- Email: [email protected]
The mechanism of selective deposition of luminescent molecules onto self-assembled monolayers using molecular dynamics
Applied Surface Science 349 (2015) 163–168
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
The mechanism of selective deposition of luminescent molecules onto self-assembled monolayers using molecular dynamics Hui Yan a , Shiling Yuan b,∗ , Suyuan Zeng c , Meiju Niu c,∗∗ a b c
School of Pharmacy, Liaocheng University, Liaocheng 252059, China Key Laboratory of Colloid and Interface Chemistry, Shandong University, Jinan 250100, China School of Chemistry and Chemical Engineering, Liaocheng University, Liaocheng 252059, China
a r t i c l e
i n f o
Article history: Received 30 March 2015 Received in revised form 27 April 2015 Accepted 28 April 2015 Available online 6 May 2015 Keywords: Self-assembled monolayers Site-selective deposition Molecular dynamics Umbrella sampling
a b s t r a c t The site-selective deposition behavior of perylene onto a self-assembled monolayers (SAMs) patterned substrate was studied using the equilibrium and steered molecular dynamics simulations. Four kinds of different densely packed SAMs were constructed on silicon oxide substrates as the patterned templates. Equilibrium MD simulations showed that the packing density of alkyl chains on the substrate could influence the deposition behavior of the organic molecules. The potential of mean force (PMF) of the deposition process of perylene onto different density packed SAMs, which was calculated by the umbrella sampling with the weighted histogram analysis method (WHAM), determined the favorite location of perylene on the SAM. The equilibrium and non-equilibrium MD methods gave the same conclusion about the deposition positions of organic molecules on the patterned substrate. In summary, this comprehensive study is expected to provide useful information for the synthesis of new functional materials. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The site-selective patterning of luminescent molecules with ordered micro- and nanoscopic arrangements has attracted much attention due to their great applications in photonics [1,2], optoelectronics [2–5], biochip-based detection [6], and biosensor arrays [7]. Recently, Chi’s group reported a bottom-up approach to fabricate organic luminescent stripe patterns [8–12]. This method is based on the selective gas deposition of organic molecules on selforganized patterned structures. The patterned structures consist of two monolayer phases rather than different chemical natures, which is the novelty differing from the other methods, such as nanoimprinting [13] and microcontact printing [14]. In Chi’s experiments [10,11], an interesting patterned structure was produced by transferring a monolayer of l-␣-dipalmitoylphosphatidylcholine (DPPC) onto mica substrates using the Langmuir–Blodgett (LB) technique. The patterned structure was comprised of liquid expanded (LE) and liquid condensed (LC) DPPC
∗ Corresponding author at: Key Laboratory of Colloid and Interface Chemistry, Department of Chemistry and Chemical Engineering, No. 27, Shanda South Road, Jinan 250100, China. Tel.: +86 531 88365896; fax: + 86 531 88564464. ∗∗ Corresponding author. E-mail addresses: [email protected] (S. Yuan), [email protected] (M. Niu). http://dx.doi.org/10.1016/j.apsusc.2015.04.211 0169-4332/© 2015 Elsevier B.V. All rights reserved.
phases. Such a structured surface can be used as a template to guide the selective deposition of both inorganic [15,16] and organic molecules [12] from the gas phase. For instance, organic luminescent molecules such as 3(5)-(9-anthryl) pyrazole (ANP) [10,12] and perylene [9] can deposit onto the LE DPPC phase areas of the stripe patterned structure at first when the amount of the organic molecules is small. However, when the amount of evaporation increases, the organic molecules begin to deposit onto the LC phase area. Similar to the DPPC template, the self-assembled monolayers (SAMs) patterned template can also be used to guide the selective deposition of luminescent molecules [17,18], as it has similar structural features with self-assembled DPPC film. In our previous studies, we performed equilibrium molecular dynamics (MD) simulations to investigate the deposition behavior of various luminescent molecules onto the self-assembled monolayers (SAMs) templates [19–21], with the luminescent molecules including perylene, rubrene, and ANP. The mechanism of selective deposition was found to be related to the different binding energies of the organic molecules and the monolayer on the silicon oxide substrate. Based on the AFM data, Chi et al. [10] postulated a difference in the energy barrier for molecules crossing from the two phases on the substrate. This notion was supported by their later molecular dynamics simulations [22]. However, further work to understand the mechanism is still necessary, especially for the investigating of binding energies
164
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
between organic molecules and different phases on the substrate. An especially useful technique in analyzing molecular interactions is umbrella sampling [23], which is used to assemble the potential of mean force (PMF). In this work, the equilibrium and non-equilibrium (named steered MD simulations [24]) simulations were carried out to study the site-selective deposition behavior of perylene onto different densely packed alkyl chains monolayer on a silicon oxide substrate. Using equilibrium MD simulations, the deposition behaviors of the perylene molecule as well as the structural features of the alkyl chains on the SAMs were investigated. From the steered MD simulations the kinetic information was obtained by calculating the potential of mean force (PMF), i.e., the free energy as a function of the separation between the center of mass (COM) of perylene and SAM. In the simulation, the PMF was calculated using umbrella sampling [23] and the weighted histogram analysis method (WHAM) [25], which are very helpful in studying the microscopic mechanism of various adsorption processes [26–28].
2. Simulation methods 2.1. Model systems The functionalized self-assembled monolayer (SAM) systems were constructed by connecting the silicon oxide substrate with 18carbon-long alkyl chains in a well-ordered arrangement according to the previous work of SAM systems on a silicon oxide substrate [19,29,30]. In this work four different densely packed SAM systems were built. The most densely packed SAM (100%-SAM) consisting of 64 alkyl chains with a surface occupied area of approximately 26.5 A˚ 2 per alkyl chain which was in accordance with the previous studies [19], was considered as the LC monolayer in Chi’s experiments. The less densely packed SAM systems (75%, 50%, and 25%-SAMs) were constructed as the LE monolayers by removing alkyl chains from the 100%-SAM, yielding three SAM systems which consisted 48, 32, and 16 alkyl chains, respectively. In each simulation system, a periodic cubic box with dimensions of 39.3 A˚ × 43.2 A˚ × 104.2 A˚ was built. For each SAM system, a 5 ns molecular simulation was performed using an NVT ensemble to obtain a stable configuration. Then, one molecule of perylene was placed onto the surface of each SAM at a distance of about 4 nm from the surface of the silicon oxide substrate.
2.2. Computational details Molecular dynamics simulations were performed using the GROMACS software package (version 4.6.3) [31]. The all-atom optimized performance for liquid systems (OPLS-AA) force field [32] was adopted for all of the potential function terms to calculate the interatomic interactions. The total potential energy was given as a combination of valence terms, including bond stretching, angle bending, torsion, and non-bonded interactions. The non-electrostatic parts of the interaction between the atoms were described by the Lennard–Jones potential, and the standard geometric combination rules were used for the van der Waals interactions between different atom species. The structure optimized calculation was performed for the perylene molecule at B3LYP/6-31G level using the Gaussian03 package [33], and the atomic electrostatic potential (ESP) charges were obtained as the partial atomic charges in coulomb interaction terms. The force field parameters for the silicon oxide substrate were taken from the work of Lorenz et al. [34]. More detailed information about the force field parameters applied in this work is summarized in the Supplementary data, Table S1.
