Methane Molecules Drive Water Molecules along Diameter-Gradient

Jun 14, 2010 - Methane Molecules Drive Water Molecules along Diameter-Gradient SWCNTs with Junctions ... (SWCNTs) of different diameters with junction...
0 downloads 0 Views 5MB Size
8676

J. Phys. Chem. B 2010, 114, 8676–8679

Methane Molecules Drive Water Molecules along Diameter-Gradient SWCNTs with Junctions H. Q. Yu,† Y. F. Li,† H. Li,*,† K. Zhang,† C. G. An,† X. F. Liu,† and K. M. Liew‡ Key Laboratory for Liquid-Solid Structural EVolution and Processing of Materials, Ministry of Education, Shandong UniVersity, Jinan 250061, China, and Department of Building and Construction, City UniVersity of Hong Kong, Kowloon, Hong Kong ReceiVed: March 29, 2010; ReVised Manuscript ReceiVed: May 23, 2010

We report the transport behavior of water molecules along a system of coaxial single-walled carbon nanotubes (SWCNTs) of different diameters with junctions under the driving force of methane molecules. The junctions are potential barriers to the transport of water molecules through SWCNTs. However, methane molecules can overcome these potential barriers and pull the water molecules across the junction region from one compartment to the next. Although a junction is an obstacle to water transport through SWCNTs, the presence of more junctions gives methane molecules a longer lasting driving force that helps them to pull the water molecules out of the SWCNTs. 1. Introduction Carbon nanotubes (CNTs) have elicited great research interest in recent years because of their wide application in molecular sieves, membranes, sensors, and nanoconveyors for the precise delivery of gases or liquids.1-12 In particular, the transport behavior of water through SWCNTs has attracted great attention because this system is very similar to that of biological membrane channels. Hummer et al.1 used classical molecular dynamics (MD) simulations to prove that the diffusion of water through small-diameter tubes (8.1 Å) follows a burst-like mechanism that stems from the presence of single-file water chains capable of moving with little resistance, which was further validated with a simple coarse-grained model of a nanotube immersed in water.13 Fast mass transport was also reported by Holt14 and Majumder,15 who revealed that the water transport rate through a CNT is more than 3 orders of magnitude faster than conventional hydrodynamic flow. However, these experimental results were questioned by Thomas,16 who used MD simulation to reassess fast water transport through CNTs by observing pressure-driven water flow through the nanotubes and concluded that the water flow enhancement was less than that previously reported.14,15 Thermal gradients imposed on the SWCNTs in the axial direction have also been used to drive water nanodroplets through single- and double-walled CNTs.17 However, with the aforementioned methods, the imposition of a high pressure or temperature gradient on the systems makes the physical and chemical properties of the water change (possibly even resulting in phase transition), which affects the accuracy of the research results. The application of an electric current to the nanotube is an effective method of driving water flow through SWCNTs,18 but is difficult to perform because the apparatus required is complicated and the water molecules may become electrolyzed by the electric current. Moreover, water usually travels through many passages connected by junctions of different diameters, rather than single diameter tubes, which can cause barriers to water transport. Although * Corresponding author. E-mail address: [email protected]. † Shandong University. ‡ City University of Hong Kong.

