecane and silica surface. Next, increasing numbers of supercritical CO2 molecules permeated through the dodecane layer and reached the silica surface. At the same time, increasing numbers of dodecane molecules adsorbed onto the silica surface were displaced by the supercritical CO2 molecules. Finally, the volume of the ropy dodecane layer expanded, and then the dodecane molecules were extracted by supercritical CO2 . Most of the dodecane molecules were dissolved in supercritical CO2 fluids. Therefore, supercritical CO2 is capable of extracting a substantial amount of dodecane from the silica surface. Next, dodecane extractions using the supercritical CO2 at 10 MPa, 30 MPa and 50 MPa were investigated. The final results of the dynamics simulations are shown in Fig. 4. After a simulation time of 5 ns, only one dodecane molecule was dissolved by the supercritical solvent at 10 MPa, while many dodecane molecules were dispersed in the supercritical solvent at 30, 50 and 100 MPa. The dissolving capacity of supercritical CO2 is found to increase with increasing pressure (the quantitative analysis is presented in the Supplementary material), which may result from the decrease in the interfacial tension of the CO2 /dodecane system with increasing pressure [36] and the enhancement of the miscibility of dodecane-supercritical CO2 . Therefore, more supercritical CO2 molecules could permeate through the dodecane layer and reach the silica surface, and more dodecane molecules on the silica surface were replaced by supercritical CO2 molecules. In addition, the higher the pressure is, the greater is the fluid density, as indicated in Table S1(c). Thus, our result indicates that the dissolving capacity of supercritical CO2 increases with increasing density, which is consistent with Chrastil’s model, in which the dissolving capacity is positively related to the solvent density [45]. However, when the pressure increased to 100 MPa, the dissolving capacity
Fig. 2. The initial model of the extraction system.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8 4
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
Fig. 3. Snapshots of dodecane extraction from the silica surface using supercritical CO2 at 100 MPa in the simulation from 0.25 ns to 5 ns.
of supercritical CO2 increased slowly, which may be attributed to an increase of viscosity and a decrease in the permeability of supercritical CO2 at high pressure. In addition, extraction at a high pressure requires a large amount of energy. Therefore, the pressure range from 30 MPa to 50 MPa may be suitable for economically effective oil extraction. Moreover, the dodecane extractions at 333.15 K and 353.15 K were also investigated. The final configurations (t = 5 ns) of dodecane extraction by supercritical CO2 at temperatures of 333.15 K, 353.15 K and 373.15 K are shown in Fig. S2. The results show that the dissolving capacity of supercritical CO2 increases with increasing temperature at high pressure, which is consistent with previously reported results [14,46]. Therefore, our MD results indicate that the dissolving capacity of supercritical CO2 increases with increasing pressure and temperature [14]. Furthermore, the final (t = 5 ns) concentration profiles of dodecane extraction at 100 MPa and 10 MPa along the X direction were calculated to investigate the detailed behavior of dodecane extraction, as shown in Fig. 5. The horizontal axis is on behalf of the distribution of dodecane along the X direction, and the vertical axis represents the relative concentration of dodecane. Taking the concentration profiles at 10 MPa as an example, we set the hydrogenated silica surface at 15 Å along the X direction and the hydroxylated silica surface at 162 Å. The dodecane concentration decreases most rapidly at 45 Å and 135 Å, which correspond to the dodecane-fluid interfaces. Therefore, the areas of 15–35 Å and 135–162 Å are the ranges of the dodecane layers with very high concentration of dodecane. The area from 45 Å to 135 Å is the range of the supercritical CO2 layer with low concentrations of dodecane, which indicates low dodecane extraction by the supercritical CO2 at 10 MPa. However, the dodecane concentration is lower in the dodecane layer and higher in the supercritical CO2
layer at 100 MPa, which indicates a higher dodecane extraction. These results demonstrate that supercritical CO2 at a high pressure has a stronger ability to dissolve dodecane, which is consistent with the above MD results. In addition, the concentration of dodecane adsorbed onto the hydroxylated silica surface is lower than that adsorbed onto the hydrogenated silica surface at 100 MPa, which shows that fewer residual dodecane molecules are on the hydroxylated silica surface. Moreover, the total concentration of dodecane extracted from the hydroxylated silica surface is higher than that extracted from the hydrogenated silica surface, which indicates that more dodecane molecules are extracted from the hydroxylated silica surface. Because there is an additional hydrogen bond interaction between CO2 and the hydroxylated silica surface [47], which increases the interaction between CO2 and the hydroxylated silica surface, more CO2 molecules diffuse through the dodecane layer and are then adsorbed onto the hydroxylated silica surface, and finally, more dodecane molecules are displaced by the supercritical CO2 . Therefore, our results indicate that the silica surface modified with polar functional groups (such as hydroxyl groups) may promote dodecane extraction with supercritical CO2 . 3.2. The effect of the polarity of kerogen moieties on extraction Supercritical CO2 at 100 MPa and 373.15 K was used to extract polar kerogen moieties. The snapshots of pyridine extraction from the silica surface with supercritical CO2 in the MD simulation from 0.25 ns to 5 ns are shown in Fig. 6. The other snapshots of quinaldine and heptylamine are shown in Figs. S3 and S4. The extraction process of pyridine with supercritical CO2 is similar to that of dodecane. Supercritical CO2 molecules first diffused and permeated through
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
5
Fig. 4. The final configurations (t = 5 ns) of dodecane extraction using supercritical CO2 at 10 MPa, 30 MPa, 50 MPa and 100 MPa.
the pyridine layer, and then the pyridine molecules adsorbed onto the silica surface were displaced by supercritical CO2 ; finally, the volume of the ropy pyridine layer expanded, and the pyridine molecules were extracted by supercritical CO2 . However, compared to the dodecane molecules, many more pyridine molecules were extracted from the silica surface into the supercritical CO2 layer at the very beginning of the simulation time (t = 0.25 ns). Therefore, supercritical CO2 is capable of effective extraction of polar kerogen moieties from the shale surface. In addition, we investigated the effect of the spacing between the two rock surfaces on pyridine
Fig. 5. The final (t = 5 ns) concentration profiles of the dodecane extraction at pressures of 100 MPa and 10 MPa along the X direction.
extraction with supercritical CO2 at 373.15 K and 100 MPa. The final molecular dynamics results (t = 5 ns) are shown in Fig. 7. When the spacing is 52, 76, 96 and 146 Å, pyridine is found to be dissolved by the supercritical CO2 . There is little difference among the extraction yields of the extraction systems with spacings of 76, 96 and 146 Å. When the spacing is too small (such as 52 Å), a lack of supercritical CO2 molecules result in a large number of residual kerogen moieties on the rock surfaces not being dissolved, thereby decreasing the extraction yield. Therefore, supercritical CO2 is capable of effective extraction of polar kerogen moieties from the shale surface (with spacing lager than 52 Å). It has been reported that the apolar oil molecules could not permeate through the water layer [48] leading to a hydrophilic rock surface, while the polar oil molecules could permeate through the water layer and be adsorbed to the rock surface, resulting in an oleophilic rock surface [41]. These results indicated that the interaction between polar oil molecules and the rock surface was higher than that between water molecules and the rock surface. However, our results showed that the supercritical CO2 molecules could permeate through the polar oil layer, consisting of molecules such as pyridine, quinaldine and heptylamine, and reach the silica surface; as a result, the silica surface became hydrophilic again, which indicates that the interactions between oil molecules and the rock surface were weakened by the injection of supercritical CO2 . Our finding is consistent with the experimental results reported recently [49]. Therefore, the injection of supercritical CO2 is able to decrease the interaction between kerogen moieties and the shale surface, thereby further changing the wettability of the shale surface, which has profound implications for studies of the wettability changes of oil shale and improved oil recovery.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8 6
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
Fig. 6. Snapshots of pyridine extraction from the silica surface using supercritical CO2 in the MD simulation from 0.25 ns to 5 ns.