All the simulations were initialized by minimizing the energies of the initial configurations with the steepest descent method. Following the minimization, a 5 ns MD simulation under canonical ensemble (NVT) for each SAM system was carried out with a time step of 1 fs. The silicon oxide substrate was considered as rigid while the alkyl chains were set as flexible through the MD simulations. The temperature was kept constant at 298 K by the Berendsen thermostat algorithm [35] with a coupling constant of 0.1 ps. Bond lengths were constrained using the LINCS algorithm [36] and periodic boundary conditions were applied in all directions. Short-range non-bonded interactions were cut off at 12 A˚ with long-range electrostatics calculated using the particle mesh Ewald method [37]. All of the configurations were visualized using VMD 1.9.1 [38]. The free energy profile for perylene to deposit from the gaseous phase onto the different densely packed SAMs was calculated on the basis of biased umbrella sampling simulations [23]. The initial configuration of the biased MD simulation was identical to that of the corresponding equilibrium MD run for each SAM system. Then the biased simulations, i.e., COM (center of mass) pulling simulations, were performed by applying an external force on the perylene molecule towards the silicon oxide substrate. Fig. S1 in the Supplementary data shows the 100%-SAM system as a schematic representation of the pulling simulation. For each SAM system, a 600 ps pulling simulation was conducted with a spring constant of 1000 kJ mol−1 nm−2 and a pull rate of 0.005 nm ps−1 . Based on the separation between the COM of perylene and the surface of the silicon oxide substrate, 16 starting configurations were derived ˚ In corresponding to the 16 sampling windows with a spacing of 2 A. each window, a 5 ns umbrella sampling simulation was performed for each SAM system. Then, the potential of mean force between the perylene and each SAM as a function of the distance between the COM of perylene and the surface of the silicon oxide substrate was calculated using the weighted histogram analysis method (WHAM) [25]. 3. Results and discussion 3.1. Deposition behavior of perylene onto different densely packed SAMs The deposition configurations of the perylene molecule on the four densely packed SAMs at the end of the MD simulation are shown in Fig. 1. From the figure, it can be seen that the organic molecule was deposited onto different positions of the SAMs. On the LE monolayer covered substrate (25%-SAM), the small molecule was adsorbed onto the surface of the substrate (Fig. 1a), while on the full LC monolayer covered substrate (100%-SAM), the small molecule was adsorbed on the surface of the SAMs (Fig. 1d). The equilibrium of the simulations was determined based on the vertical distance between the COM of the perylene molecule and the surface of the SiO2 substrate during the MD run. These distances with time are shown in Fig. 2. It should be noted that the distance in each SAM system remained steady during the last 4.5 ns of the simulation, which confirmed each system had reached equilibrium. Another significant observation is that the perylene molecule was deposited deeply on to the alkyl chains film with a decrease in cover density of the SAM. This is probably due to the steric hindrance of the more densely packed alkyl chains film, which hindered the perylene molecule from submerging into the film. 3.2. Structural properties of different densely packed SAMs As discussed above, the perylene molecule would submerge deeply into the hydrocarbon film with the decrease of cover density
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
165
Fig. 1. Snapshots of the configurations of the four different densely packed SAMs systems at the end of the simulation. (a) 25%-SAM, (b) 50%-SAM, (c) 75%-SAM and (d) 100%-SAM. Perylene molecules are represented by red VDW spheres for clarity, while the SAMs are in ball and stick model. The atom coloring scheme is O, red; Si, yellow; C, cyan; and H, white. (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.)
Fig. 2. Time profiles of the vertical distance between the COM of the perylene molecule and the surface of the SAM in each simulation system.
of SAMs. We considered that the packing density was responsible to this phenomenon. Taking this into account, the structure of the alkyl chains on the substrate was analyzed. First, the locations of different components in each densely packed SAM were characterized by calculating the average number density distribution functions along the z-axis, which is in the direction normal to the plane of the substrate (Fig. 3). The different components including alkyl chains and substrate were separately computed. The distributions of perylene were also plotted to show the deposition locations of perylene molecules on different densely packed SAMs. From the density distribution, it was observed that as the cover density of SAM increased, the thickness of the alkyl chain film on the substrate increased. For the LC monolayer covered SAM (100%SAM), a plateau is found from the density distribution profile. This indicates that the arrangement of the alkyl chains in the film is very ordered. According to the distribution, the thickness of the full covered film could be estimated to be approximately 1.5 nm. For the
Fig. 3. Number density profiles of components for each SAM system in the direction along the z-axis of the simulation cell. Panels a and b correspond to the distribution of hydrocarbon chains and perylene molecule, respectively. The curves of each system are indicated in the figure.
two less densely packed SAMs (50% and 75%), two layers with different densities of alkyl chains were found from the profiles. That is, the density of the film which is adjacent to the SiO2 substrate is larger than that on the surface of the SAM, which means the hydrocarbon chains bend to the substrate. A sharp peak was observed from the profile of the 25%-SAM, which means that a thick alkyl
166
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
Fig. 4. (a) Order parameter SZ of the hydrocarbon chains in films of the four SAM systems. (b) Probability distribution of the tilt angle between the hydrocarbon vector and the z-axis of the simulation cell for each SAM system.