much work focuses on the transport of water confined in a single-diameter SWCNT,19,20 little research has been conducted on the transport behavior of water molecules in a diametergradient SWCNT with many junctions. The manipulation of the flow of water molecules through diameter-gradient SWCNTs connected by junctions with different diameters is a challenging issue and has not been well tackled. In this study, MD simulation is performed to investigate how methane molecules pull water molecules through a diametergradient SWCNT with junctions. The model is highly appropriate for studying the influence of junctions on the transport of water through SWCNTs with different diameters. Understanding the mechanism of water transport in these carbon-nanotubebased membranes is crucial in the design of nanomachinery and synthetic membranes, and for the development of novel drugdelivery devices. 2. Computational Details All simulations are performed using the classical molecular dynamics method. The COMPASS force field is adopted to model the interaction between molecules in systems.21 Each system consists of coaxial carbon nanotubes with different diameters, which is linked by a junction region with a pentagon-heptagon (5-7) topological structure, as illustrated in Figure 1. Such connections have been directly observed by Iijima and others with transmission electronmicroscope (TEM).22,23 The nanotubes are zigzag (n, 0) SWCNTs with indices n ) 12, 13, 14, 15, 16, and 17, the diameters of which are 9.39, 10.18, 10.96, 11.74, 12.53, and 13.31 Å, respectively. Both bottlelike and terrace-like SWCNTs are used in the simulation. Figure 1 shows the bottle-like SWCNTs, which comprise a (12, 0) SWCNT connected to (13, 0), (14, 0), and (15, 0) SWCNTs, respectively. Figure 2 shows the terrace-like SWCNTs, in which (12, 0), (13, 0), (14, 0), and (15, 0) SWCNTs are connected with smooth junctions formed by five- and seven-membered rings to form a (12, 0) - (13, 0) - (14, 0) - (15, 0) seriesconnected SWCNT system. In the same way, other seriesconnected SWCNT systems are also fabricated successfully. The length of all the series-connected SWCNT systems is 83.6 Å. The coordinates of the SWCNTs are fixed. The quantity of

10.1021/jp102810j  2010 American Chemical Society Published on Web 06/14/2010

Water Molecules along SWCNTs

Figure 1. Snapshots of the initial configurations of three bottle-like SWCNT systems (a) a (12, 0) SWCNT connected with (13, 0), (b) a (12, 0) SWCNT connected with (14, 0), and (c) a (12, 0) SWCNT connected with (15, 0) SWCNT. The SWCNTs are gray, the hydrogen atoms are represented by white spheres, the oxygen atoms are in red, and the carbon atoms of methane are blue.

J. Phys. Chem. B, Vol. 114, No. 26, 2010 8677

Figure 3. Snapshots of water molecules dragged by methane molecules in three bottle-like systems at t ) 6000 ps obtained by MD simulation.

Figure 4. Evolution curves of the potential energy of the three systems with time t. Figure 2. Snapshots of the initial configurations of three terrace-like SWCNT systems (a) (12, 0) - (13, 0) - (14, 0) - (15, 0), (b) (12, 0) (14, 0) - (15, 0) - (16, 0), and (c) (12, 0) - (15, 0) - (16, 0) - (17, 0).

methane molecules is chosen as 50. The cutoff distance is set to 10.0 Å, and the time step is 1 fs. Data are recorded every 0.5 ps for further analysis. Initially, 20 water molecules are inserted randomly into the interior of the (12, 0) SWCNT, and 5000 time steps are run to relax the initial configuration to obtain the lowest energy structure. Finally, the methane molecules are arranged in another nanotube connected with a (12, 0) SWCNT to be used to drag the water molecules from the inside of the (12, 0) SWCNT. 3. Results and Discussion Figure 3 shows the translocation displacement L of water molecules dragged by methane molecules through three bottlelike SWCNTs at 6000 ps. Figure 3c demonstrates that the displacement distance L of water molecules in the (12, 0) - (15, 0) system is the longest and differs from the behavior of the system in the (12, 0) - (13, 0) SWCNT shown in Figure 3a. Obviously, the motion of the water molecules is attributed to the methane molecules that generate the driving force to make the water molecules flow through the SWCNT. Because methane molecules confined in the SWCNT are not at the lowest energy status, they continuously diffuse with an initial acceleration. Subsequently, the water molecules will follow the methane molecules through the SWCNT due to the van der Waals attraction between the methane molecules and the water molecules. The methane molecules are responsible for the driving force, and therefore the translocation of the water molecules is dominated by the methane molecules.