Moreover, we found that the calculated interaction energies between the kerogen moieties and the silica (at the initial and final simulation time) mostly consisted of electrostatic interaction energies, as shown in Table S2 and Fig. 8, which indicated that the electrostatic interaction played a significant role in the total kerogenshale interaction. These primarily electrostatic interactions
have also been found in the Na-montmorillonite system [38]. To investigate the effect of the polarity of kerogen moieties on the extraction of the kerogen moieties with supercritical CO2 , the dipole moments of the four molecules were calculated by the first-principle simulation; the details of the simulation have been reported in our previous work [50–52]. The dipole moments of
Fig. 7. Final configurations (t = 5 ns) of pyridine extraction using supercritical CO2 at 373.15 K and 100 MPa with a spacing of (a) 52 Å, (b) 76 Å, (c) 96 Å and (d) 146 Å.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
7
Fig. 9. The final (t = 5 ns) concentration profiles of dodecane and pyridine along the X direction. Fig. 8. The interaction energies (total and electrostatic) between the different kerogen moieties and silica at the initial simulation time (t = 0) and at the final simulation time (t = 5 ns).
dodecane, heptylamine, quinaldine and pyridine molecules are 0.0002, 1.2643, 1.7851 and 2.2882 Debye, respectively. The dipole moment of the pyridine calculated in our simulations is close to the values (2.20 ∼ 2.36 Debye) reported experimentally and theoretically [53,54]. According to the dipole moments of the four molecules, the order of the polarity of the molecules is pyridine > quinaldine > heptylamine > dodecane. Dodecane is seen as an apolar molecule due to their very small dipole moment, and the others are seen as polar molecules for relatively large dipole moments. The interaction between the apolar molecules (dodecane) and the silica is weaker than that between the polar kerogen moieties (pyridine, quinaldine and heptylamine) and the silica at both the initial and final simulation times, as shown in Fig. 8, because the polarity of the kerogen moieties can enhance the kerogen–silica interaction. In addition, these interaction energies were attractive at both the initial and final simulation times, and the attractive interaction between the kerogen moieties and the silica was reduced at the final simulation time because of the injection of CO2 , which led to the ease of extraction of the kerogen moieties adsorbed on the silica surface into the supercritical CO2 fluids. Moreover, the electronic kerogen moieties-shale interaction is defined as qi qj rij ,
Eelectronic =
where i and j are random atoms of kerogen
i,j
molecules and shale, respectively, qi and qj are the atomic charges of atoms i and j, respectively, is the dielectric constant, and rij is the i-j atomic separation distance. In other words, the electronic interaction is inversely proportional to the atomic separation distance (r ij ). The kerogen moieties are far away from the shale surface after dissolving in the supercritical CO2 , and the atomic separation distance (rij ) increases. Therefore, the electronic kerogen moietiesshale interaction is greatly reduced. In addition, the electrostatic interaction played a significant role in the total interaction between the kerogen moieties and the silica. Thus, the total non-bond interactions between the kerogen moieties and silica were reduced as shown in Fig. 8. Moreover, there was a much greater loss of the interaction between the polar kerogen moieties and the silica surface, which means that the polar kerogen moieties-silica interaction is much further reduced and that supercritical CO2 extraction is much more effective for polar kerogen moiety extraction from the silica surface compared to the extraction of dodecane. Furthermore, the final concentration profiles (t = 5 ns) of pyridine and dodecane along the X direction were calculated to compare the detailed behaviour of the extraction of polar kerogen moieties and the extraction of apolar kerogen moieties, as shown in
Fig. 9. The concentration of pyridine in the supercritical CO2 layer is found to be higher than that of dodecane due to the much more reduced interaction between the polar kerogen moieties and the silica surface (as analysed above). However, there are two sharp peaks of the concentration curve of pyridine, which means that a very thin layer of the pyridine remained on the silica surface caused by the strong interaction between the polar kerogen moieties (very close to the surface) and the silica surface. In addition, the peak of pyridine concentration at the hydroxylated surface is much higher than that at the hydrogenated surface because the interaction between the polar kerogen moieties and the polar group-modified surface is much stronger than that between the polar kerogen moieties and the apolar group modified surface. Therefore, it may be a slightly difficult for the remaining polar oil to be extracted from the polar silica surface. 4. Conclusions In summary, by using molecular dynamics simulations, the supercritical extraction process and mechanism of representative kerogen moieties with supercritical CO2 are elucidated, as are the effect of other influential factors, such as pressure, temperature, spacing between the shale surfaces, functional groups modified on the shale surface and polarity of kerogen moieties. Supercritical CO2 is found to be capable of extracting a substantial amount of kerogen moieties from the shale surface. In addition, hydroxyl functional groups modified on the shale surface promote the extraction of kerogen moieties with supercritical CO2. It is also found that the dissolving capacity of supercritical CO2 effectively increases with increasing pressure before the pressure reaches a critical value (approximately 50 MPa) and then increases slowly and that the dissolving capacity of the supercritical CO2 increases with increasing temperature at high pressure. To conclude, all of these good extractions from the shale surface result from the decrease of the electronic interaction of kerogen-shale after the injection of supercritical CO2 , which further changes the wettability of the shale surface and has profound implications for studies of the wettability changes of oil shale and the future exploitation of oil shale. Acknowledgements This work is supported by the Natural Science Foundation of China (41330313), the Taishan Scholar Foundation (ts20130929), the Fundamental Research Funds for the Central Universities (13CX05009A, 14CX05013A, and 15CX08009A), and the Graduate Innovation Fund of China University of Petroleum (YCX2014070 and YCX2015057).