chains film forms on the substrate. The intensity of the density profile of 25%-SAM is significantly weaker than those of the other three SAMs, which indicates that the alkyl chains do not cover the entire substrate. Number density profiles of the perylene molecule in the zdirection provide further insights into the deposition location on the SAMs (Fig. 3b). On the 100%-SAM, the perylene molecule is located on the top surface of the SAM, while on the 50% and 75%SAM, the perylene molecules submerge into the films, located at the region of low density as discussed above. The location of the perylene molecule is almost coincidental with that of the alkyl chains on the 25%-SAM, indicating that the perylene molecule was deposited onto the SiO2 substrate. 3.3. Orientation of alkyl chains on different densely packed SAMs To better quantify the conformations of the alkyl chains on different densely packed SAMs, the order parameter of the hydrocarbon films was calculated which is defined as: SZ =
3 1 cos2 − 2 2
where is the angle between the z-axis and the molecular axis of the respective Cn−1 –Cn+1 bond (n = 2,3,. . .,17, C1 is the carbon atom which is connected to the substrate, and C18 is the terminal atom of the tail) in the hydrocarbon chains. These results are shown in Fig. 4a. From this figure it can be seen that, for the 100%-SAM, values for each carbon seem to change smoothly without sharp variation. For the 75%-SAM, values reach a plateau starting from the first two carbon atoms. This suggests that those hydrocarbon chains have a similar inclination in the two more densely packed SAMs. For the 25% and 50%-SAMs, the variation of these values is very obvious, suggesting that the hydrocarbon chains are more disordered in these two cases. More detailed information on the orientation of the hydrocarbon chains can be obtained from the probability distribution of the tilt angle ˛ between the hydrocarbon chain vector and the z-axis. The vector connecting the C1 atom and the terminal C18 atom is defined as the chain vector. Fig. 4b shows the probability distribution r(˛) of the tilt angle between the tail vectors with the normal of the surface of the SAM system. A sharp peak can be observed from the angle distribution of the 100%-SAM, suggesting the arrangement of the alkyl chains in the LC monolayer is well ordered with an average tilt angle of approximately 41.4◦ . In the case of the LE monolayers of the other three SAM systems, the distributions are in wide ranges especially for the 50%-SAM system, which indicates that the chains are very disordered. The tilt angle distribution of the chains on the 25%-SAM is mainly around 85.2◦ , which means the chains are almost perpendicular to the normal of the SAM. Since the
packing density of the alkyl chains of the 25%-SAM is much lighter, the chains tend to bend towards the surface of the substrate and form a thin film. 3.4. Diffusion properties of perylene on the SAMs It has been reported that the site-selective deposition behavior of perylene on different densely SAMs is a diffusion-controlled process [10,22]. To investigate the diffusion properties of the organic molecules on the films, we have calculated the mean square displacements (MSDs) of perylene molecules in the xy-plane of the four SAM systems over the last 500 ps MD trajectories, as shown in Fig. 5. From the figure, it can be seen that the diffusion of the perylene molecule on the LC monolayer covered SAM (100%-SAM) is significantly higher than it is on the other three SAMs. This is because that the perylene molecule was deposited onto the surface of the hydrocarbon film due to the steric hindrance of the densely packed hydrocarbon chains on the substrate, which makes the organic molecule diffuse more freely on the top surface of the SAM. The diffusion of the perylene molecule on the 75%-SAM is the lowest among the four SAM systems, and it is very similar to that of the 25%-SAM. While, the diffusion of the perylene molecule on the 50%-SAM is much more pronounced than the above two SAM systems. Because the perylene molecule can submerge into the alkyl chain film as discussed above, we consider that the motion of the perylene molecule should be related to the packing density of the surrounding alkyl chains. From Fig. 3, the locations of the perylene molecule on the three kinds of less densely packed SAM (25%, 50%, and 75%-SAM) are approximately 1.5, 2.1 and 2.5 nm of the simulation cell, respectively. The corresponding number densities of the alkyl chains at the three locations are 90, 79 and 92 nm−3 ,
Fig. 5. Mean square displacements for perylene on the four different densely packed SAMs.
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
respectively. This indicates that the diffusion of perylene is confined with the increased density of the surrounding alkyl chains. The lower density of the alkyl chains around perylene on the 50%SAM is mainly due to the less ordered arrangement of the chains, while for the 25%-SAM, the chains bend towards the surface of the SiO2 substrate, resulting in a higher density of the film. Since the film on the 25%-SAM did not cover the whole surface of the substrate, the perylene molecule would locate at the porosities on the surface with a limited mobility. 3.5. Deposition PMF of different densely packed SAM systems The free energy of deposition of a perylene molecule from the gaseous phase onto the different densely packed SAMs was calculated using umbrella sampling [23] and the weighted histogram analysis method (WHAM) [25]. As discussed above in Section 2, a single perylene molecule was pulled towards the surface of the SiO2 substrate and the associated potential of mean force (PMF) was calculated during this process. In this case, the reaction coordinate corresponds to the z-axis of the simulation box. The PMFs as a function of the vertical distance between the COM of a perylene molecule and the surface of the substrate were calculated using the GROMACS tool g wham [31], shown in Fig. 6. From Fig. 6, satisfactory convergence can be achieved for the long sampling simulations. The global minimums of the four potential of mean force profiles occur at <0.50, 0.75, 1.11, and 1.93 nm for the four SAM systems with the cover density increased. These minimums are comparable to the deposition locations of the perylene molecule on the four kinds of SAM shown in Fig. 2. This indicates that the deposition positions of the perylene molecule on the different SAM systems predicted using steered molecular dynamics methods (SMD) agree well with those obtained by the equilibrium MD method. Since the minimums of the PMF curves represent the stable binding sites between the perylene molecule and SAM, they can be used to explain the preferable deposition locations of perylene onto the SAM. The binding energy G of a perylene molecule adsorbed on each kind of SAM can be determined from the corresponding PMFs. The overall order of binding energies for the four kinds of SAM systems is as follows: G25%-SAM ≈ G75%-SAM > G50%-SAM > G100%-SAM . It should be noted that the global minimum is similar for the 25% and 75%-SAM systems. This means perylene binds to these two kinds of SAM more stably than the other two SAM systems. The higher binding energies in these two SAM systems are mainly due to the higher density of alkyl chains around the perylene molecules as discussed above. In contrast, the binding energy of 50%-SAM system is smaller due to the less ordered and loosely packed alkyl chains around the
167
perylene molecule. The less stable binding between perylene and SAM causes much more pronounced displacement of the perylene molecule on the film. For the 100%-SAM, it can be seen that there is another pronounced minimum located at around 1.0 nm besides the global minimum. The global minimum represents the preferred location of perylene on the 100%-SAM, which is on the top surface of the film. However, the perylene did not submerge into the film during the equilibrium MD simulation. This is because the perylene molecule was pulled into the film under an external force which was applied on perylene in the steered MD simulation. The configuration of perylene submerged into the film is provided in the Supplementary data, Fig. S2. It can be seen that the perylene molecule is located at the gap between the well-ordered hydrocarbon chains. Due to the well-ordered chains of the film, the gap can provide much free space to take in the perylene. Thus, the binding between perylene and SAM is quite unstable. The profile of the external force with time evolution and configurations at different times are provided in Figs. S3 and S4. The maximum value occurs at about 340 ps at which time the perylene molecule is about to submerge into the film under the external force. This indicates that large force is needed to pull the perylene molecule into the film. In other words, it is difficult for the perylene molecule to enter into the film of a fully packed SAM. Based on the above analysis, we can conclude that perylene prefers to deposit onto the less densely packed SAM, i.e., the LE phase on the substrate. Therefore, site-selective deposition on SAMs with a striped pattern composed of the expanded and condensed alkyl chains phase in the stripes may occur. Since the binding energy between perylene and the expanded alkyl chains phase is higher than that of the condensed alkyl chains phase, the organic molecules that deposit on the condensed phase can move to the expanded phase of the SAM.