When the water molecules reach a junction region, they oscillate due to the potential barrier of the junction, which counteracts the attractive force of the methane molecules. If the attractive force of the methane molecules overcomes the potential barrier caused by the junction, then the methane molecules are able to drag the water molecules to another compartment; otherwise, they become jammed in the junction region. Figure 4 shows the evolution of the potential energy in the three bottle-like systems with time. It is worthy of note that the magnitudes of the potential energy of the three systems decrease as the diameter of the SWCNTs increases. This is because the water molecules are more stable inside the bigger SWCNT than they are inside the smaller SWCNT, as demonstrated in previous work.24 The potential energy barriers in the (12, 0) - (13, 0) SWCNT, the (12, 0) - (14, 0) SWCNT, and the (12, 0) - (15, 0) SWCNT are 75, 100, and 168 kJ, respectively, which indicates that a slight change in the diameter of a junction causes significant variation in the potential energy. In the (12, 0) - (15, 0) system, it takes the water molecules 19 330 ps to cross the junction region, whereas in the (12, 0) - (13, 0) SWCNT, it takes them about 20 550 ps to cross the junction region. The systems tend to equilibrate once the water molecules have crossed the junction region. When the methane molecules leave the interior of the SWCNTs, the attractive interaction between the methane molecules and water molecules weakens. Finally, the methane-water attractive interaction is equal to that of SWCNT-water and SWCNT-methane, which means that the whole system reaches equilibrium. Hence the number of the methane molecules in the SWCNT also affects the transport behavior of the water molecules through the nanotube. More methane molecules means a larger driving force, and therefore it is easier for the water molecules to cross the junction region. As mentioned in Figure 4, the potential energy curves display

8678

J. Phys. Chem. B, Vol. 114, No. 26, 2010

Yu et al.

Figure 5. Sequential snapshots of water molecules drawn by methane molecules in the three systems at time t ) 20 640 ps.

Figure 7. Simulation snapshots of water molecules in three bottlelike SWCNTs without methane molecules at t ) 20 640 ps.

Figure 6. Translocation displacement L(t) in the axial direction as a function of time t of water molecules dragged by methane molecules in three bottle-like SWCNTs.

Figure 8. Evolution of the translocation displacement L(t) of water molecules driven by methane molecules in three different terrace-like SWCNTs. The inset shows the dependence of the translocation displacement L(t) of the water molecules on time (t ) 0-100 ps).

clear fluctuations around the average value, which is caused by the vibration of the water molecules in the interior of the SWCNTs at their equilibrium position in the axial direction. The final configurations of the three systems can be seen in Figure 5. The maximum value of the translocation displacement of the water molecules in the (12, 0) - (13, 0), (12, 0) - (14, 0), and (12, 0) - (15, 0) SWCNTs is 18.037, 23.708, and 34.703 Å, respectively. The corresponding average velocities of the three systems are 9 × 10-4, 1.15 × 10-3, and 1.7 × 10-3 Å/ps. As shown in Figure 6, the displacement L(t) of the water molecules displays a roughly linear dependence on time t for the different diameters of SWCNTs to which the (12, 0) SWCNT is connected. Hence, the translocation velocity of the water molecules increases with the diameter of the SWCNTs. However, for a simple SWCNT-water system without the methane molecules, as shown in Figure 7, there is still a quite small displacement L of water moving toward the next compartment, which is much less than those driven by the methane molecules. Due to the interaction of the long-lasting hydrogen bonds between water molecules, it is a collective motion behavior for the confined water molecules, which is consistent with the ballistic-type mechanism of Striolo’s results.25 When the water molecules reach a junction region, they oscillate backward and forward along the axial direction, but at last they would not spontaneously enter the next compartment due to the barrier of the junction. This is different from Hummer’s calculation in which it was found that a one-dimensionally ordered chain of water can fill a nonpolar CNT spontaneously and continuously with little resistance.1 Likewise, our finding differs greatly from the previous investigations of fast water transport through a single-diameter CNT reported by Holt,14 Majumder,15 and Thomas.16 To further examine the effect of a series-connected SWCNT system on the transport of water molecules, a system with more