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
8
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.supflu.2015.07. 005 References [1] B. Hascakir, T. Babadagli, S. Akin, Experimental and numerical simulation of oil recovery from oil shales by electrical heating, Energy Fuels 22 (2008) 3976–3985. [2] S.H. Deng, Z.J. Wang, Q. Gu, F.Y. Meng, J.F. Li, H.Y. Wang, Extraction hydrocarbons from Huadian oil shale by sub-critical water, Fuel Process. Technol. 92 (2011) 1062–1067. [3] W.E. Rudzinski, T.M. Aminabhavi, A review on extraction and identification of crude oil and related products using supercritical fluid technology, Energy Fuels 14 (2000) 464–475. [4] J.W. Smith, L.W. Higby, Preparation of organic concentrate from Green River oil shale, Anal. Chem. 32 (1960) 1718. [5] J. Reisberg, The beneficiation of Green River oil shale by pelletization. In Oil Shale, Tar Sands, and Related Materials 1981, pp. 155. [6] R.A. Ibrahimov, K.K. Bissada, Comparative analysis and geological significance of kerogen isolated using open-system (palynological) versus chemically and volumetrically conservative closed-system methods, Org. Geochem. 41 (2010) 800–811. [7] R. Gupta, D. Gidaspow, D.T. Wasan, Electrostatic beneficiation of eastern oil shales, Chem. Eng. Commun. 108 (1991) 49–66. [8] J.G. Na, C.H. Im, S.H. Chung, K.B. Lee, Effect of oil shale retorting temperature on shale oil yield and properties, Fuel 95 (2012) 131–135. [9] A. Sing, M. Canel, Comparison of retoring and supercritical techniques on gaining liquid products from Goynuk oil shale (Turkey), Energy Sources 26 (2004) 739–749. [10] J.D. Tucker, B. Massri, S.A. Lee, Comparison of retorting and supercritical extraction techniques on El-Lajjun oil shale, Energy Sources 22 (2000) 453–463. [11] H. Hu, J. Zhang, S. Guo, G. Chen, Extraction of Huadian oil shale with water in sub- and supercritical states, Fuel 78 (1999) 645–651. [12] M.L. G-Hourcade, C. Torrente, M.A. Galán, Study of the solubility of kerogen from oil shales (Puertollano, Spain) in supercritical toluene and methanol, Fuel 86 (2007) 698–705. [13] M.C. Torrente, M.A. Galán, Extraction of kerogen from oil shale (Puertollano, Spain) with supercritical toluene and methanol mixtures, Ind. Eng. Chem. Res. 50 (2011) 1730–1738. [14] A. Mamdouh, A. Awni, A. Hussein, et al., CO2 spupercritical fluid extraction of Jordanian oil shale utilizing different co-solvents, Fuel Process. Technol. 92 (2011) 2016–2023. [15] C. Song, CO2 conversion and utilization: An overview, ACS Symp. Ser. 809 (2002) 2–30. [16] E. Reverchon, I. De Marco, Supercritical fluid extraction and fractionation of natural matter, J. Supercrit. Fluids 38 (2006) 146–166. [17] E. Sabio, M. Lozano, V. Montero de Espinosa, R.L. Mendes, et al., Lycopene and â-carotene extraction from tomato processing waste using supercritical CO2 , Ind. Eng. Chem. Res. 42 (2003) 6641–6646. [18] G. Panfili, L. Cinquanta, A. Fratianni, R. Cubadda, Extraction of wheat germ oil by supercritical CO2 : oil and defatted cake characterization, J. Am. Oil Chem. Soc. 80 (2003) 157–161. [19] J. Min, S. Li, J. Hao, N. Liu, Supercritical CO2 extraction of jatropha oil and solubility correlation, J. Chem. Eng. Data 55 (2010) 3755–3758. [20] E. Conde, A. Moure, H. Dominguez, Supercritical CO2 extraction of fatty acids, phenolics and fucoxanthin from freeze–dried sargassum muticum, J. Appl. Phycol. (2014). [21] M. Godec, V. Kuuskraa, V. Van Leeuwen, L.S. Melzer, N. Wildgust, CO2 storage in depleted oil fields: the worldwide potential for carbon dioxide enhanced oil recovery, Energy Procedia 4 (2011) 2162–2169. [22] S. Thomas, Enhanced oil recovery-an overview, Oil Gas Sci. Technol.-Rev IFP 63 (2008) 9–19. [23] P.L. Bondor, Applications of carbon dioxide in enhanced oil recovery, Energy Convers. Manage. 33 (1992) 579–586. [24] D.D. Jia, S.F. Li, L. Xiao, Supercritical CO2 extraction of plumula nelumbinis oil: experiments and modeling, J. Supercrit. Fluids 50 (2009) 229–234. [25] J.M. del Valle, Extraction of natural compounds using supercritical CO2 : going from the laboratory to the industrial application, J. Supercrit. Fluids 96 (2015) 180–199. [26] H. Lin, T. Fujii, B. Takisawa, T. Takahashi, T. Hashida, Experimental evaluation of interactions in supercritical CO2 /water/rock minerals system under geologic CO2 sequestration conditions, J. Mater. Sci. 43 (2008) 2307–2315.
[27] J.W. Jung, J. Wan, Supercritical CO2 and ionic strength effects on wettability of silica surfaces: equilibrium contact angle measurements, Energy Fuels 26 (2012) 6053–6059. [28] L.S. Lara, M.F. Michelon, C.R. Miranda, Molecular dynamics studies of fluid/oil interfaces for improved oil recovery process, J. Phys. Chem. B 116 (2012) 14667–14676. [29] M. Koel, S. Ljovin, Y. Bondar, Supercritical carbon dioxide extraction of estonian oil shale, Oil shale 17 (2000) 225–232. [30] N. Olukcu, J. Yanik, M. Saglam, M. Yüksel, M. Karaduman, Solvent effect on the extraction of Beypazari oil shale, Energy Fuels 13 (1999) 895–902. [31] M. Vandenbroucke, C. Largeau, Kerogen origin, evolution and structure, Org. Geochem. 38 (2007) 719–833. [32] M.O. Anita, S.O.P. Ian, R.B. Shyam, S.S. Mark, et al., Three-dimensional structure of the siskin green river oil shale kerogen model: a comparison between calculated and observed properties, Energy Fuels 27 (2013) 702–710. [33] M. Siskin, G. Brons, J.F. Payack, Disruption of kerogen mineral interactions in oil shales, Energy Fuels 1 (1987) 248–252. [34] M. Siskin, G. Brons, J. Payack, Disruption of kerogen-mineral interactions in rundle ramsay crossing oil-shale, Energy Fuels 3 (1989) 108–109. [35] J.W. Larsen, S. Li, Solvent swelling studies of green river kerogen, Energy Fuels 8 (1994) 932–936. [36] S.R. Kelemen, M. Afeworki, M.L. Gorbaty, M. Sansone, P.J. Kwiatek, et al., Direct characterization of kerogen by X-ray and solid-state 13 C nuclear magnetic resonance methods, Energy Fuels 21 (2007) 1548–1561. [37] W.K. Albert, B. Fancoise, G.H. Patrick, Quantitative analysis of long chain fatty acids presentin a type I kerogen using electrospray ionization fouriertransform ion cyclotron resonance mass spectrometry: compared with BF3 /MeOH methylation/GC-FID, J. Am. Soc. Mass Spectrom. 25 (2014) 880–890. [38] D.R. Katti, B. Upadhyay, K.S. Katti, Molecular interactions of kerogen moieties with Na-montmorillonite: an experimental and modeling study, Fuel 130 (2014) 34–45. [39] H. Sun, COMPASS: an ab initio force-field optimized for condensed-phase applications-overview with details on alkane and benzene compounds, J. Phys. Chem. B 102 (1998) 338–7364. [40] S.H. Garofalini, Molecular dynamics computer simulations of silica surface structure and adsorption of water molecules, J. Non-Cryst. Solids 120 (1990) 1–12. [41] N.R. Tummala, L. Shi, A. Striolo, Melocular dynamics simulations of surfactants at the silica–water interface: anionic vs nonionic headgroups, J. Colloid Interface Sci. 362 (2011) 135–143. [42] Q. Liu, S. Yuan, H. Yan, X. Zhao, Mechanism of oil detachment from a silica surface in aqueous surfactant solutions: molecular dynamics simulations, J. Phys. Chem. B 11 (2012) 2867–2875. [43] C.G. Aimoli, E.J. Maginn, C.R.A. Abreu, Forec filed comparison and thermodynamic property calculation of supercritical CO2 and CH4 using molecular dynamics simulations, Fluid Phase Equilib. 368 (2014) 80–90. [44] A.C.T. Duin, S.R. Larter, A computational chemical study of penetration and displacement of water films near mineral surface, Geochem. Trans. 6 (2001). [45] J. Chrastil, Solubility of solids and liquids in supercritical gases, J. Phys. Chem. 86 (1982) 3016–3021. [46] S. Zhao, An experimental investigation into the solubility of Moringa oleifera oil in supercritical carbon dioxide, J. Food Eng. 138 (2014) 1–10. [47] Y. Qin, X. Yan, Y. Zhu, J. Ping, Molecular dynamics simulation of interaction between supercritical CO2 fluid and modified silica surfaces, J. Phys. Chem. C 112 (2008) 12815–12824. [48] M. Salehi, S.J. Johnson, J.T. Liang, Mechanistic study of wettability alteration using surfactants with applications in naturally fractured reservoirs, Langmuir 24 (2008) 14099–14107. [49] Y. Kim, J. Wan, T.J. Kneafsey, T.K. Tokunaga, Dewetting of silica surfaces upon reactions with supercritical CO2 and brine: pore-scale studies in micromodels, Environ. Sci. Technol. 46 (2012) 4228–4235. [50] T. Zhang, Q.Z. Xue, M.X. Shan, Z.Y. Jiao, X.Y. Zhou, C.C. Ling, Z.F. Yan, Adsorption and catalytic activation of O2 molecule on the surface of Au-doped graphene under an external electric field, J. Phys. Chem. C 116 (2012) 19918–19924. [51] Z.L. Liu, Q.Z. Xue, T. Zhang, C.C. Ling, M.X. Shan, Carbon doping of hexagonal boron nitride by using CO molecules, J. Phys. Chem. C 117 (2013) 9332–9339. [52] T.T. Wu, Q.Z. Xue, C.C. Ling, M.X. Shan, Z.L. Liu, Y.T. Tao, X.F. Li, Fluorine-modified porous graphene as membrane for CO2 /N2 separation: molecular dynamic and first principles simulations, J. Phys. Chem. C 118 (2014) 7369–7376. [53] P. Löwdin, Calculation of electric dipole moments of some heterocyclics, J. Phys. Chem. 19 (1951) 1323. [54] H. Soscún, O. Castellano, Y. Bermúdez, C. Toro-Mendoza, et al., Linear and nonlinear optical properties of pyridine N-oxide molecule, J. Mol. Struct. THEOCHEM 592 (2002) 19–28.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