4. Conclusions We studied the deposition behavior of the organic molecule perylene onto different densely packed SAMs using equilibrium and biased MD simulations. The equilibrium MD results show that the perylene molecule deposits on the different positions of the different densely packed SAMs. With decreased cover density of SAM, the arrangement of the hydrocarbon chains on the substrate will become less ordered and perylene will submerge deeply into the alkyl chains film. On the LE monolayer covered SAM, the hydrocarbon chains bend towards the surface of the substrate, producing higher density of alkyl chains near the substrate and lower density on the surface region of the film. The diffusion of perylene is related to the density of the surrounding alkyl chains. That is, with a higher density of the alkyl chains around the perylene molecule, the diffusion of perylene will be confined. Using umbrella sampling and the weighted histogram analysis method, the potential of mean force (PMF) of the deposition process of a perylene molecule from the gaseous phase onto the different densely packed SAMs was calculated. From the PMF curves, the binding energy between perylene and each kind of SAM can be estimated. Based on the binding energies, the site-selective deposition of perylene on different densely packed SAMs can be explained.
Acknowledgments
Fig. 6. Potential of mean force along the reaction coordinate for a perylene molecule pulled from gaseous phase onto different densely packed SAMs (0 represents the surface of the silicon oxide substrate).
This work was supported by the National Natural Science Foundation of China (No. 21203084). Thanks to Dr. Edward C. Mignot, Shandong University, for linguistic advice.
168
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
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.apsusc.2015.04. 211 References [1] M. Gaal, C. Gadermaier, H. Plank, E. Moderegger, A. Pogantsch, G. Leising, E. List. Adv. Matter 15 (2003) 1165. [2] C. Kallinger, M. Hilmer, A. Haugeneder, M. Perner, W. Spirkl, U. Lemmer, J. Feldmann, U. Scherf, K. Müllen, A. Gombert, V. Wittwer, Adv. Matter 10 (1998) 920. [3] Y. Koide, Q. Wang, J. Cui, D.D. Benson, T.J. Marks, J. Am. Chem. Soc. 122 (2000) 11266. [4] T. Lee, J. Zaumseil, Z. Bao, J.W.P. Hsu, J.A. Rogers, PNAS 101 (2004) 429. [5] J.G.C. Veinot, H. Yan, S.M. Smith, J. Cui, Q. Huang, T.J. Marks, Nano Lett. 2 (2002) 333. [6] B.J. Cheek, A.B. Steel, M.P. Torres, Y. Yu, H. Yang, Anal. Chem. 73 (2001) 5777. [7] B.M. Lingerfelt, H. Mattoussi, E.R. Goldman, J.M. Mauro, G.P. Anderson, Anal. Chem. 75 (2003) 4043. [8] X.D. Chen, A.L. Rogach, D.V. Talapin, H. Fuchs, L.F. Chi, J. Am. Chem. Soc. 128 (2006) 9592. [9] J. Hao, N. Lu, L. Li, M. Hirtz, L. Gao, W. Wang, C. Du, H. Fuchs, L.F. Chi, Soft Matter 6 (2010) 5302. [10] J. Hao, N. Lu, Q. Wu, W. Hu, X. Chen, H. Zhang, Y. Wu, Y. Wang, L.F. Chi, Langmuir 24 (2008) 5315. [11] M. Hirtz, H. Fuchs, L.F. Chi, J. Phys. Chem. B 112 (2008) 824. [12] L. Gao, N. Lu, J. Hao, W. Hu, G. Shi, Y. Wang, L.F. Chi, Langmuir 25 (2009) 3894. [13] W. Hu, N. Lu, H. Zhang, Y. Wang, N. Kehagias, V. Reboud, C.M. Sotomayor, Torres, J. Hao, W. Li, H. Fuchs, L.F. Chi, Adv. Mater 19 (2007) 2119. [14] H.Y. Choi, S.H. Kim, J. Jang, Adv. Mater. 16 (2004) 732. [15] M. Gleiche, L.F. Chi, H. Fuchs, Nature 403 (2000) 173. [16] M. Gleiche, L.F. Chi, E. Gedig, H. Fuchs, ChemPhysChem 2 (2001) 187. [17] J.M. Choi, S.H. Jeong, D.K. Hwang, S. Im, B.H. Lee, M.M. Sung, Org. Electron 10 (2009) 199. [18] S. Steudel, D. Janssen, S. Verlaak, J. Genoe, P. Heremans, Appl. Phys. Lett. 85 (2004) 5550. [19] Z. Xu, M. Hirtz, S.L. Yuan, C.B. Liu, L.F. Chi, Chem. Phys. Lett. 507 (2011) 138. [20] C. Zhang, Z. Xu, H. Yan, F. Gao, S.L. Yuan, Chem. Phys. Lett. 571 (2013) 38.
[21] C. Zhang, C. Du, H. Yan, S.L. Yuan, L.F. Chi, RSC Adv. 3 (2013) 15404. [22] M. Hirtz, N. Kumar, J.H. Franke, J. Hao, N. Lu. H. Fuchs, L.F. Chi. J. Colloid Interface Sci. 389 (2013) 206. [23] G.M. Torrie, J.P. Valleau, J. Comput. Phys. 23 (1977) 187. [24] S. Izrailec, S. Stepaniants, B. Isralewitz, D. Kosztin, H. Lu, F. Molnar, W. Wriggers, K. Schulten, Steered Molecular Dynamics, Springer-Verlag, Berlin, 1998. [25] S. Kumbia, D. Bouzida, R.H. Swendsen, P.A. Kollman, J.M. Rosenberg, J. Comput. Chem. 13 (1992) 1011. [26] A. Striolo, J.M. Prausnitz, J. Chem. Phys. 114 (2001) 8565. [27] Y. Sun, B.N. Dominy, R.A. Latour, J. Comput. Chem. 28 (2007) 1883. [28] Z. Xu, X. Yang, Z. Yang, J. Phys. Chem. B 112 (2008) 13802. [29] A. Hozumi, S. Asakura, A. Fuwa, N. Shirahata, T. Kameyama, Langmuir 21 (2005) 8234. [30] D. Janssen, R. DePalma, S. Verlaak, P. Heremans, W. Dehaen, Thin Solid Films 515 (2006) 1433. [31] D. van der Spoel, E. Lindahl, B. Hess, the GROMCAS Development Team, GROMACS User Manual Version 4.6.3, www.gromacs.org 2013. [32] W.L. Jorgensen, D.S. Maxwell, J. Tirada-Rives, J. Am. Chem. Soc. 118 (1996) 11225. [33] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Jr. Montgomery, T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision B.05; Gaussian, Inc., Pittsburgh, PA, 2003. [34] C.D. Lorenz, E.B. Webb III, M.J. Stevens, M. Chandross, G.S. Grest, Tribol. Lett. 19 (2005) 93. [35] H.J.C. Berendsen, J.P.M. Postma, W.F. van Gunsteren, A. DiNola, J.R. Haak, J. Chem. Phys. 81 (1984) 3684. [36] B. Hess, H. Bekker, H.J.C. Berendsen, J.G.E.M. Fraaije, J. Comput. Chem. 18 (1997) 1463. [37] U. Essman, L. Perera, M.L. Berkowitz, T. Darden, H. Lee, L.G. Pedersen, J. Chem. Phys. 103 (1995) 8577. [38] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics 14 (1996) 33.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
The mechanism of selective deposition of luminescent molecules onto self-assembled monolayers using molecular dynamics Hui Yan a , Shiling Yuan b,∗ , Suyuan Zeng c , Meiju Niu c,∗∗ a b c
School of Pharmacy, Liaocheng University, Liaocheng 252059, China Key Laboratory of Colloid and Interface Chemistry, Shandong University, Jinan 250100, China School of Chemistry and Chemical Engineering, Liaocheng University, Liaocheng 252059, China
a r t i c l e
i n f o
Article history: Received 30 March 2015 Received in revised form 27 April 2015 Accepted 28 April 2015 Available online 6 May 2015 Keywords: Self-assembled monolayers Site-selective deposition Molecular dynamics Umbrella sampling
