ARTICLE pubs.acs.org/Langmuir
Steric Blocking as a Tool To Control Molecular Film Geometry at a Metal Surface Kin L. Wong,†,‡ Zhihai Cheng,† Greg Pawin,†,‡ Dezheng Sun,† Ki-Young Kwon,†,§ Daeho Kim,† Robert Carp,† Michael Marsella,† and Ludwig Bartels*,† †
Departments of Chemistry and Physics, University of California—Riverside, Riverside, California 92521, United States Departments of Chemistry and Electrical Engineering, University of California—Los Angeles, Los Angeles, California 90095, United States § Department of Chemistry, Gyeongsang National University, Jinju, Gyeongnam 660-701, South Korea ‡
ABSTRACT: The application of steric blocking in surface science is exemplified by the control of surface patterns through the selective methylation of pentacenetetrone. Pentacenetetrones interact (with one another) on Cu(111) via intermolecular hydrogen bonding involving the carbonyl oxygen and the adjacent hydrogen atoms. Steric blocking of the intermolecular interaction by the successive insertion of inert methyl groups at terminal locations transforms a dense molecular pattern first into isolated double rows and eventually into single rows in a highly predictable fashion. Density functional theory modeling reveals the underlying energetics.
’ INTRODUCTION Control of the geometry and morphology of patterns formed by organic molecules at surfaces has attracted increasing interest over the past decade. This is partially driven by the discovery of new methods of pattern control and partially by the emergence of applications of molecular layers in a variety of fields ranging from electron injection layers in electronic/optoelectronic applications to the lubrication of hard disks and to antifouling coatings for maritime use. Intermolecular interactions as diverse as covalent, coordinative, and hydrogen bonds have been utilized, and a large number of research projects focus on the optimization and strengthening of intermolecular bonds in tailoring film growth. On the whole, great success has been achieved. (For reviews, see refs 1 3.) Here we show that the opposite of strengthening bonds, namely, steric blocking (i.e., the prevention of access to an otherwise present interaction site through the placement of an inert group), can similarly control film patterns in a highly predictable fashion. Arguably, steric blocking is just another way of optimizing bonding (i.e., toward its reduction). It results in less tightly bonded structures that, depending on the application, may be desirable. Steric blocking is commonly employed in solution-phase and biological chemistry, where it is of great practical importance. In scanning the surface science literature, we find that whereas steric blocking has been used implicitly in many prior studies, it has not been conceptualized explicitly or attained the same interest as in other fields of chemistry.4 Steric blocking is important in the formation of homochiral surface domains, where the chirality of the adsorbed species is a consequence of the steric blocking of a molecular rotational degree of freedom by the substrate or the blocking of particular interaction sites.5 12 There have been many incidents in which concepts related to steric blocking have been observed, including the dependence of r 2011 American Chemical Society
surface patterns on the protonation of terephtalic acid,13 the interdigitation of alkyl chains,14 and the competition between coadsorbates in the formation of metal organic networks.15 To elucidate this concept in a systematic fashion, we prepared coverages of a family of related compounds: pentacenetetrone (PT), methyl-pentacenetetrone (MPT), and dimethyl-pentacenetetrone (DMTP) as shown in Scheme 1. We find that these molecules interact via hydrogen bonding between the carbonyl oxygen atoms and exclusively the hydrogen atoms in R positions on the aromatic ring systems (Scheme 1). Previous work showed that similar hydrogen bonds in anthraquinone on a Cu(111) surface are capable of arranging the molecules into rows,16 although these bonds are insignificant in the solution phase. This notion is generally ascribed to charge transfer from the substrate to the aromatic system, leading to a slight extension and higher susceptibility of the C H bonds.16 21 In contrast, there is no evidence of hydrogen bonding to atoms attached to saturated carbon centers.22 This background renders the PT, MPT, and DMPT family of molecules an ideal test case for the exploration of steric blocking at surfaces. Moreover, the comparatively low bond energy and otherwise high mobility of arenes on Cu(111) permits the optimization of the film structure at relatively low temperatures (avoiding desorption or decomposition) and prevents capture of the system in metastable states.