ARTICLE IN PRESS
SUPFLU-3377; No. of Pages 8
J. of Supercritical Fluids xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu
Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations Tiantian Wu a,b , Qingzhong Xue a,b,∗ , Xiaofang Li b , Yehan Tao b , Yakang Jin b , Cuicui Ling b , Shuangfang Lu c a
State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580, Shandong, PR China College of Science and Key Laboratory of New Energy Physics & Materials Science in Universities of Shandong, China University of Petroleum, Qingdao 266580, Shandong, PR China c Institute of Unconventional Oil & Gas and New Energy, China University of Petroleum, Qingdao 266580, Shandong, PR China b
a r t i c l e
i n f o
Article history: Received 6 February 2015 Received in revised form 3 July 2015 Accepted 4 July 2015 Available online xxx Keywords: Extraction process Supercritical extraction mechanisms Electronic interaction
a b s t r a c t The extraction process and mechanism of kerogen moieties with supercritical CO2 are elucidated using molecular dynamics simulations. It is demonstrated that supercritical CO2 can effectively dissolve the kerogen moieties adsorbed onto the shale surface, and the kerogen moieties dissolved in supercritical CO2 can be easily extracted from oil shale, because the interaction between the kerogen moieties dissolved in supercritical CO2 and the shale surface is greatly reduced. The dissolving capacity of supercritical CO2 is found to effectively increase with increasing pressure before the pressure reaches a critical value (approximately 50 MPa) and then increases slowly. Moreover, the dissolving capacity of supercritical CO2 increases with increasing temperature at high pressure, which is consistent with experimental results. In addition, the hydroxyl functional groups modified on the shale surface promote the extraction of kerogen moieties with supercritical CO2, and the polar kerogen moieties were more easily dissolved in supercritical CO2 . © 2015 Elsevier B.V. All rights reserved.
1. Introduction The worldwide potential amount of oil shale has been proved to be approximately 80 billion tons [1], which suggests that oil shale is one of the most important sources of liquid fuel after natural petroleum [2]. Using modern petroleum refining technologies, oil shale can be refined into marketable organic products [3]. Shale oil, called “artificial oil”, is produced from organic matter in oil shale, including saturates, aromatics, resins and asphaltenes, by using retorting or pyrolysis [1,2]. The major organic matter in oil shale is kerogen, which is a complex mixture of organic materials tightly adsorbed onto the shale surface and is difficult to be exploited. Therefore, the kerogen extraction from the shale surface is of crucial importance for effective oil extraction from oil shale. The traditional experimental approaches to extract kerogen from oil shale include differential wettability, sink-float separations, chemical methods and pyrolysis [1,4–8]. Differential
∗ Corresponding author at: State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580, Shandong, PR China. Fax: +86 0532 86981169. E-mail address: [email protected] (Q. Xue).
wettability methods are based on the principle of differential wetting of kerogen and minerals. However, the oil shale is required to be of fine particle sizes for a full wetting and successful future separation. Taking advantage of the differential gravity of kerogen and minerals, kerogen and minerals were separated using sink-float separation methods. Although kerogen is not altered chemically by the sink-float method, only low yields of highly enriched organic matters are obtained. Chemical methods, always result in major structural changes in kerogen by chemical agents and are also not reliable for separating kerogen from oil shale for characterization studies. Pyrolysis requires a large amount of energy and results in many harmful toxic compounds. Recently, supercritical fluid extraction (SFE), which uses gases such as carbon dioxide, ethane, propane, toluene and water under supercritical conditions, was reported to be competitive as a new extraction technology. A supercritical fluid has similar density and dissolving ability to a liquid, and has similar diffusivity, viscosity, and surface tension to a gas. Therefore, the great selectivity and rapid mass transfer of supercritical fluid make SFE an efficient, environmentally friendly, and attractive separation technique compared to conventional extraction methods [9,10]. Investigations of kerogen extraction from oil shale with supercritical fluids have been reported [11–14]. However, to our knowledge, the mechanism of SFE of kerogen from the
http://dx.doi.org/10.1016/j.supflu.2015.07.005 0896-8446/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model
ARTICLE IN PRESS
SUPFLU-3377; No. of Pages 8
T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
2
shale surface is still unclear, understanding this mechanism is very important to designing effective methods for oil extraction from oil shale. Supercritical CO2 has attracted considerable attention in many fields such as solvents, material syntheses, supercritical extraction and enhanced oil recovery [15–25]. The introduction of supercritical CO2 was found to be a key factor to encourage the dissolution in the water/granite system [26]. Supercritical CO2 can change the wetting of the surface of silica rock [27], increase oil saturation above the residual saturation, and reduce oil viscosity [21], which can lead to oil flow with better mobility. Moreover, supercritical CO2 has a stronger interaction with aromatic molecules, which has unique properties as an improved oil recovery fluid compared to other supercritical fluids such as N2 and CH4 [28]. Although supercritical CO2 is a nonpolar solvent, it still has good attraction to polar solutes and many large organic molecules due to its large quadrupole moment [3]; in addition, a wide variety of compounds could be extracted by supercritical CO2 [3,9,10,29,30]. Therefore, supercritical CO2 with a low critical point (Tc = 304.3 K and Pc = 7.38 MPa) [16], a high permeability, a great selectivity and a high dissolving ability has the advantage of being able to extract large organic molecules (such as kerogen). Even though many investigations of structures and models of kerogen have been reported [31–37], it is still difficult to investigate a complete kerogen. The kerogen moieties identified in kerogen have been used to evaluate the properties of kerogen [33,38]. In addition, these kerogen moieties are also products of oil shale, the extraction mechanism of which may be important for future oil extraction from oil shale. Therefore, in this study, we investigated the extraction of four representative kerogen moieties with supercritical CO2 using molecular dynamics (MD) simulations. The microscopic behavior of kerogen extraction was revealed using MD simulations, and the supercritical extraction mechanism was proposed. Moreover, the effects of pressure, temperature, spacing between the shale surfaces, functional groups modified on the shale surface and the polarity of kerogen moieties on the extraction of the kerogen moieties with supercritical CO2 were also investigated.