a b s t r a c t The site-selective deposition behavior of perylene onto a self-assembled monolayers (SAMs) patterned substrate was studied using the equilibrium and steered molecular dynamics simulations. Four kinds of different densely packed SAMs were constructed on silicon oxide substrates as the patterned templates. Equilibrium MD simulations showed that the packing density of alkyl chains on the substrate could influence the deposition behavior of the organic molecules. The potential of mean force (PMF) of the deposition process of perylene onto different density packed SAMs, which was calculated by the umbrella sampling with the weighted histogram analysis method (WHAM), determined the favorite location of perylene on the SAM. The equilibrium and non-equilibrium MD methods gave the same conclusion about the deposition positions of organic molecules on the patterned substrate. In summary, this comprehensive study is expected to provide useful information for the synthesis of new functional materials. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The site-selective patterning of luminescent molecules with ordered micro- and nanoscopic arrangements has attracted much attention due to their great applications in photonics [1,2], optoelectronics [2–5], biochip-based detection [6], and biosensor arrays [7]. Recently, Chi’s group reported a bottom-up approach to fabricate organic luminescent stripe patterns [8–12]. This method is based on the selective gas deposition of organic molecules on selforganized patterned structures. The patterned structures consist of two monolayer phases rather than different chemical natures, which is the novelty differing from the other methods, such as nanoimprinting [13] and microcontact printing [14]. In Chi’s experiments [10,11], an interesting patterned structure was produced by transferring a monolayer of l-␣-dipalmitoylphosphatidylcholine (DPPC) onto mica substrates using the Langmuir–Blodgett (LB) technique. The patterned structure was comprised of liquid expanded (LE) and liquid condensed (LC) DPPC
∗ Corresponding author at: Key Laboratory of Colloid and Interface Chemistry, Department of Chemistry and Chemical Engineering, No. 27, Shanda South Road, Jinan 250100, China. Tel.: +86 531 88365896; fax: + 86 531 88564464. ∗∗ Corresponding author. E-mail addresses: [email protected] (S. Yuan), [email protected] (M. Niu). http://dx.doi.org/10.1016/j.apsusc.2015.04.211 0169-4332/© 2015 Elsevier B.V. All rights reserved.
phases. Such a structured surface can be used as a template to guide the selective deposition of both inorganic [15,16] and organic molecules [12] from the gas phase. For instance, organic luminescent molecules such as 3(5)-(9-anthryl) pyrazole (ANP) [10,12] and perylene [9] can deposit onto the LE DPPC phase areas of the stripe patterned structure at first when the amount of the organic molecules is small. However, when the amount of evaporation increases, the organic molecules begin to deposit onto the LC phase area. Similar to the DPPC template, the self-assembled monolayers (SAMs) patterned template can also be used to guide the selective deposition of luminescent molecules [17,18], as it has similar structural features with self-assembled DPPC film. In our previous studies, we performed equilibrium molecular dynamics (MD) simulations to investigate the deposition behavior of various luminescent molecules onto the self-assembled monolayers (SAMs) templates [19–21], with the luminescent molecules including perylene, rubrene, and ANP. The mechanism of selective deposition was found to be related to the different binding energies of the organic molecules and the monolayer on the silicon oxide substrate. Based on the AFM data, Chi et al. [10] postulated a difference in the energy barrier for molecules crossing from the two phases on the substrate. This notion was supported by their later molecular dynamics simulations [22]. However, further work to understand the mechanism is still necessary, especially for the investigating of binding energies
164
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
between organic molecules and different phases on the substrate. An especially useful technique in analyzing molecular interactions is umbrella sampling [23], which is used to assemble the potential of mean force (PMF). In this work, the equilibrium and non-equilibrium (named steered MD simulations [24]) simulations were carried out to study the site-selective deposition behavior of perylene onto different densely packed alkyl chains monolayer on a silicon oxide substrate. Using equilibrium MD simulations, the deposition behaviors of the perylene molecule as well as the structural features of the alkyl chains on the SAMs were investigated. From the steered MD simulations the kinetic information was obtained by calculating the potential of mean force (PMF), i.e., the free energy as a function of the separation between the center of mass (COM) of perylene and SAM. In the simulation, the PMF was calculated using umbrella sampling [23] and the weighted histogram analysis method (WHAM) [25], which are very helpful in studying the microscopic mechanism of various adsorption processes [26–28].
2. Simulation methods 2.1. Model systems The functionalized self-assembled monolayer (SAM) systems were constructed by connecting the silicon oxide substrate with 18carbon-long alkyl chains in a well-ordered arrangement according to the previous work of SAM systems on a silicon oxide substrate [19,29,30]. In this work four different densely packed SAM systems were built. The most densely packed SAM (100%-SAM) consisting of 64 alkyl chains with a surface occupied area of approximately 26.5 A˚ 2 per alkyl chain which was in accordance with the previous studies [19], was considered as the LC monolayer in Chi’s experiments. The less densely packed SAM systems (75%, 50%, and 25%-SAMs) were constructed as the LE monolayers by removing alkyl chains from the 100%-SAM, yielding three SAM systems which consisted 48, 32, and 16 alkyl chains, respectively. In each simulation system, a periodic cubic box with dimensions of 39.3 A˚ × 43.2 A˚ × 104.2 A˚ was built. For each SAM system, a 5 ns molecular simulation was performed using an NVT ensemble to obtain a stable configuration. Then, one molecule of perylene was placed onto the surface of each SAM at a distance of about 4 nm from the surface of the silicon oxide substrate.
2.2. Computational details Molecular dynamics simulations were performed using the GROMACS software package (version 4.6.3) [31]. The all-atom optimized performance for liquid systems (OPLS-AA) force field [32] was adopted for all of the potential function terms to calculate the interatomic interactions. The total potential energy was given as a combination of valence terms, including bond stretching, angle bending, torsion, and non-bonded interactions. The non-electrostatic parts of the interaction between the atoms were described by the Lennard–Jones potential, and the standard geometric combination rules were used for the van der Waals interactions between different atom species. The structure optimized calculation was performed for the perylene molecule at B3LYP/6-31G level using the Gaussian03 package [33], and the atomic electrostatic potential (ESP) charges were obtained as the partial atomic charges in coulomb interaction terms. The force field parameters for the silicon oxide substrate were taken from the work of Lorenz et al. [34]. More detailed information about the force field parameters applied in this work is summarized in the Supplementary data, Table S1.