junctions, such as the terrace-like SWCNT, is designed to perform another simulation. As shown in the inset in Figure 8, during the first 100 ps in the simulation, the translocation velocity of the water molecules increases with the pore size, as in Figure 6. The reason is that, during the first 100 ps, the translocation displacement of water molecules in the (12, 0) (13, 0) - (14, 0) - (15, 0), (12, 0) - (14, 0) - (15, 0) - (16, 0), and (12, 0) - (15, 0) - (16, 0) - (17, 0) SWCNTs is 1.545, 1.564, and 1.573 Å, respectively, less than those in the (12, 0) - (13, 0), (12, 0) - (14, 0), and (12, 0) - (15, 0) SWCNTs (2.106, 2.360, and 2.863 Å). This indicates that the methane molecules in the terrace-like SWCNTs just start to diffuse. More junctions of the terrace-like system do not affect the diffusion of the methane molecules. The transport behavior of the water molecules in the terrace-like SWCNTs is similar to that in the bottle-like SWCNTs. Over time, an unexpected opposite trend is observed in Figure 8, in that the axial translocation velocity of the water molecules in the terrace-like SWCNTs increases sharply with the decrease of the diameters of the SWCNTs connected to the (12, 0) SWCNT. This indicates that the narrower terrace-like SWCNTs give the methane molecules a greater driving force. Consequently, the translocation velocity of the water molecules increases slowly as the diameters of the SWCNTs connected to the (12, 0) SWCNT increase during the initial 100 ps but decreases rapidly with the increase in diameter after 100 ps. Figure 9 shows the final snapshots of the three terrace-like systems at t ) 18 000 ps. It is worth noting that the water molecules escape completely from the (12, 0) - (13, 0) - (14, 0) - (15, 0) SWCNT. The translocation displacement L (length 81.296 Å) of the water molecules is the longest in this SWCNT (as shown in Figure 9a), whereas the translocation displacement L (length 35.201 Å) of the water molecules in the (12, 0) - (15, 0) - (16, 0) - (17, 0) SWCNT is the shortest (as shown in Figure

Water Molecules along SWCNTs

J. Phys. Chem. B, Vol. 114, No. 26, 2010 8679 National Basic Research Program of China (2007CB613901). This work is supported by a grant from the National Science Fund for Distinguished Young Scholars (No. 50625101) and Scientific Research Foundation for Returned Scholars (JIAO WAI SI LIU2007-1108), Ministry of Education of China. We also appreciate the support from the Natural Science Fund for Distinguished Young Scholars of Shandong (JQ200817). This work is also supported by the Natural Science Fund of Shandong Province (ZR2009FM043), by the Ph.D. Dot Programs Foundation of ministry of education of China (No. 20090131110025) and by the National Science Fund for Distinguished Young Scholars (No. 2009JQ014) from Shandong University. References and Notes

Figure 9. Simulation snapshots of three terrace-like SWCNTs at time t ) 1.8 × 104 ps.

9c). These results conflict with those in Figure 5. Moreover, the water molecules in the terrace-like SWCNTs move further than those in the bottle-like SWCNTs over the same interval. A comparison of the results for the transport of water molecules in the bottle-like SWCNTs with those for the terrace-like SWCNTs confirms that the greater number of junctions in the terrace-like SWCNTs gives the methane molecules a more sustained driving force, which favors the rapid transport of the water molecules along the SWCNTs. 4. Conclusion MD simulation has been performed to study the continuous methane-driven transport of water molecules along bottle-like and terrace-like SWCNTs connected by junctions to SWCNTs with different diameters. Methane molecules can overcome the potential barrier caused by the junctions and successfully pull the water molecules across the junction regions from one compartment to the next. In the bottle-like SWCNTs, the junctions are detrimental obstacles to the transport of water through the nanotube, especially when the CNTs have small diameters. In contrast, for the terrace-like SWCNTs with more junctions, the junctions become a driving force that accelerates the diffusion of the methane molecules. Thus, the translocation displacement of water molecules in the terrace-like SWCNTs is much longer than that in the bottle-like SWCNTs. These results have implications for the design of nanoscale molecular diffusive drivers as motors for the transport of liquid or gas molecules. This study also indicates that the filling/emptying transition in functionalized nanotubes can potentially be used as a wetting-dewetting nanoscale device, which would have wide application in nanotechnology and biotechnology. Acknowledgment. We acknowledge the support from the National Natural Science Foundation of China (Grant Nos. 50971081, 50831003, 50871062, and 50772061) and the