Ecross−term = + + +
b b
F
b
b
+
− 0
Fb b − b0
F − 0
ij
− 0
b − b0
(3)
× [F1 cos + F2 cos 2 + F3 cos 3]
× [V1 cos + V2 cos 2 + V3 cos 3]
K cos − 0
× − 0
Enon-bond =
Fb (b − b0 ) × [V1 cos + V2 cos 2 + V3 cos 3]
+
Fb (b − b0 ) − 0
b
+
Fbb (b − b0 ) b − b0
Aij r9ij
−
Bij r6ij
+ ij
qi qj erij
+ EH-bond
(4)
The valence energy, Evalence , is generally accounted for by terms of bond stretching, valence angle bending, dihedral angle torsion, and inversion. The cross-term interaction energy, Ecross−term , accounts for factors such as bond or angle distortions caused by nearby atoms to accurately reproduce the dynamic properties of molecules. The non-bonding interaction term, Enon−bond , accounts for the interactions between non-bonded atoms and results mainly from van der Waals (vdW) interactions. In Eqs. (1)–(4), q is the atomic charge, is the dielectric constant, and rij is the i-j atomic separation distance. b and b are the lengths of two adjacent bonds, is the two-bond angle, is the dihedral torsion angle, and is the out of plane angle. b0 , ki (i = 2 − 4), 0 , Hi (i = 2 − 4), i0 (i = 1 − 3), Vi (i = 1 − 3), Fbb , b0 , F , 0 , Fb , Fb , Fb , Fi (i = 1 − 3), F , K , Aij , and Bij are fitted from quantum mechanical calculations and are implemented into the Discover module of the Materials Studio. 2.1. Oil shale surface
2. Models and methods MD simulations are implemented by the DISCOVER code embedded in the Material Studio software developed by Accelrys Inc. The condensed-phase optimized molecular potential for atomistic simulation studies (COMPASS) module in the Material Studio software was used to conduct force-field computations to account for the interactions between atoms and molecules [39]. This is the first ab initio force field that is parametrized and validated using condensed-phase properties in addition to various ab initio and empirical data. The force field is expressed as a sum of valence (or bonding), cross-terms, and non-bonding interactions: Etotal = Evalence + Ecross−term + Enon−bond
Evalence = + + +
2
3
K2 (b − b0 ) + K3 (b − b0 ) + K4 (b − b0 )
b
H2 − 0
2
+ H3 − 0
3
(1)
4
+ H4 − 0
4
V1 1 − cos( − 10 )] + V2 [1 − cos(2 − 20 )] + V3 [1 − cos(3−30 )
Kx 2 + EUB
x
(2)
The most important inorganic constituents of oil shale are clay, calcite, and silica. Silica is the major mineral constituent in most shale formations. Therefore, the silica surface was investigated as oil shale surface in this work [40–42]. The initial silica lattice was derived from the database of the Material Studio software. The (100) silica surface was cleaved with a thickness of 1.2 nm. To investigate the effect of functional groups modified on the shale surface on the extraction of kerogen moieties with supercritical CO2 , the silica surfaces were modified with OH (strong polarity) and H (weak polarity). The structures of the modified silica surface are shown in Fig. S1. 2.2. Supercritical CO2 Carbon dioxide in the gas phase was built in a 281 × 281 × 281 Å3 box with 600 CO2 molecules. To obtain supercritical CO2 , we performed a constant-pressure/constant-temperature dynamics (NPT) simulation controlled by the Andersen thermostat method, with a fixed time step of 1 fs. Data were collected every 5 ps, and the fullprecision trajectory was then recorded. By changing the simulation pressure and temperature, the supercritical CO2 in the equilibrium state was finally obtained. Table S1 shows the density of supercritical CO2 at different temperatures and pressures. The densities of supercritical CO2 at 333.15 K and 373.15 K calculated in our simulation are quite similar to those in the database reported by Aimoli group [43].
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
3
Fig. 1. Molecular structures of (a) dodecane, (b) pyridine, (c) quinaldine, and (d) heptylamine.
2.3. Kerogen moieties The structures of kerogen moieties used in our work are shown in Fig. 1. To ensure a similar thickness of the layer of kerogen moieties and simultaneously fix the atom numbers, the numbers of dodecane, pyridine, quinaldine and heptylamine molecules used in our simulations are 50, 100, 50, and 100, respectively. 2.4. Extraction model According to the previously reported models [44], the extraction system is composed of the shale surfaces, the supercritical CO2 layer and the layer of kerogen moieties, with a total size of 35 × 36 × 177 Å3 , as shown in Fig. 2. The whole system was simulated using a constant-volume/constant-temperature dynamics (NVT) ensemble for 5 ns. The NVT dynamics simulation parameters were the same as those of the NPT simulation. Periodic boundary conditions were applied in the whole process. Subsequently, the results were analysed. 3. Results and discussion 3.1. The effects of pressure, temperature and functional group modification on the silica surface on extraction We first investigated dodecane extraction from the silica surface with supercritical CO2 at 373.15 K and 100 MPa. Fig. 3 shows the snapshots of dodecane extraction in the simulation from 0.25 ns to 5 ns. Initially, the dodecane molecules were tightly adsorbed onto the silica surface. First, several supercritical CO2 molecules permeated into the dodecane layer, and the dodecane molecules far away from the silica surface were first extracted to achieve a stronger
non-bonding energy between dodecane and supercritical CO2 compared to that between dodecane and silica surface. Next, increasing numbers of supercritical CO2 molecules permeated through the dodecane layer and reached the silica surface. At the same time, increasing numbers of dodecane molecules adsorbed onto the silica surface were displaced by the supercritical CO2 molecules. Finally, the volume of the ropy dodecane layer expanded, and then the dodecane molecules were extracted by supercritical CO2 . Most of the dodecane molecules were dissolved in supercritical CO2 fluids. Therefore, supercritical CO2 is capable of extracting a substantial amount of dodecane from the silica surface. Next, dodecane extractions using the supercritical CO2 at 10 MPa, 30 MPa and 50 MPa were investigated. The final results of the dynamics simulations are shown in Fig. 4. After a simulation time of 5 ns, only one dodecane molecule was dissolved by the supercritical solvent at 10 MPa, while many dodecane molecules were dispersed in the supercritical solvent at 30, 50 and 100 MPa. The dissolving capacity of supercritical CO2 is found to increase with increasing pressure (the quantitative analysis is presented in the Supplementary material), which may result from the decrease in the interfacial tension of the CO2 /dodecane system with increasing pressure [36] and the enhancement of the miscibility of dodecane-supercritical CO2 . Therefore, more supercritical CO2 molecules could permeate through the dodecane layer and reach the silica surface, and more dodecane molecules on the silica surface were replaced by supercritical CO2 molecules. In addition, the higher the pressure is, the greater is the fluid density, as indicated in Table S1(c). Thus, our result indicates that the dissolving capacity of supercritical CO2 increases with increasing density, which is consistent with Chrastil’s model, in which the dissolving capacity is positively related to the solvent density [45]. However, when the pressure increased to 100 MPa, the dissolving capacity
Fig. 2. The initial model of the extraction system.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8 4
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
Fig. 3. Snapshots of dodecane extraction from the silica surface using supercritical CO2 at 100 MPa in the simulation from 0.25 ns to 5 ns.