All the simulations were initialized by minimizing the energies of the initial configurations with the steepest descent method. Following the minimization, a 5 ns MD simulation under canonical ensemble (NVT) for each SAM system was carried out with a time step of 1 fs. The silicon oxide substrate was considered as rigid while the alkyl chains were set as flexible through the MD simulations. The temperature was kept constant at 298 K by the Berendsen thermostat algorithm [35] with a coupling constant of 0.1 ps. Bond lengths were constrained using the LINCS algorithm [36] and periodic boundary conditions were applied in all directions. Short-range non-bonded interactions were cut off at 12 A˚ with long-range electrostatics calculated using the particle mesh Ewald method [37]. All of the configurations were visualized using VMD 1.9.1 [38]. The free energy profile for perylene to deposit from the gaseous phase onto the different densely packed SAMs was calculated on the basis of biased umbrella sampling simulations [23]. The initial configuration of the biased MD simulation was identical to that of the corresponding equilibrium MD run for each SAM system. Then the biased simulations, i.e., COM (center of mass) pulling simulations, were performed by applying an external force on the perylene molecule towards the silicon oxide substrate. Fig. S1 in the Supplementary data shows the 100%-SAM system as a schematic representation of the pulling simulation. For each SAM system, a 600 ps pulling simulation was conducted with a spring constant of 1000 kJ mol−1 nm−2 and a pull rate of 0.005 nm ps−1 . Based on the separation between the COM of perylene and the surface of the silicon oxide substrate, 16 starting configurations were derived ˚ In corresponding to the 16 sampling windows with a spacing of 2 A. each window, a 5 ns umbrella sampling simulation was performed for each SAM system. Then, the potential of mean force between the perylene and each SAM as a function of the distance between the COM of perylene and the surface of the silicon oxide substrate was calculated using the weighted histogram analysis method (WHAM) [25]. 3. Results and discussion 3.1. Deposition behavior of perylene onto different densely packed SAMs The deposition configurations of the perylene molecule on the four densely packed SAMs at the end of the MD simulation are shown in Fig. 1. From the figure, it can be seen that the organic molecule was deposited onto different positions of the SAMs. On the LE monolayer covered substrate (25%-SAM), the small molecule was adsorbed onto the surface of the substrate (Fig. 1a), while on the full LC monolayer covered substrate (100%-SAM), the small molecule was adsorbed on the surface of the SAMs (Fig. 1d). The equilibrium of the simulations was determined based on the vertical distance between the COM of the perylene molecule and the surface of the SiO2 substrate during the MD run. These distances with time are shown in Fig. 2. It should be noted that the distance in each SAM system remained steady during the last 4.5 ns of the simulation, which confirmed each system had reached equilibrium. Another significant observation is that the perylene molecule was deposited deeply on to the alkyl chains film with a decrease in cover density of the SAM. This is probably due to the steric hindrance of the more densely packed alkyl chains film, which hindered the perylene molecule from submerging into the film. 3.2. Structural properties of different densely packed SAMs As discussed above, the perylene molecule would submerge deeply into the hydrocarbon film with the decrease of cover density
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
165
Fig. 1. Snapshots of the configurations of the four different densely packed SAMs systems at the end of the simulation. (a) 25%-SAM, (b) 50%-SAM, (c) 75%-SAM and (d) 100%-SAM. Perylene molecules are represented by red VDW spheres for clarity, while the SAMs are in ball and stick model. The atom coloring scheme is O, red; Si, yellow; C, cyan; and H, white. (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.)
Fig. 2. Time profiles of the vertical distance between the COM of the perylene molecule and the surface of the SAM in each simulation system.
of SAMs. We considered that the packing density was responsible to this phenomenon. Taking this into account, the structure of the alkyl chains on the substrate was analyzed. First, the locations of different components in each densely packed SAM were characterized by calculating the average number density distribution functions along the z-axis, which is in the direction normal to the plane of the substrate (Fig. 3). The different components including alkyl chains and substrate were separately computed. The distributions of perylene were also plotted to show the deposition locations of perylene molecules on different densely packed SAMs. From the density distribution, it was observed that as the cover density of SAM increased, the thickness of the alkyl chain film on the substrate increased. For the LC monolayer covered SAM (100%SAM), a plateau is found from the density distribution profile. This indicates that the arrangement of the alkyl chains in the film is very ordered. According to the distribution, the thickness of the full covered film could be estimated to be approximately 1.5 nm. For the
Fig. 3. Number density profiles of components for each SAM system in the direction along the z-axis of the simulation cell. Panels a and b correspond to the distribution of hydrocarbon chains and perylene molecule, respectively. The curves of each system are indicated in the figure.
two less densely packed SAMs (50% and 75%), two layers with different densities of alkyl chains were found from the profiles. That is, the density of the film which is adjacent to the SiO2 substrate is larger than that on the surface of the SAM, which means the hydrocarbon chains bend to the substrate. A sharp peak was observed from the profile of the 25%-SAM, which means that a thick alkyl
166
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
Fig. 4. (a) Order parameter SZ of the hydrocarbon chains in films of the four SAM systems. (b) Probability distribution of the tilt angle between the hydrocarbon vector and the z-axis of the simulation cell for each SAM system.