’ EXPERIMENTAL SECTION This project is a combination of scanning tunneling microscopy (STM) with organic synthesis and density functional theory (DFT) modeling of adsorption structures. All STM measurements proceeded on sputter-and-anneal cleaned samples, on which the reactants were Received: March 17, 2011 Published: June 14, 2011 8735
dx.doi.org/10.1021/la2015435 | Langmuir 2011, 27, 8735–8737
Langmuir
ARTICLE
Scheme 1. Pentacenetetrone (PT) and the Substitution Sites (Gray) for Methyl-PT (MPT) and Dimethyl-PT (DMTP)a
a Hydrogen bonding in the molecular film employs the hydrogen atoms at R positions exclusively; steric blocking at β positions can prevent access to them.
deposited at cryogenic temperatures (∼100 K) followed by annealing to room temperature for ∼1 h. All coverages fill only ∼20% of the sample surface. All images were obtained at ∼85 K. PT is available commercially. DMPT was synthesized via the dimerization of appropriately functionalized napthaquinones as described in ref 23. MPT was obtained via dimerization of a mixture of methylated and unmethylated napthaquinones. This results in a mixture of PT, MPT, and DMPT, which was separated chromatographically. Given the small difference between the species, the separation was not perfect, and coverages of MPT contain unavoidable traces of DMPT and PT. Density functional theory calculations use the VASP code24 with the generalized-gradient approximation25 for the exchange-correlation functional and the plane-wave pseudopotential method26 with ultrasoft pseudopotentials.27 Our calculations use a 3 11 substrate atom supercell with three substrate layers. The supercell size and geometry are taken to be rigid. The geometric degrees of freedom of all atoms are optimized so that the remaining forces are less than 0.02 eV/Å with the exception of the bottom substrate layer, which was kept fixed in all three dimensions. For calculations lifted from the substrate, also the y degree of freedom of one carbon atom was fixed to prevent the molecules from rotating in an arrangement incompatible with the substrate.
’ RESULTS AND DISCUSSION STM images of PT, MPT, and DMPT patterns, respectively, are shown in panels a c of Figure 1. The molecules appear as elongated features in which the oxygen atoms that anchor the molecules on the substrate appear as slight indentations. Similar appearances of covalently bonded oxygen atoms have been reported for a number of molecules.16,23,28 Figure 1a shows that PT forms dense molecular islands corresponding to a 0 1 7 0 @ A 3 2 superstructure (Figure 1d) in which rows of molecules aligned along the substrate [110] direction are offset by half of their length and three substrate atomic rows ([156] direction). This structure allows the molecules to form hydrogen bonds via each of their carbonyl oxygen and R hydrogen atoms to its four nearest neighbors (i.e., a total of eight hydrogen bonds per molecule). This results in the formation of a continuously interacting film (Figure 1g). The insertion of a methyl group in one of the terminal β positions (resulting in MPT) prevents access to the R hydrogen atom and thus sterically blocks the formation of the continuous film observed for PT (Figure 1b,h). Rather than forming a looser continuous pattern, the molecules adopt double rows running in the [156] direction, which allow the placement of all steric blocks to the outside of the double row while maintaining within the
Figure 1. (a c) STM images of PT (bias 2.1 V, current 0.09 nA, 53 Å 38 Å), MPT (bias 2.8 V, current 0.06 nA, 55 Å 40 Å), and DMPT (bias 2.7 V, current 0.13 nA, 55 Å 40 Å) and (d f) corresponding model of the molecules on the substrate. PT, MPT, and DMPT form extended islands, double rows, and single rows on Cu(111), respectively. (Insets g and h) Intermolecular hydrogen bonding (orange) and how the methyl group provides steric blocking (red stars) of this arrangement, respectively.