of supercritical CO2 increased slowly, which may be attributed to an increase of viscosity and a decrease in the permeability of supercritical CO2 at high pressure. In addition, extraction at a high pressure requires a large amount of energy. Therefore, the pressure range from 30 MPa to 50 MPa may be suitable for economically effective oil extraction. Moreover, the dodecane extractions at 333.15 K and 353.15 K were also investigated. The final configurations (t = 5 ns) of dodecane extraction by supercritical CO2 at temperatures of 333.15 K, 353.15 K and 373.15 K are shown in Fig. S2. The results show that the dissolving capacity of supercritical CO2 increases with increasing temperature at high pressure, which is consistent with previously reported results [14,46]. Therefore, our MD results indicate that the dissolving capacity of supercritical CO2 increases with increasing pressure and temperature [14]. Furthermore, the final (t = 5 ns) concentration profiles of dodecane extraction at 100 MPa and 10 MPa along the X direction were calculated to investigate the detailed behavior of dodecane extraction, as shown in Fig. 5. The horizontal axis is on behalf of the distribution of dodecane along the X direction, and the vertical axis represents the relative concentration of dodecane. Taking the concentration profiles at 10 MPa as an example, we set the hydrogenated silica surface at 15 Å along the X direction and the hydroxylated silica surface at 162 Å. The dodecane concentration decreases most rapidly at 45 Å and 135 Å, which correspond to the dodecane-fluid interfaces. Therefore, the areas of 15–35 Å and 135–162 Å are the ranges of the dodecane layers with very high concentration of dodecane. The area from 45 Å to 135 Å is the range of the supercritical CO2 layer with low concentrations of dodecane, which indicates low dodecane extraction by the supercritical CO2 at 10 MPa. However, the dodecane concentration is lower in the dodecane layer and higher in the supercritical CO2
layer at 100 MPa, which indicates a higher dodecane extraction. These results demonstrate that supercritical CO2 at a high pressure has a stronger ability to dissolve dodecane, which is consistent with the above MD results. In addition, the concentration of dodecane adsorbed onto the hydroxylated silica surface is lower than that adsorbed onto the hydrogenated silica surface at 100 MPa, which shows that fewer residual dodecane molecules are on the hydroxylated silica surface. Moreover, the total concentration of dodecane extracted from the hydroxylated silica surface is higher than that extracted from the hydrogenated silica surface, which indicates that more dodecane molecules are extracted from the hydroxylated silica surface. Because there is an additional hydrogen bond interaction between CO2 and the hydroxylated silica surface [47], which increases the interaction between CO2 and the hydroxylated silica surface, more CO2 molecules diffuse through the dodecane layer and are then adsorbed onto the hydroxylated silica surface, and finally, more dodecane molecules are displaced by the supercritical CO2 . Therefore, our results indicate that the silica surface modified with polar functional groups (such as hydroxyl groups) may promote dodecane extraction with supercritical CO2 . 3.2. The effect of the polarity of kerogen moieties on extraction Supercritical CO2 at 100 MPa and 373.15 K was used to extract polar kerogen moieties. The snapshots of pyridine extraction from the silica surface with supercritical CO2 in the MD simulation from 0.25 ns to 5 ns are shown in Fig. 6. The other snapshots of quinaldine and heptylamine are shown in Figs. S3 and S4. The extraction process of pyridine with supercritical CO2 is similar to that of dodecane. Supercritical CO2 molecules first diffused and permeated through
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
5
Fig. 4. The final configurations (t = 5 ns) of dodecane extraction using supercritical CO2 at 10 MPa, 30 MPa, 50 MPa and 100 MPa.
the pyridine layer, and then the pyridine molecules adsorbed onto the silica surface were displaced by supercritical CO2 ; finally, the volume of the ropy pyridine layer expanded, and the pyridine molecules were extracted by supercritical CO2 . However, compared to the dodecane molecules, many more pyridine molecules were extracted from the silica surface into the supercritical CO2 layer at the very beginning of the simulation time (t = 0.25 ns). Therefore, supercritical CO2 is capable of effective extraction of polar kerogen moieties from the shale surface. In addition, we investigated the effect of the spacing between the two rock surfaces on pyridine
Fig. 5. The final (t = 5 ns) concentration profiles of the dodecane extraction at pressures of 100 MPa and 10 MPa along the X direction.
extraction with supercritical CO2 at 373.15 K and 100 MPa. The final molecular dynamics results (t = 5 ns) are shown in Fig. 7. When the spacing is 52, 76, 96 and 146 Å, pyridine is found to be dissolved by the supercritical CO2 . There is little difference among the extraction yields of the extraction systems with spacings of 76, 96 and 146 Å. When the spacing is too small (such as 52 Å), a lack of supercritical CO2 molecules result in a large number of residual kerogen moieties on the rock surfaces not being dissolved, thereby decreasing the extraction yield. Therefore, supercritical CO2 is capable of effective extraction of polar kerogen moieties from the shale surface (with spacing lager than 52 Å). It has been reported that the apolar oil molecules could not permeate through the water layer [48] leading to a hydrophilic rock surface, while the polar oil molecules could permeate through the water layer and be adsorbed to the rock surface, resulting in an oleophilic rock surface [41]. These results indicated that the interaction between polar oil molecules and the rock surface was higher than that between water molecules and the rock surface. However, our results showed that the supercritical CO2 molecules could permeate through the polar oil layer, consisting of molecules such as pyridine, quinaldine and heptylamine, and reach the silica surface; as a result, the silica surface became hydrophilic again, which indicates that the interactions between oil molecules and the rock surface were weakened by the injection of supercritical CO2 . Our finding is consistent with the experimental results reported recently [49]. Therefore, the injection of supercritical CO2 is able to decrease the interaction between kerogen moieties and the shale surface, thereby further changing the wettability of the shale surface, which has profound implications for studies of the wettability changes of oil shale and improved oil recovery.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8 6
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
Fig. 6. Snapshots of pyridine extraction from the silica surface using supercritical CO2 in the MD simulation from 0.25 ns to 5 ns.
Moreover, we found that the calculated interaction energies between the kerogen moieties and the silica (at the initial and final simulation time) mostly consisted of electrostatic interaction energies, as shown in Table S2 and Fig. 8, which indicated that the electrostatic interaction played a significant role in the total kerogenshale interaction. These primarily electrostatic interactions
have also been found in the Na-montmorillonite system [38]. To investigate the effect of the polarity of kerogen moieties on the extraction of the kerogen moieties with supercritical CO2 , the dipole moments of the four molecules were calculated by the first-principle simulation; the details of the simulation have been reported in our previous work [50–52]. The dipole moments of
Fig. 7. Final configurations (t = 5 ns) of pyridine extraction using supercritical CO2 at 373.15 K and 100 MPa with a spacing of (a) 52 Å, (b) 76 Å, (c) 96 Å and (d) 146 Å.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
7
Fig. 9. The final (t = 5 ns) concentration profiles of dodecane and pyridine along the X direction. Fig. 8. The interaction energies (total and electrostatic) between the different kerogen moieties and silica at the initial simulation time (t = 0) and at the final simulation time (t = 5 ns).