chains film forms on the substrate. The intensity of the density profile of 25%-SAM is significantly weaker than those of the other three SAMs, which indicates that the alkyl chains do not cover the entire substrate. Number density profiles of the perylene molecule in the zdirection provide further insights into the deposition location on the SAMs (Fig. 3b). On the 100%-SAM, the perylene molecule is located on the top surface of the SAM, while on the 50% and 75%SAM, the perylene molecules submerge into the films, located at the region of low density as discussed above. The location of the perylene molecule is almost coincidental with that of the alkyl chains on the 25%-SAM, indicating that the perylene molecule was deposited onto the SiO2 substrate. 3.3. Orientation of alkyl chains on different densely packed SAMs To better quantify the conformations of the alkyl chains on different densely packed SAMs, the order parameter of the hydrocarbon films was calculated which is defined as: SZ =
3 1 cos2 − 2 2
where is the angle between the z-axis and the molecular axis of the respective Cn−1 –Cn+1 bond (n = 2,3,. . .,17, C1 is the carbon atom which is connected to the substrate, and C18 is the terminal atom of the tail) in the hydrocarbon chains. These results are shown in Fig. 4a. From this figure it can be seen that, for the 100%-SAM, values for each carbon seem to change smoothly without sharp variation. For the 75%-SAM, values reach a plateau starting from the first two carbon atoms. This suggests that those hydrocarbon chains have a similar inclination in the two more densely packed SAMs. For the 25% and 50%-SAMs, the variation of these values is very obvious, suggesting that the hydrocarbon chains are more disordered in these two cases. More detailed information on the orientation of the hydrocarbon chains can be obtained from the probability distribution of the tilt angle ˛ between the hydrocarbon chain vector and the z-axis. The vector connecting the C1 atom and the terminal C18 atom is defined as the chain vector. Fig. 4b shows the probability distribution r(˛) of the tilt angle between the tail vectors with the normal of the surface of the SAM system. A sharp peak can be observed from the angle distribution of the 100%-SAM, suggesting the arrangement of the alkyl chains in the LC monolayer is well ordered with an average tilt angle of approximately 41.4◦ . In the case of the LE monolayers of the other three SAM systems, the distributions are in wide ranges especially for the 50%-SAM system, which indicates that the chains are very disordered. The tilt angle distribution of the chains on the 25%-SAM is mainly around 85.2◦ , which means the chains are almost perpendicular to the normal of the SAM. Since the
packing density of the alkyl chains of the 25%-SAM is much lighter, the chains tend to bend towards the surface of the substrate and form a thin film. 3.4. Diffusion properties of perylene on the SAMs It has been reported that the site-selective deposition behavior of perylene on different densely SAMs is a diffusion-controlled process [10,22]. To investigate the diffusion properties of the organic molecules on the films, we have calculated the mean square displacements (MSDs) of perylene molecules in the xy-plane of the four SAM systems over the last 500 ps MD trajectories, as shown in Fig. 5. From the figure, it can be seen that the diffusion of the perylene molecule on the LC monolayer covered SAM (100%-SAM) is significantly higher than it is on the other three SAMs. This is because that the perylene molecule was deposited onto the surface of the hydrocarbon film due to the steric hindrance of the densely packed hydrocarbon chains on the substrate, which makes the organic molecule diffuse more freely on the top surface of the SAM. The diffusion of the perylene molecule on the 75%-SAM is the lowest among the four SAM systems, and it is very similar to that of the 25%-SAM. While, the diffusion of the perylene molecule on the 50%-SAM is much more pronounced than the above two SAM systems. Because the perylene molecule can submerge into the alkyl chain film as discussed above, we consider that the motion of the perylene molecule should be related to the packing density of the surrounding alkyl chains. From Fig. 3, the locations of the perylene molecule on the three kinds of less densely packed SAM (25%, 50%, and 75%-SAM) are approximately 1.5, 2.1 and 2.5 nm of the simulation cell, respectively. The corresponding number densities of the alkyl chains at the three locations are 90, 79 and 92 nm−3 ,
Fig. 5. Mean square displacements for perylene on the four different densely packed SAMs.
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
respectively. This indicates that the diffusion of perylene is confined with the increased density of the surrounding alkyl chains. The lower density of the alkyl chains around perylene on the 50%SAM is mainly due to the less ordered arrangement of the chains, while for the 25%-SAM, the chains bend towards the surface of the SiO2 substrate, resulting in a higher density of the film. Since the film on the 25%-SAM did not cover the whole surface of the substrate, the perylene molecule would locate at the porosities on the surface with a limited mobility. 3.5. Deposition PMF of different densely packed SAM systems The free energy of deposition of a perylene molecule from the gaseous phase onto the different densely packed SAMs was calculated using umbrella sampling [23] and the weighted histogram analysis method (WHAM) [25]. As discussed above in Section 2, a single perylene molecule was pulled towards the surface of the SiO2 substrate and the associated potential of mean force (PMF) was calculated during this process. In this case, the reaction coordinate corresponds to the z-axis of the simulation box. The PMFs as a function of the vertical distance between the COM of a perylene molecule and the surface of the substrate were calculated using the GROMACS tool g wham [31], shown in Fig. 6. From Fig. 6, satisfactory convergence can be achieved for the long sampling simulations. The global minimums of the four potential of mean force profiles occur at <0.50, 0.75, 1.11, and 1.93 nm for the four SAM systems with the cover density increased. These minimums are comparable to the deposition locations of the perylene molecule on the four kinds of SAM shown in Fig. 2. This indicates that the deposition positions of the perylene molecule on the different SAM systems predicted using steered molecular dynamics methods (SMD) agree well with those obtained by the equilibrium MD method. Since the minimums of the PMF curves represent the stable binding sites between the perylene molecule and SAM, they can be used to explain the preferable deposition locations of perylene onto the SAM. The binding energy G of a perylene molecule adsorbed on each kind of SAM can be determined from the corresponding PMFs. The overall order of binding energies for the four kinds of SAM systems is as follows: G25%-SAM ≈ G75%-SAM > G50%-SAM > G100%-SAM . It should be noted that the global minimum is similar for the 25% and 75%-SAM systems. This means perylene binds to these two kinds of SAM more stably than the other two SAM systems. The higher binding energies in these two SAM systems are mainly due to the higher density of alkyl chains around the perylene molecules as discussed above. In contrast, the binding energy of 50%-SAM system is smaller due to the less ordered and loosely packed alkyl chains around the
167
perylene molecule. The less stable binding between perylene and SAM causes much more pronounced displacement of the perylene molecule on the film. For the 100%-SAM, it can be seen that there is another pronounced minimum located at around 1.0 nm besides the global minimum. The global minimum represents the preferred location of perylene on the 100%-SAM, which is on the top surface of the film. However, the perylene did not submerge into the film during the equilibrium MD simulation. This is because the perylene molecule was pulled into the film under an external force which was applied on perylene in the steered MD simulation. The configuration of perylene submerged into the film is provided in the Supplementary data, Fig. S2. It can be seen that the perylene molecule is located at the gap between the well-ordered hydrocarbon chains. Due to the well-ordered chains of the film, the gap can provide much free space to take in the perylene. Thus, the binding between perylene and SAM is quite unstable. The profile of the external force with time evolution and configurations at different times are provided in Figs. S3 and S4. The maximum value occurs at about 340 ps at which time the perylene molecule is about to submerge into the film under the external force. This indicates that large force is needed to pull the perylene molecule into the film. In other words, it is difficult for the perylene molecule to enter into the film of a fully packed SAM. Based on the above analysis, we can conclude that perylene prefers to deposit onto the less densely packed SAM, i.e., the LE phase on the substrate. Therefore, site-selective deposition on SAMs with a striped pattern composed of the expanded and condensed alkyl chains phase in the stripes may occur. Since the binding energy between perylene and the expanded alkyl chains phase is higher than that of the condensed alkyl chains phase, the organic molecules that deposit on the condensed phase can move to the expanded phase of the SAM.