double row the same intermolecular spacing and interaction as observed for PT. Figure 1e shows a model of this structure. Note that the methyl steric block does not displace the hydrogen atom through which PT molecules interact to form a close-packed island. Rather, it prohibits access to it through its bulk. Finally, the insertion of another steric block at the β position diagonally opposite (leading to DMPT) prohibits access to the R-hydrogen atoms employed in the bonding between the pair of molecules abreast in the MPT row. Figure 1c,f shows that singlefile rows of molecules running in the same [156] direction result. Within the rows, DMPT molecules maintain the same intermolecular geometry/interaction as observed in PT films, indicating that the steric blocking is highly local. All rows found on the surface run exclusively diagonally across the molecules, placing the blocking group at the side of the row. The presence of the steric block(s) renders both MPT and DMPT chiral when confined in 2D. Similarly, their arrangement of interaction sites with neighboring molecules becomes chiral. Because all nonblocked interaction sites are used in surface pattern formation, for geometric reasons continuous and extended patterns of these species can form only in homochiral ensembles. As a consequence, we find that MPT and DMPT rows are homochiral. Separate rows of both surface enantiomers are found to accommodate the racemic mixture of surface species generated at adsorption. Because coverages of MPT contain traces of PT and DMPT as discussed in the Experimental Section, the perfect rows found for MPT are shorter than those for DMPT. In summary, our experiments affirm that steric blocking is a viable concept for the control of surface patterns. Our finding hinges, however, on the presence of an intermolecular hydrogen bond in the first place that may be blocked. To validate its 8736
dx.doi.org/10.1021/la2015435 |Langmuir 2011, 27, 8735–8737
Langmuir
Figure 2. For the investigation of the presence of hydrogen bonding, we employed different tiling with the same supercell (green), leading to (left) interacting and (right) noninteracting surface patterns.
presence, we performed DFT modeling: in particular, we estimate the hydrogen bond strength by a comparison of the total energy of a row of interacting PT molecules (E1D) to that of an open 2D pattern of noninteracting PT molecules (E2D) (Figure 2a,b), where in the former, hydrogen bonds (marked orange) can be present and in the latter they cannot. These systems can be set up from identical supercells (green) through the application of different periodicity. To account for any computational artifacts associated with the different supercell periodicity, we also calculate the total energy for the substrate alone (E1Ds,2Ds). Given that there are two hydrogen bonds between adjacent molecules in the row and two bonding partners along the row, the intermolecular bond energy is calculated to be [(E1D E2D) (E1Ds E2Ds)]/4 = 0.04 eV, which compares well with the experimentally obtained hydrogen bond energy of 0.05 eV in anthraquinone molecular rows.29 To corroborate our result further, we perform the same calculation for the molecules lifted ∼5.6 Å above the substrate, where no charge transfer from the substrate to the aromatic moiety occurs. Here the hydrogen atoms are expected to be unsusceptible to hydrogen bonding. In good agreement, we find an insignificant amount of intermolecular repulsion (∼7 meV). In summary, it has been shown that the concept of steric blocking can be transferred to the control of pattern formation at surfaces: when pentacene derivates are used, the insertion of a blocking group as small (∼3% of the molecular mass) and as inert as a methyl group can direct the shape of a molecular film in a highly predictable fashion, even though the actual functional groups involved in the original intermolecular interaction, the carbonyl oxygen atoms and the hydrogen atoms at R positions, are not modified. Steric blocking is used widely and successfully in solution-phase chemistry; explicitly conceptualizing it in the context of surface pattern formation facilitates a more direct transfer of successful solution-phase approaches to surface pattern formation.