dodecane, heptylamine, quinaldine and pyridine molecules are 0.0002, 1.2643, 1.7851 and 2.2882 Debye, respectively. The dipole moment of the pyridine calculated in our simulations is close to the values (2.20 ∼ 2.36 Debye) reported experimentally and theoretically [53,54]. According to the dipole moments of the four molecules, the order of the polarity of the molecules is pyridine > quinaldine > heptylamine > dodecane. Dodecane is seen as an apolar molecule due to their very small dipole moment, and the others are seen as polar molecules for relatively large dipole moments. The interaction between the apolar molecules (dodecane) and the silica is weaker than that between the polar kerogen moieties (pyridine, quinaldine and heptylamine) and the silica at both the initial and final simulation times, as shown in Fig. 8, because the polarity of the kerogen moieties can enhance the kerogen–silica interaction. In addition, these interaction energies were attractive at both the initial and final simulation times, and the attractive interaction between the kerogen moieties and the silica was reduced at the final simulation time because of the injection of CO2 , which led to the ease of extraction of the kerogen moieties adsorbed on the silica surface into the supercritical CO2 fluids. Moreover, the electronic kerogen moieties-shale interaction is defined as qi qj rij ,
Eelectronic =
where i and j are random atoms of kerogen
i,j
molecules and shale, respectively, qi and qj are the atomic charges of atoms i and j, respectively, is the dielectric constant, and rij is the i-j atomic separation distance. In other words, the electronic interaction is inversely proportional to the atomic separation distance (r ij ). The kerogen moieties are far away from the shale surface after dissolving in the supercritical CO2 , and the atomic separation distance (rij ) increases. Therefore, the electronic kerogen moietiesshale interaction is greatly reduced. In addition, the electrostatic interaction played a significant role in the total interaction between the kerogen moieties and the silica. Thus, the total non-bond interactions between the kerogen moieties and silica were reduced as shown in Fig. 8. Moreover, there was a much greater loss of the interaction between the polar kerogen moieties and the silica surface, which means that the polar kerogen moieties-silica interaction is much further reduced and that supercritical CO2 extraction is much more effective for polar kerogen moiety extraction from the silica surface compared to the extraction of dodecane. Furthermore, the final concentration profiles (t = 5 ns) of pyridine and dodecane along the X direction were calculated to compare the detailed behaviour of the extraction of polar kerogen moieties and the extraction of apolar kerogen moieties, as shown in
Fig. 9. The concentration of pyridine in the supercritical CO2 layer is found to be higher than that of dodecane due to the much more reduced interaction between the polar kerogen moieties and the silica surface (as analysed above). However, there are two sharp peaks of the concentration curve of pyridine, which means that a very thin layer of the pyridine remained on the silica surface caused by the strong interaction between the polar kerogen moieties (very close to the surface) and the silica surface. In addition, the peak of pyridine concentration at the hydroxylated surface is much higher than that at the hydrogenated surface because the interaction between the polar kerogen moieties and the polar group-modified surface is much stronger than that between the polar kerogen moieties and the apolar group modified surface. Therefore, it may be a slightly difficult for the remaining polar oil to be extracted from the polar silica surface. 4. Conclusions In summary, by using molecular dynamics simulations, the supercritical extraction process and mechanism of representative kerogen moieties with supercritical CO2 are elucidated, as are the effect of other influential factors, such as pressure, temperature, spacing between the shale surfaces, functional groups modified on the shale surface and polarity of kerogen moieties. Supercritical CO2 is found to be capable of extracting a substantial amount of kerogen moieties from the shale surface. In addition, hydroxyl functional groups modified on the shale surface promote the extraction of kerogen moieties with supercritical CO2. It is also found that the dissolving capacity of supercritical CO2 effectively increases with increasing pressure before the pressure reaches a critical value (approximately 50 MPa) and then increases slowly and that the dissolving capacity of the supercritical CO2 increases with increasing temperature at high pressure. To conclude, all of these good extractions from the shale surface result from the decrease of the electronic interaction of kerogen-shale after the injection of supercritical CO2 , which further changes the wettability of the shale surface and has profound implications for studies of the wettability changes of oil shale and the future exploitation of oil shale. Acknowledgements This work is supported by the Natural Science Foundation of China (41330313), the Taishan Scholar Foundation (ts20130929), the Fundamental Research Funds for the Central Universities (13CX05009A, 14CX05013A, and 15CX08009A), and the Graduate Innovation Fund of China University of Petroleum (YCX2014070 and YCX2015057).
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005
G Model SUPFLU-3377; No. of Pages 8
ARTICLE IN PRESS T. Wu et al. / J. of Supercritical Fluids xxx (2015) xxx–xxx
8
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.supflu.2015.07. 005 References [1] B. Hascakir, T. Babadagli, S. Akin, Experimental and numerical simulation of oil recovery from oil shales by electrical heating, Energy Fuels 22 (2008) 3976–3985. [2] S.H. Deng, Z.J. Wang, Q. Gu, F.Y. Meng, J.F. Li, H.Y. Wang, Extraction hydrocarbons from Huadian oil shale by sub-critical water, Fuel Process. Technol. 92 (2011) 1062–1067. [3] W.E. Rudzinski, T.M. Aminabhavi, A review on extraction and identification of crude oil and related products using supercritical fluid technology, Energy Fuels 14 (2000) 464–475. [4] J.W. Smith, L.W. Higby, Preparation of organic concentrate from Green River oil shale, Anal. Chem. 32 (1960) 1718. [5] J. Reisberg, The beneficiation of Green River oil shale by pelletization. In Oil Shale, Tar Sands, and Related Materials 1981, pp. 155. [6] R.A. Ibrahimov, K.K. Bissada, Comparative analysis and geological significance of kerogen isolated using open-system (palynological) versus chemically and volumetrically conservative closed-system methods, Org. Geochem. 41 (2010) 800–811. [7] R. Gupta, D. Gidaspow, D.T. Wasan, Electrostatic beneficiation of eastern oil shales, Chem. Eng. Commun. 108 (1991) 49–66. [8] J.G. Na, C.H. Im, S.H. Chung, K.B. Lee, Effect of oil shale retorting temperature on shale oil yield and properties, Fuel 95 (2012) 131–135. [9] A. Sing, M. Canel, Comparison of retoring and supercritical techniques on gaining liquid products from Goynuk oil shale (Turkey), Energy Sources 26 (2004) 739–749. [10] J.D. Tucker, B. Massri, S.A. Lee, Comparison of retorting and supercritical extraction techniques on El-Lajjun oil shale, Energy Sources 22 (2000) 453–463. [11] H. Hu, J. Zhang, S. Guo, G. Chen, Extraction of Huadian oil shale with water in sub- and supercritical states, Fuel 78 (1999) 645–651. [12] M.L. G-Hourcade, C. Torrente, M.A. Galán, Study of the solubility of kerogen from oil shales (Puertollano, Spain) in supercritical toluene and methanol, Fuel 86 (2007) 698–705. [13] M.C. Torrente, M.A. Galán, Extraction of kerogen from oil shale (Puertollano, Spain) with supercritical toluene and methanol mixtures, Ind. Eng. Chem. Res. 50 (2011) 1730–1738. [14] A. Mamdouh, A. Awni, A. Hussein, et al., CO2 spupercritical fluid extraction of Jordanian oil shale utilizing different co-solvents, Fuel Process. Technol. 92 (2011) 2016–2023. [15] C. Song, CO2 conversion and utilization: An overview, ACS Symp. Ser. 809 (2002) 2–30. [16] E. Reverchon, I. De Marco, Supercritical fluid extraction and fractionation of natural matter, J. Supercrit. Fluids 38 (2006) 146–166. [17] E. Sabio, M. Lozano, V. Montero de Espinosa, R.L. Mendes, et al., Lycopene and â-carotene extraction from tomato processing waste using supercritical CO2 , Ind. Eng. Chem. Res. 42 (2003) 6641–6646. [18] G. Panfili, L. Cinquanta, A. Fratianni, R. Cubadda, Extraction of wheat germ oil by supercritical CO2 : oil and defatted cake characterization, J. Am. Oil Chem. Soc. 80 (2003) 157–161. [19] J. Min, S. Li, J. Hao, N. Liu, Supercritical CO2 extraction of jatropha oil and solubility correlation, J. Chem. Eng. Data 55 (2010) 3755–3758. [20] E. Conde, A. Moure, H. Dominguez, Supercritical CO2 extraction of fatty acids, phenolics and fucoxanthin from freeze–dried sargassum muticum, J. Appl. Phycol. (2014). [21] M. Godec, V. Kuuskraa, V. Van Leeuwen, L.S. Melzer, N. Wildgust, CO2 storage in depleted oil fields: the worldwide potential for carbon dioxide enhanced oil recovery, Energy Procedia 4 (2011) 2162–2169. [22] S. Thomas, Enhanced oil recovery-an overview, Oil Gas Sci. Technol.-Rev IFP 63 (2008) 9–19. [23] P.L. Bondor, Applications of carbon dioxide in enhanced oil recovery, Energy Convers. Manage. 33 (1992) 579–586. [24] D.D. Jia, S.F. Li, L. Xiao, Supercritical CO2 extraction of plumula nelumbinis oil: experiments and modeling, J. Supercrit. Fluids 50 (2009) 229–234. [25] J.M. del Valle, Extraction of natural compounds using supercritical CO2 : going from the laboratory to the industrial application, J. Supercrit. Fluids 96 (2015) 180–199. [26] H. Lin, T. Fujii, B. Takisawa, T. Takahashi, T. Hashida, Experimental evaluation of interactions in supercritical CO2 /water/rock minerals system under geologic CO2 sequestration conditions, J. Mater. Sci. 43 (2008) 2307–2315.