4. Conclusions We studied the deposition behavior of the organic molecule perylene onto different densely packed SAMs using equilibrium and biased MD simulations. The equilibrium MD results show that the perylene molecule deposits on the different positions of the different densely packed SAMs. With decreased cover density of SAM, the arrangement of the hydrocarbon chains on the substrate will become less ordered and perylene will submerge deeply into the alkyl chains film. On the LE monolayer covered SAM, the hydrocarbon chains bend towards the surface of the substrate, producing higher density of alkyl chains near the substrate and lower density on the surface region of the film. The diffusion of perylene is related to the density of the surrounding alkyl chains. That is, with a higher density of the alkyl chains around the perylene molecule, the diffusion of perylene will be confined. Using umbrella sampling and the weighted histogram analysis method, the potential of mean force (PMF) of the deposition process of a perylene molecule from the gaseous phase onto the different densely packed SAMs was calculated. From the PMF curves, the binding energy between perylene and each kind of SAM can be estimated. Based on the binding energies, the site-selective deposition of perylene on different densely packed SAMs can be explained.
Acknowledgments
Fig. 6. Potential of mean force along the reaction coordinate for a perylene molecule pulled from gaseous phase onto different densely packed SAMs (0 represents the surface of the silicon oxide substrate).
This work was supported by the National Natural Science Foundation of China (No. 21203084). Thanks to Dr. Edward C. Mignot, Shandong University, for linguistic advice.
168
H. Yan et al. / Applied Surface Science 349 (2015) 163–168
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.apsusc.2015.04. 211 References [1] M. Gaal, C. Gadermaier, H. Plank, E. Moderegger, A. Pogantsch, G. Leising, E. List. Adv. Matter 15 (2003) 1165. [2] C. Kallinger, M. Hilmer, A. Haugeneder, M. Perner, W. Spirkl, U. Lemmer, J. Feldmann, U. Scherf, K. Müllen, A. Gombert, V. Wittwer, Adv. Matter 10 (1998) 920. [3] Y. Koide, Q. Wang, J. Cui, D.D. Benson, T.J. Marks, J. Am. Chem. Soc. 122 (2000) 11266. [4] T. Lee, J. Zaumseil, Z. Bao, J.W.P. Hsu, J.A. Rogers, PNAS 101 (2004) 429. [5] J.G.C. Veinot, H. Yan, S.M. Smith, J. Cui, Q. Huang, T.J. Marks, Nano Lett. 2 (2002) 333. [6] B.J. Cheek, A.B. Steel, M.P. Torres, Y. Yu, H. Yang, Anal. Chem. 73 (2001) 5777. [7] B.M. Lingerfelt, H. Mattoussi, E.R. Goldman, J.M. Mauro, G.P. Anderson, Anal. Chem. 75 (2003) 4043. [8] X.D. Chen, A.L. Rogach, D.V. Talapin, H. Fuchs, L.F. Chi, J. Am. Chem. Soc. 128 (2006) 9592. [9] J. Hao, N. Lu, L. Li, M. Hirtz, L. Gao, W. Wang, C. Du, H. Fuchs, L.F. Chi, Soft Matter 6 (2010) 5302. [10] J. Hao, N. Lu, Q. Wu, W. Hu, X. Chen, H. Zhang, Y. Wu, Y. Wang, L.F. Chi, Langmuir 24 (2008) 5315. [11] M. Hirtz, H. Fuchs, L.F. Chi, J. Phys. Chem. B 112 (2008) 824. [12] L. Gao, N. Lu, J. Hao, W. Hu, G. Shi, Y. Wang, L.F. Chi, Langmuir 25 (2009) 3894. [13] W. Hu, N. Lu, H. Zhang, Y. Wang, N. Kehagias, V. Reboud, C.M. Sotomayor, Torres, J. Hao, W. Li, H. Fuchs, L.F. Chi, Adv. Mater 19 (2007) 2119. [14] H.Y. Choi, S.H. Kim, J. Jang, Adv. Mater. 16 (2004) 732. [15] M. Gleiche, L.F. Chi, H. Fuchs, Nature 403 (2000) 173. [16] M. Gleiche, L.F. Chi, E. Gedig, H. Fuchs, ChemPhysChem 2 (2001) 187. [17] J.M. Choi, S.H. Jeong, D.K. Hwang, S. Im, B.H. Lee, M.M. Sung, Org. Electron 10 (2009) 199. [18] S. Steudel, D. Janssen, S. Verlaak, J. Genoe, P. Heremans, Appl. Phys. Lett. 85 (2004) 5550. [19] Z. Xu, M. Hirtz, S.L. Yuan, C.B. Liu, L.F. Chi, Chem. Phys. Lett. 507 (2011) 138. [20] C. Zhang, Z. Xu, H. Yan, F. Gao, S.L. Yuan, Chem. Phys. Lett. 571 (2013) 38.
[21] C. Zhang, C. Du, H. Yan, S.L. Yuan, L.F. Chi, RSC Adv. 3 (2013) 15404. [22] M. Hirtz, N. Kumar, J.H. Franke, J. Hao, N. Lu. H. Fuchs, L.F. Chi. J. Colloid Interface Sci. 389 (2013) 206. [23] G.M. Torrie, J.P. Valleau, J. Comput. Phys. 23 (1977) 187. [24] S. Izrailec, S. Stepaniants, B. Isralewitz, D. Kosztin, H. Lu, F. Molnar, W. Wriggers, K. Schulten, Steered Molecular Dynamics, Springer-Verlag, Berlin, 1998. [25] S. Kumbia, D. Bouzida, R.H. Swendsen, P.A. Kollman, J.M. Rosenberg, J. Comput. Chem. 13 (1992) 1011. [26] A. Striolo, J.M. Prausnitz, J. Chem. Phys. 114 (2001) 8565. [27] Y. Sun, B.N. Dominy, R.A. Latour, J. Comput. Chem. 28 (2007) 1883. [28] Z. Xu, X. Yang, Z. Yang, J. Phys. Chem. B 112 (2008) 13802. [29] A. Hozumi, S. Asakura, A. Fuwa, N. Shirahata, T. Kameyama, Langmuir 21 (2005) 8234. [30] D. Janssen, R. DePalma, S. Verlaak, P. Heremans, W. Dehaen, Thin Solid Films 515 (2006) 1433. [31] D. van der Spoel, E. Lindahl, B. Hess, the GROMCAS Development Team, GROMACS User Manual Version 4.6.3, www.gromacs.org 2013. [32] W.L. Jorgensen, D.S. Maxwell, J. Tirada-Rives, J. Am. Chem. Soc. 118 (1996) 11225. [33] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Jr. Montgomery, T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision B.05; Gaussian, Inc., Pittsburgh, PA, 2003. [34] C.D. Lorenz, E.B. Webb III, M.J. Stevens, M. Chandross, G.S. Grest, Tribol. Lett. 19 (2005) 93. [35] H.J.C. Berendsen, J.P.M. Postma, W.F. van Gunsteren, A. DiNola, J.R. Haak, J. Chem. Phys. 81 (1984) 3684. [36] B. Hess, H. Bekker, H.J.C. Berendsen, J.G.E.M. Fraaije, J. Comput. Chem. 18 (1997) 1463. [37] U. Essman, L. Perera, M.L. Berkowitz, T. Darden, H. Lee, L.G. Pedersen, J. Chem. Phys. 103 (1995) 8577. [38] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics 14 (1996) 33.