’ AUTHOR INFORMATION
ARTICLE
(3) Barth, J. V. Surf. Sci. 2009, 603, 1533–1541. (4) McMurry, J.; Castellion, M. E. Fundamentals of Organic and Biological Chemistry, 2nd ed.; Prentice Hall: Upper Saddle River, NJ, 1999. (5) Gopakumar, T. G.; Matino, F.; Schwager, B.; Bannwarth, A.; Tuczek, F.; Kroger, J.; Berndt, R. J. Phys. Chem. C 2010, 114, 18247– 18251. (6) Bohringer, M.; Morgenstern, K.; Schneider, W.; Berndt, R.; Mauri, F.; De Vita, A.; Car, R. Phys. Rev. Lett. 1999, 83, 324–327. (7) Barlow, S. M.; Louafi, S.; Le Roux, D.; Williams, J.; Muryn, C.; Haq, S.; Raval, R. Surf. Sci. 2005, 590, 243–263. (8) Rao, B.; Kwon, K.; Zhang, J.; Liu, A.; Bartels, L. Langmuir 2004, 20, 4406–4412. (9) Schunack, M.; Laegsgaard, E.; Stensgaard, I.; Johannsen, I.; Besenbacher, F. Angew. Chem., Int. Ed. 2001, 40, 2623–2626. (10) Spillmann, H.; Dmitriev, A.; Lin, N.; Messina, P.; Barth, J. V.; Kern, K. J. Am. Chem. Soc. 2003, 125, 10725–10728. (11) Xu, B.; Tao, C. G.; Williams, E. D.; Reutt-Robey, J. E. J. Am. Chem. Soc. 2006, 128, 8493–8499. (12) Ernst, K. H. Supramol. Chirality 2006, 265, 209–252. (13) Ca~ nas-Ventura, M. E.; Klappenberger, F.; Clair, S.; Pons, S.; Kern, K.; Brune, H.; Strunskus, T.; Woll, Ch.; Fasel, R.; Barth, J. V. J. Chem. Phys. 2006, 125, 184710. (14) Tahara, K.; Lei, S. B.; Adisoejoso, J.; De Feyter, S.; Tobe, Y. Chem. Commun. 2010, 46, 8507–8525. (15) Langner, A.; Tait, S. L.; Lin, N.; Chandrasekar, R.; Ruben, M.; Kern, K. Chem. Commun. 2009, 2502–2504. (16) Pawin, G.; Wong, K. L.; Kwon, K. Y.; Bartels, L. Science 2006, 313, 961–962. (17) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford University Press: Oxford, U.K., 1997. (18) Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond; Oxford University Press: Oxford, U.K., 2001. (19) Yokoyama, T.; Yokoyama, S.; Kamikado, T.; Okuno, Y.; Mashiko, S. Nature 2001, 413, 619–621. (20) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S. J.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347–1363. (21) Berland, K.; Einstein, T. L.; Hyldgaard, P. Phys. Rev. B 2009, 80, 155431. (22) Kwon, K. Y.; Pawin, G.; Wong, K. L.; Peters, E.; Kim, D.; Hong, S.; Rahman, T. S.; Marsella, M.; Bartels, L. J. Am. Chem. Soc. 2009, 131, 5540–5545. (23) Cheng, Z.; Chu, E. S.; Sun, D.; Kim, D.; Zhu, Y.; Luo, M.; Pawin, G.; Wong, K. L.; Kwon, K.-Y.; Carp, R.; Marsella, M.; Bartels, L. J. Am. Chem. Soc. 2010, 132, 13578. (24) Kresse, G.; Hafner, J. Phys. Rev. B 1993, 47, 558–561. (25) Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244–13249. (26) Payne, M. C.; Teter, M. P.; Allan, D. C.; Arias, T. A.; Joannopoulos, J. D. Rev. Mod. Phys. 1992, 64, 1045–1097. (27) Vanderbilt, D. Phys. Rev. B 1990, 41, 7892–7895. (28) Loui, A.; Chiang, S. Appl. Surf. Sci. 2004, 237, 559–564. (29) Pawin, G.; Solanki, U.; Kwon, K. Y.; Wong, K. L.; Lin, X.; Jiao, T.; Bartels, L. J. Am. Chem. Soc. 2007, 129, 12056–12057.
Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This work was supported by U.S. Department of Energy grant DE-FG02-07ER15842 and by NSF grant 0749949.
’ NOTE ADDED AFTER ASAP PUBLICATION This paper was published on the Web on June 14, 2011. Changes were made to the Abstract and the last sentence of the Experimental Section. The corrected version was reposted on June 22, 2011.
’ REFERENCES (1) Bartels, L. Nat. Chem. 2010, 2, 87–95. (2) Elemans, J. A. A. W.; Lei, S. B.; De Feyter, S. Angew. Chem., Int. Ed. 2009, 48, 7298–7332. 8737
dx.doi.org/10.1021/la2015435 |Langmuir 2011, 27, 8735–8737