[27] J.W. Jung, J. Wan, Supercritical CO2 and ionic strength effects on wettability of silica surfaces: equilibrium contact angle measurements, Energy Fuels 26 (2012) 6053–6059. [28] L.S. Lara, M.F. Michelon, C.R. Miranda, Molecular dynamics studies of fluid/oil interfaces for improved oil recovery process, J. Phys. Chem. B 116 (2012) 14667–14676. [29] M. Koel, S. Ljovin, Y. Bondar, Supercritical carbon dioxide extraction of estonian oil shale, Oil shale 17 (2000) 225–232. [30] N. Olukcu, J. Yanik, M. Saglam, M. Yüksel, M. Karaduman, Solvent effect on the extraction of Beypazari oil shale, Energy Fuels 13 (1999) 895–902. [31] M. Vandenbroucke, C. Largeau, Kerogen origin, evolution and structure, Org. Geochem. 38 (2007) 719–833. [32] M.O. Anita, S.O.P. Ian, R.B. Shyam, S.S. Mark, et al., Three-dimensional structure of the siskin green river oil shale kerogen model: a comparison between calculated and observed properties, Energy Fuels 27 (2013) 702–710. [33] M. Siskin, G. Brons, J.F. Payack, Disruption of kerogen mineral interactions in oil shales, Energy Fuels 1 (1987) 248–252. [34] M. Siskin, G. Brons, J. Payack, Disruption of kerogen-mineral interactions in rundle ramsay crossing oil-shale, Energy Fuels 3 (1989) 108–109. [35] J.W. Larsen, S. Li, Solvent swelling studies of green river kerogen, Energy Fuels 8 (1994) 932–936. [36] S.R. Kelemen, M. Afeworki, M.L. Gorbaty, M. Sansone, P.J. Kwiatek, et al., Direct characterization of kerogen by X-ray and solid-state 13 C nuclear magnetic resonance methods, Energy Fuels 21 (2007) 1548–1561. [37] W.K. Albert, B. Fancoise, G.H. Patrick, Quantitative analysis of long chain fatty acids presentin a type I kerogen using electrospray ionization fouriertransform ion cyclotron resonance mass spectrometry: compared with BF3 /MeOH methylation/GC-FID, J. Am. Soc. Mass Spectrom. 25 (2014) 880–890. [38] D.R. Katti, B. Upadhyay, K.S. Katti, Molecular interactions of kerogen moieties with Na-montmorillonite: an experimental and modeling study, Fuel 130 (2014) 34–45. [39] H. Sun, COMPASS: an ab initio force-field optimized for condensed-phase applications-overview with details on alkane and benzene compounds, J. Phys. Chem. B 102 (1998) 338–7364. [40] S.H. Garofalini, Molecular dynamics computer simulations of silica surface structure and adsorption of water molecules, J. Non-Cryst. Solids 120 (1990) 1–12. [41] N.R. Tummala, L. Shi, A. Striolo, Melocular dynamics simulations of surfactants at the silica–water interface: anionic vs nonionic headgroups, J. Colloid Interface Sci. 362 (2011) 135–143. [42] Q. Liu, S. Yuan, H. Yan, X. Zhao, Mechanism of oil detachment from a silica surface in aqueous surfactant solutions: molecular dynamics simulations, J. Phys. Chem. B 11 (2012) 2867–2875. [43] C.G. Aimoli, E.J. Maginn, C.R.A. Abreu, Forec filed comparison and thermodynamic property calculation of supercritical CO2 and CH4 using molecular dynamics simulations, Fluid Phase Equilib. 368 (2014) 80–90. [44] A.C.T. Duin, S.R. Larter, A computational chemical study of penetration and displacement of water films near mineral surface, Geochem. Trans. 6 (2001). [45] J. Chrastil, Solubility of solids and liquids in supercritical gases, J. Phys. Chem. 86 (1982) 3016–3021. [46] S. Zhao, An experimental investigation into the solubility of Moringa oleifera oil in supercritical carbon dioxide, J. Food Eng. 138 (2014) 1–10. [47] Y. Qin, X. Yan, Y. Zhu, J. Ping, Molecular dynamics simulation of interaction between supercritical CO2 fluid and modified silica surfaces, J. Phys. Chem. C 112 (2008) 12815–12824. [48] M. Salehi, S.J. Johnson, J.T. Liang, Mechanistic study of wettability alteration using surfactants with applications in naturally fractured reservoirs, Langmuir 24 (2008) 14099–14107. [49] Y. Kim, J. Wan, T.J. Kneafsey, T.K. Tokunaga, Dewetting of silica surfaces upon reactions with supercritical CO2 and brine: pore-scale studies in micromodels, Environ. Sci. Technol. 46 (2012) 4228–4235. [50] T. Zhang, Q.Z. Xue, M.X. Shan, Z.Y. Jiao, X.Y. Zhou, C.C. Ling, Z.F. Yan, Adsorption and catalytic activation of O2 molecule on the surface of Au-doped graphene under an external electric field, J. Phys. Chem. C 116 (2012) 19918–19924. [51] Z.L. Liu, Q.Z. Xue, T. Zhang, C.C. Ling, M.X. Shan, Carbon doping of hexagonal boron nitride by using CO molecules, J. Phys. Chem. C 117 (2013) 9332–9339. [52] T.T. Wu, Q.Z. Xue, C.C. Ling, M.X. Shan, Z.L. Liu, Y.T. Tao, X.F. Li, Fluorine-modified porous graphene as membrane for CO2 /N2 separation: molecular dynamic and first principles simulations, J. Phys. Chem. C 118 (2014) 7369–7376. [53] P. Löwdin, Calculation of electric dipole moments of some heterocyclics, J. Phys. Chem. 19 (1951) 1323. [54] H. Soscún, O. Castellano, Y. Bermúdez, C. Toro-Mendoza, et al., Linear and nonlinear optical properties of pyridine N-oxide molecule, J. Mol. Struct. THEOCHEM 592 (2002) 19–28.
Please cite this article in press as: T. Wu, et al., Extraction of kerogen from oil shale with supercritical carbon dioxide: Molecular dynamics simulations, J. Supercrit. Fluids (2015), http://dx.doi.org/10.1016/j.supflu.2015.07.005