(1) Hummer, G.; Rasaiah, J. C.; Noworyt, J. P. Nature 2001, 414, 188– 190. (2) Dellago, C.; Naor, M. M.; Hummer, G. Phys. ReV. Lett. 2003, 90, 105902/1–4. (3) Sunand, L.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122, 12340– 12345. (4) Miller, S. A.; Young, V. Y.; Martin, C. R. J. Am. Chem. Soc. 2001, 123, 12335–12342. (5) Wang, Q. Nano Lett. 2009, 9, 245–249. (6) Ren, Z. F.; Huang, Z. P.; Xu, J. W.; Wang, J. H.; Bush, P.; Siegal, M. P.; Provencio, P. P. Science 1998, 282, 1105–1107. (7) Power, T. D.; Skoulidas, A. I.; Sholl, D. S. J. Am. Chem. Soc. 2002, 124, 1858–1859. (8) Wang, Q.; Challa, S. R.; Sholl, D. S.; Johnson, J. K. Phys. ReV. Lett. 1999, 82, 956–959. (9) Wang, Q. Carbon 2009, 47, 1870–1873. (10) Skoulidas, A. I.; Ackerman, D. M.; Johnson, J. K.; Sholl, D. S. Phys. ReV. Lett. 2002, 89, 185901. (11) Yu, H. Q.; Li, H.; Zhang, J. X.; Liu, X. F.; Liew, K. M. Carbon 2010, 48, 417–423. (12) Li, H.; Zhang, X. Q.; Liew, K. M. J. Chem. Phys. 2008, 128, 034707/1–5. (13) Maibaum, L.; Chandler, D. J. Phys. Chem. B 2003, 107, 1189– 1193. (14) Holt, J. K.; Park, H. G.; Wang, Y. M.; Stadermann, M.; Artyukhin, A. B.; Grigoropoulos, C. P. Science 2006, 312, 1034–1037. (15) Majumder, M.; Chopra, N.; Andrews, R.; Hinds, B. J. Nature 2005, 438, 44. (16) Thomas, J. A.; McGaughey, A. J. H. Nano Lett 2008, 8, 2788– 2793. (17) Zambrano, H. A.; Walther, J. H.; Koumoutsakos, P.; Sbalzarini, I. F. Nano Lett. 2009, 9, 66–71. (18) Zhao, Y. C.; Song, L.; Deng, K.; Liu, Z.; Zhang, Z. X.; Yang, Y. L. AdV. Mater. 2008, 20, 1772–1776. (19) Maniwa, Y.; Matsuda, K.; Kyakuno, H.; Ogasawara, S.; Hibi, T.; Kadowaki, H.; Suzuki, S.; Achiba, Y.; Kataura, H. Nature 2007, 6, 135– 141. (20) Kalra, A.; Hummer, G.; Garde, S. J. Phys. Chem. B 2004, 108, 544–549. (21) Teleman, O.; Jo¨nsson, B.; Engstro¨m, S. Mol. Phys. 1987, 60, 193– 203. (22) Iijima, S.; Ichihashi, T.; Ando, Y. Nature 1992, 356, 776–778. (23) Zhen, Y.; Henk, W. Ch. P.; Leon, B.; Cees, D. Nature 1999, 40, 273–276. (24) Maniwa, Y.; Kataura, H.; Abe, M.; Udaka, A.; Suzuki, S.; Achiba, Y.; Kira, H.; Matsuda, K.; Kadowaki, H.; Okabe, Y. Chem. Phys. Lett. 2005, 401, 534–538. (25) Striolo, A. Nano Lett. 2006, 6, 633–639.

JP102810J